diff --git a/.devcontainer/devcontainer.json b/.devcontainer/devcontainer.json new file mode 100644 index 0000000000..c39e1225f7 --- /dev/null +++ b/.devcontainer/devcontainer.json @@ -0,0 +1,18 @@ +{ + "build": { "dockerfile": "../docker/Dockerfile" }, + "runArgs": [ + "--gpus", + "all", + "--shm-size", + "8g" + ], + "customizations": { + "vscode": { + "extensions": [ + "ms-python.python", + "ms-python.vscode-pylance" + ] + } + }, + "shutdownAction": "none" +} \ No newline at end of file diff --git a/.gitignore b/.gitignore index 787d13ec67..b1aea68659 100644 --- a/.gitignore +++ b/.gitignore @@ -30,7 +30,7 @@ MANIFEST # before PyInstaller builds the exe, so as to inject date/other infos into it. *.manifest *.spec - +nohup.out # Installer logs pip-log.txt pip-delete-this-directory.txt @@ -62,7 +62,8 @@ instance/ # Scrapy stuff: .scrapy - +wandb +test_commands.sh # Sphinx documentation docs/en/_build/ docs/zh_cn/_build/ @@ -118,3 +119,7 @@ mmseg/.mim # Pytorch *.pth + +logs/ +*.png +*.out \ No newline at end of file diff --git a/README_amc-cbn.md b/README_amc-cbn.md new file mode 100644 index 0000000000..b0047ff7d4 --- /dev/null +++ b/README_amc-cbn.md @@ -0,0 +1,32 @@ +# README_AMC-CBN + +## Dataset +- ADE20K: Download ADE20K dataset from http://data.csail.mit.edu/places/ADEchallenge/ADEChallengeData2016.zip +- Angiography: + - `mmseg/datasets/angiography.py` + +## Settings +- logger: change `vis_backends` type to `WandbVisBackend` in `configs/_base_/default_runtime.py` +- wandb logging url: https://wandb.ai/amccbn/mmsegmentation-tool +- Adjust the number of epochs depending on the batch size. + +## Installation +- Download pretrained vit from https://github.com/open-mmlab/mmpretrain/tree/master/configs/vision_transformer and move it to `pretrain/` +- Used [vit checkpoint](vit_vit-b16_mln_upernet_8xb2-80k_ade20k-512x512.py) for training vit_vit-b16_mln_upernet_8xb2-80k_ade20k-512x512.py +- `pip install wandb` + +## Train +``` +tools/dist_train.sh configs/vit/vit_vit-b16_mln_upernet_8xb2-80k_ade20k-512x512.py [num_gpus] --work-dir logs/vit-upernet-ade20k +tools/dist_train.sh configs/upernet/upernet_r101_4xb4-160k_ade20k-512x512.py [num_gpus] --work-dir logs/res-101-upernet-ade20k + +tools/dist_train.sh configs/upernet/upernet_r101_4xb4-160k_cag-512x512.py [num_gpus] --work-dir logs/res-101-upernet-cag +``` + +## Transforms +- Additional augmentation functions using albumentation can be found in `mmseg/datasets/transforms/transforms.py` + +## Issues +- To create instance of `dataset` in the jupyter notebook, `init_default_scope('mmseg')` must be called. + - https://mmsegmentation.readthedocs.io/en/latest/advanced_guides/datasets.html#main-interfaces +- After adding a custom dataset class, you need to register the class to `datasets/__init__.py` file. diff --git a/a.ipynb b/a.ipynb new file mode 100644 index 0000000000..5525c580ae --- /dev/null +++ b/a.ipynb @@ -0,0 +1,430 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 2, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "[ WARN:0@36.485] global loadsave.cpp:248 findDecoder imread_('/workspaces/mmsegmentation-1/OUPUT/MAEOUT_240422/00000.png'): can't open/read file: check file path/integrity\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + } + ], + "source": [ + "import cv2\n", + "import matplotlib.pyplot as plt\n", + "import numpy as np\n", + "\n", + "# Define the paths to the images\n", + "image_path_output = '/workspaces/mmsegmentation-1/OUPUT/MAEOUT_240422/00000.png'\n", + "image_path_img = '/workspaces/mmsegmentation-1/data/cag/images/test/00000.png'\n", + "image_path_label = '/workspaces/mmsegmentation-1/data/cag/annotations/test/00000.png'\n", + "\n", + "\n", + "# Load the images using OpenCV\n", + "label_img = cv2.imread(image_path_label, 0)\n", + "img = cv2.imread(image_path_img)\n", + "output_img = cv2.imread(image_path_output, 0)\n", + "\n", + "# Create masks for pixels where label and output are both 1, label is 1 but output is 0, and label is 0 but output is 1\n", + "true_mask = (label_img == 1) & (output_img == 1)\n", + "false_mask = (label_img == 1) & (output_img == 0)\n", + "the_false_mask = (label_img == 0) & (output_img == 1)\n", + "\n", + "# Create an array to store the visualization\n", + "visualization = np.zeros((label_img.shape[0], label_img.shape[1], 3), dtype=np.uint8)\n", + "\n", + "# Define colors\n", + "green = [0, 0, 255] # Red for overlapping parts\n", + "yellow = [0, 255, 255] # Yellow for parts where prediction is not possible\n", + "blue = [0, 255, 0] # Green for parts where prediction is better\n", + "\n", + "# Assign colors to the visualization array based on the masks\n", + "visualization[true_mask] = green\n", + "visualization[false_mask] = yellow\n", + "visualization[the_false_mask] = blue\n", + "\n", + "# Display the result\n", + "plt.imshow(cv2.cvtColor(visualization, cv2.COLOR_BGR2RGB))\n", + "plt.axis('off')\n", + "plt.show()\n" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "00009.png\n", + "IoU: 19.76\n", + "F1 Score: 33.00\n", + "Precision: 32.68\n", + "Recall: 33.33\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "00072.png\n", + "IoU: 28.44\n", + "F1 Score: 44.28\n", + "Precision: 51.37\n", + "Recall: 38.91\n" + ] + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAOcAAADnCAYAAADl9EEgAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/OQEPoAAAACXBIWXMAAAsTAAALEwEAmpwYAADTAUlEQVR4nOz9eXRk6XneCT439gAigMCaQCLXyszaN1YVWcWdlMSdsinJsiQfj6dlj8eyZLq7rdGRrNOyelrd1qjbY7vHPqe9qD1W227ZlmTaFEWJNlncRVaxqlisYu2V+4LEDsQKILb5I+r3xXNvAshk6Uy3/+A9ByeRiBv3fsu7PO/zvt/3Rf1+X9+/vn99//rP70r9n92A71/fv75/7X19Xzm/f33/+s/0+r5yfv/6/vWf6fV95fz+9f3rP9Pr+8r5/ev713+mV+agD6Mo6r/xr5Ksrv8tiiJJuuEe/8w/T6VS6vV6ez7rVt/X7/f3fPZe30nex//3azd/T6VS4bN+vx/7br/fVyqV2rOtez0r2Qae5X9PpVKxf/3v3hbem81mb+gD7en1eje8J4qi8KxMJqN0Oq10Oq1UKhX+zefzyuVyyuVyKhQK4e98N4qi8GzaFEWR0um0stmsCoWCcrmcstmsstls6EOv1wtt5u+ZTEaZTEb9fl/dbledTid8P51Oq9vtqt/vK5vNhnal0+nY/el0WqOjo+r1emo2m9rZ2VGv11On09Hu7q663W4Yk3a7rW63G9rU6XTU6/WCPLbb7dBWPuNdyKuPHePW7XbVbrdjc817fS65v9PpqFqt6ty5czp//rzW19eHQmDXgcpZKBRCQxiodDodhMEnDoFwAUyn07GG8rtPciaTCZ3ySc/lcqGjmUwmJlR0mO/RDn82z0u2m3v4jj8TAaNNCLwLCRPpk0M/XJGYUO5xReB5vDeTyQRFo/8IcyaTCcqSzWZVLBY1MjKiYrGoUqmkVCqldrutdrsd5qDT6cT+xt9zuZyKxaIKhYIKhYLy+bzy+XxoR7FYVLlc1sjISGgv48G4RlGkTqejbrcbflA62sj4uWwkDYsbOhSj2+3GZIo5Tv7NDaYbN+TH/2W+mUeupKK6IjYaDe3s7KjdbmtnZyfMSafTUafTUTabDX/jh/lCqdEF3o/sSNL29raWl5f1wgsv6PHHH99X/w5Uzocffljz8/NhwjudjvL5fFA6t3w+YK5kCK1Pgk8232cgXdnodC6XUyaTiVlV9248G6VhkBEQv98NBe3PZrNByVAoBKzT6cQUTJJyuVxoW9II0aaksQoDboLLO12AaA/9zefzwVjl83mVy2WNjo7GjGav1wtegnni993dXXU6ndBuhKRQKKhUKqlQKIR5yefz4dnS3qhnPxRyq9d+38nlcjd93l5t8N+ZH/9bchwYN0mxMeQHWdje3tbOzo5yuZw6nc4N6Agj2263g7LyPve8yBP/lwaGIZfLaXR0VPl8ft/+HqicpVJJ8/PzarVayufz6nQ6wapiKfAC/J9GJgcKhcED0QH3Uu45/P9RFAWrj3BlMpnQWX7HyvIe3ptOp7W7uxsUyr2VGwIGLgk3MUhh0Ox9fM/bu5+XcKPR7Xa1u7sbg4W0I51Ox4xQoVBQNpvVyMiI8vm8+v2+Wq1WDOqiYAgBAtlqtbS7uxuDmOl0Wq1WK+Zpd3d3wxiBmPZTAv+bG9Q3e/kY3WqItF+b9gp9ut2u6vW6tre3g+Fzr+9yw/fS6XSA9Sgqnr3b7Yb/I8P+LIylIzcMHoqOHPq8JK8DlXNxcVGzs7MqFApBqba3t2PertfrBcvhwhpFUcDwPhCulJ1ORzs7O+F7KB6dR/iIU9wbMgkoyvb2dhDkVqsVJsW9JoMG7mewECwUjnbiXRFevHK/31en04lBdYdItCsZW+MhUVJJwXIyeQ7JeL/D6Xa7rd3dXbXb7dAe+sPnCAXQDK/qcVq/31etVlOj0VA+nw+KzTjn8/lgvJLGyvuDkmNMvC/7KfRe160o3q1cez2HmBWoz5yBMHzsmXvGy50JUNhDBQws/zIGzAdjjRzgfCTFOIO9rgOVs9Pp6MqVKzp16lRMyBFOYi3vlMcKrqwoA99HmdwDuvXJ5/OxGJXBZLCBx+6poyjS9vZ2zBMiuOl0OsCJZEyAIXAiQlLwUrwLmMuPK6d7bp6LoiW9J5bUiQXGD5hEvAdK4ZkIAc9zxII1RiERDvrOPVhrPOvU1FQQ0na7HTMgxLzuGXk/yi8ptFtS8PT/Z1/IYiaTUalUUiaTCc6g1+up1WoFBXWPisFhvlGofD6vZrMZ5hZ5xSBKCkYTpMl4M6boQSaTCTK+33UztlbNZlOLi4u6/fbbQ0O8wSgP1gdhcjbTSRdXjCQJ4gLnxBPPxQoxAAizkzR8x4kVV3qUiuck4RTP9riS99IXFMTj4mS8hNHB0gItkwYll8vFFMnjUYdPKDJtdlICjyApoBHuYw4QQrfcjGGz2Qzz0Wq1JCnMkRtIF3iU1+ciycjuBU33k7O94sn9rv2e6/OefD4wlfiPPjmSGxkZiYU3bgQd/eEoXG6deEQO8/n8DTxKLpcLz/B+73UdqJxYiM3NTa2trWlyclLSIHhnUhwW0hCE3d0/cZJ3gEHButBpfwafO8xDcIrFYhBKV26/x1lGjAkTidDznaTwOvvrzK4ktVotFQqFwIw6ivBYk4n2MU0aJTy0GzxnL4n1HYL52O3s7MRimaQBdSaSvjpj7d6EuNPRiqMmxsvHEajY6/VCX2jfQZfP0Z/k8j5jrImbk4QhirpXjOtGzdEdaCuJkJKpIA99XGYcChMu7O7uBqJzv+vA0WPyJOnq1avK5XKqVCqShhCLyfLYBIEiPgUGIrjuZeg8HcfieoCOdUM4XeH9vf53BtxjMjce/N/bjlLxbqfIuYdn8QyUkjgQOOdQx9tGykEa5nuz2WwwSgj2zs5OMGJYfDxkMubf2dkJc+asIHOXz+dj9/Eu4l2HstzH3NM+5kBSzPjRJifiQFgIXzIk8rlnDvj/m7mQRZ+TdrsdUkbO5LvcFAqFYNi2t7fDs1x+GA/nFBgbly9XRAyfpCDzKHPSiRxEpB2onOPj4xodHQ2Ctri4qH6/r0qlEibDYyJeDG3t1r9YLMboZSZld3dXkmLMKekLFNsVxvNqTg5hvRwSel6Ke3m3dCMJ4wQSgggUcQ8hKXhNf45DFRcYJ0rILTJZPNcNEePiUIi4eK9x9u962MC9tNfHEBKH2MuVq9frqdFoBPjMeAGzXcEwaBjXnZ2doLxA+WKxuGfI4ULMPCc9yc2gH3KWhLTE7A5JiYX5P7Hhzs5O8JI8R1JwGoQdxN8+zp5qA2U4uvDQB1lzLuYg1HBTQqhUKml7eztM4uXLl9XtdjU+Ph6LNYBGURSpXq9LUmDJULJMJqPt7e2QLyWH5DAQ7+gD6NaGvzHR/B9BosMMpjOzHrwjnMBMLJxPNoJFjJkc3GQ73Ui4Ivf7/ZDgJx3kSXCPYXlH8gJ2Qdpwn8fOHrfyPMbKISxjg7EEynrMj5cdGRmJVcswlgj/yMhIkBU8vRuP7e3toBgeo0lDiJycdxfevRQzGTKlUqkA+xkXSJgkl8APcgsP4KEBY+iw2NNe3jY32si4cxX8i1I7T7BXfOzXgcrZ7/dVKpU0NjamarWqjY0Ntdttra6uKoqiAHF90MHSwDcXGmkQJzIQPtDuBVFYPAyCj1V1L5kscnDlRXFQKM8dAh1pg8cK7lFJRCeZTE9MO5uKx0YB6PPY2FgoKPAUE5aVmMknVBomu0EBSZLHWfD9BNfhnhQPV6Io0ujoaHiH/50UEuEJgk47gecIqKTYfeRTMYieEvL5wqhHURQ8mHtb2sO7vH94c4fhHmtjROiLhy5OQiEXjJl7NveGHmO6Ana73SDvLoPIbzJ84rM/ESEE1c6kU7+4urqq3d1dTU5OqlgshsGmE0AzlMoZPyY2ybRizd0CY9kZKLwcA+VewoXaDQD3+0QjMLlcTvV6PQalaSPCkIwZneXjHcSb3Avbm8/nNTExEWhznuEeg3c7A5s0VsnLFc09tocOriT0hf4nyS8MhL+X3yG/PL5nHJg3F2Q3uMSkPNsNK8gp+XfnJ3iXj5mHIQ61k/Gckz481/9G2z1ltrOzE7gUJ84cyrthwFt737mS4+2e3A3AftdNUykQHYVCQbOzs1pdXVW1WlWr1dLOzo52dnY0Pz8f8kiSYoSFJ7WZCCBts9mMQQGHo+5VEEZXVGAnk0SnPb8IO+Z5KVfiJFuGd3VyhNivVCrFgnwf4KQnBCqNjo5qZGQkBsc9hgJFQJ6k0/FKJvqA8Hl85uQWIYXHS65kbtTwIN435te9AO9zJXBlSXo1H0e8lXMOkkKM7IUQzK8bKby2K7kTKklBz2azQRbpK2PCve5xnZgpFAphHnZ2dlSv19VsNmPkpUPVZMxM/0iReNjjBSE8x70ncvKmlBMB5MWjo6MqFApaWlpSo9EIcejVq1c1OzuriYmJGCRk4BCCQqEQSBKsFcE4jCJKwXclxbwQbWJA3Pow8cQvDBBwwycfi+leGBhNG2jP1tZWIA8QJOJwJgnFhgzJ5/MaGxuLWWw3Dp7v5B7vBwoDOvCY2AWXZ/C7e2EXagSZ9gHre71eLO3gzKbHVqAcr0fFA3mqwT3/XjGe9xFF5v9e1JAki4hby+VygL/IBk6k2WwqlUqpVCoFPsGVi/b63HoZJ+xus9nU1tZWUPYkmmCskT9nqV3xndBEbhgP+v6my/ewns4qplIpHTlyRM1mUysrK8GTra2tqd1ua2ZmJlguJtEhCZbXB81zZiMjI7GqIxTZUzNeEM6zUMJUKqXt7e1gBPCaTCLfcS/mS5K43MN1u11Vq1WNjIwEj+DkAbkrXwpFDTKTk4R8+0EwnptMiTghlLwcZrswueHg8vjY0wooB/e69ff54j0eo+3nARFi+pxML/EDIwoaAwIzT3yPNBLGQBquLuEHpFcsFmOFJM46e8zKYgoPOUqlkvr9fqjHTaIR+hxFg3JRyB0P6TBkjgK40AcM2JtSTs8LOczp9XoaHx9XLpfT2tqaGo2GOp2ONjc31Ww2NTc3FwTdBxWL5flPBNGhgQskXpC2ONvn7anX6zGm0Ev28Dg8m9yXW1ba5oyfe8N+v69GoxEEhndj/fDOpI3S6XTwTG5FkxaciUS4nMVLxld4TY+1GSugL4gEz4aXp697QXzmqNFoBMNH+5xs4d9GoxGMYVLRGPNarRYKxrmcxUZZkIdmsxnKDkFaxKz0GQOd5Ci8ZI6cZK1WC+tLaaPLMn9zxODKB5qo1+shDKCPTgoxp06k+VjQ51QqFfLVTg55/vR7Uk6E1ku03OoUi0XNzc0FJhfLdfHiRVUqFVUqlRhZ4Urua+WkYZCNYsB+8V6HcpLCoNA+FNchp0Mzz6t62sfJkmazGZ7v6++cHHHPwgATO7iHRjHdi2HlnexIxlcYEveyTmQ4S+iTj7Jwj3tYhIix9f4Qi/Ju2ozS0h7QkKMd3oPQ0TY4g2q1qlqtFtaQOjED+40BcKPpywoxVigdsb97Yo/n3JjxHQyBF4hICugGBeFzxjubzWpsbCy2iJs++kofR4XOnTghmjQG6MFBpNBNY04sE4NJ3pKXgdNLpZKWl5fV6XRUr9e1vLys3d1djY+PxybGoVShUFC9Xo/FSPzw/L2gmVt9n2wUy70Yk4FXyefzsSIJhAorzEQh/EyUD7KTE+5RWHPp+VnajUKhiEAuh70ek3hRhpMZbsxQOhQrSW55OzB07XY7eFJXcIeesJVuMLnPPSnQGzlAaTBsxH3MVz6fD6kSVwrmq9/vB2LR59MrphhvlIT43ivVWCyNrHU6nVjxPnLj/2eeHP14PMnYYCg6nU4oanB2HCPR7w+LRKR4ignZ5e9vSjn9pf1+P1hAhAernclkNDk5qfHxcS0uLobi6e3tbW1vb6tcLmtqaiqQQy5AHsc4++Ze2i0pVp37nZEDBrm3QSkYDGeMURKEDIOBN+R+Jw5SqUHCG/jHfR67+dg4nPJx8++4EUCIXRk9TvQ0Q7FYDEuhYNAnJydjgohldmEkjnLjgiIQs7H7AjE894F2PL3B/KAk/MzOzgY54t5MJqNisahisRgzSkBgoGgqlYoVR6C4Pj70A86AoojR0dFYvEgbnXfgfh9PV3bGbWdnJ0BjVqT495MGJhkOpdPpsITRK4do10HXLbG1QMixsbFQ6cFEYrGBPPPz84qiSOvr64G2J26Ym5sLioJ1LpfL2t7eDgrt3gQv5JDSl+J40hwomFRKJ26AXL72DmWE6XNI6BYWhcXLQzi4gLmSu9fmHgxP0iO5J0Ho/LmxCbMJdqPkXsrjYqfuXTAcNifDBdrgxSSSYmw1/RodHQ0w3kmiVCql0dFRzc7OqlarhXeyaB6jDHQFtTDeQEhQER610xnsv0NbXfEYG+clkrlfLo+/3eB7/EqcCAz2EAgvznhy+Tv4PvJOe3jnQfGmdIurUrBaxFbAQQYRy4rHm5ycVKlU0uLiYphEUi4TExMaHx8PnXRYKA1We3jpVRKT8x4YWc+5+SC6omB5qfZxFhDvgWUn+OczWFzGwrdpYYw8ZnFPnqTJnQwgXnIDRAkaypvMHTJeCBPvYSy9OsuZY56HoNIHFmH7/HnM6+QR/2JcPD4mJeWowVEAMaVDO49xnemUhuSNQ3MMCNDa5aPdbmtjY0OdTkdjY2NKp9OhrNArlNwQecUV8oIs4HRADs59OCnm44tRZz69NtrHxI0HxulNKSdecXR0VKVSKaawkB8O2TwOKRaLWlhYUD6f1+bmZvjs2rVrqtVqmp2d1cjIyA0kCkqWyWQCe8jEoVTQ7UnLhaIkSSYnYeiXM5pALay0x4a8F+9BbawzpfyejGU83+rtIvb1Nu1FYCXzkw6HkwwhQuCESdKL9PvDemKH/sTQvAPrT7tBGr1eL5YiwvvxbCeyaBfjy7vwSE4SOrnmf2dceBZz4rlJN1StVisQVNKwEi3JmjvTS9uZM28Dc8I7eSbz0mg0go54zOnkGH9joQRj4/noN6WcWC1npcgfJqGS57Cc4ZudnVWpVFK1WlWj0VAmk1G1WlWz2dTMzExYIwoEgGGjM8BQh39JwaDTfOYJfbyDFycTL7iXcMtPHOOem0F3JUxCwmRqBLjlNLwrE1bfvaJ7xqRgeO7T3+UkE7GPe1Q+w7vyzmKxGJL5Hv8A8+EXPB72MkUuhB7DRLudA/A+O4nCWPMe5h9j4CGAp9wcEfFMZCWZW/aVKE7cef7et8tB9kFabnCQ0/Hx8aAPrtQ+Bm643Kh4nxwGf0/KiXZTB+v1lAisEzc0hElEyPL5vGZmZpTL5bS+vh4sD5VG5XI5QC0Gmklw+EQsJMVXWfBeLDOwlO8iPHgBLveQKDFwEgUeHR19I07KK+q21d3tKer3lY0iZftdpfIpdTpSKpULAukwhwllItyAeXvxXExq0gszuW6R3et4qsQFgoQ84YgbE8+J+tz5WHluThpW5GDoHHb6MzqdjprNZoif3eDQf5cT5pt7t7e3lU4P9qTF2Lmz8H64ItJnN5LuoQjDvK9JgwK76yQkF22lXbVaLfSNOXPD7XPioWAyrfM9KydBvU+2d8rXIboCYdWw9CjH7Oyscrlc2Fiq3+8HBmx0dFTj4+MhXkrGY57PSwozA+JCgoDRZqAZ8arnvYjF6E+xWFSv19Po6Kiq1eqApd78hj74hU+p/00TiNmUok9EqtZOqVa7fSB4D0bq39NTq3VY1eo96lf66uWG0ExKqdOZUCo1hEKucG44khPtRtD/j8A4ucG4w+gS37ElCVCeGmdfs8g48S7GGuGlbaxY4ZnMJwhoY2MjPIu58tSEIwreIQ0YUuQCAcZgoiz84CkReCp+fCtMH8NerxfCIhSdccYz3+zycIcwgvy5x7aOMB1luqE56LqpchJYU/VCB90KI1S+343HDU6elEolFYvF4DXpGFanXC4HxWEAXMERPDrnE80Ee0zM91BQ2u6sJcpJn7Ho1WpVURRpc3NTLxX+SH/m3y/GB2hR0nekEV3RnL48+Fth8NPrZdTtFRS9XdJhSRVJvyz1Klltbr5NOzsKKKDVmlOt9n5lMvep0+lre7sVlCBJPnlahbHlb41GI+yFihH1kkTiMt/qE2bSmVL/3efPSyLhCFqtVqiNxvsR8+2VMsCA+xIzDLnndmFJ6/V68OL8ixIx/8R05HA9beMkD3IEX4GHdo/mRBVj7IY+6QCIQXk3aMJL+nBmqVQqFNV7tdh+101raz1eQSEdKnh8g8J4zotOOmuYy+V06NAhbW5uqlqtBkuyurqqVqul0dFRjY2NxdhELsfz5MpQLPYoov4VhaS4PpnqwKJ77AdhglXOZrPaqm7pI99oHDRUw2t78JNSRynVpf9kn31J0m9Isz/4udjI9/tSr/cpVatnlEpl1e9LOzsVra19QPX6Q5JG1ekMBBEhTNZrsjPcxsaGJicnlUqlAsmFV6DgwwkP5s8VCmFn7jxEYH6SUA6yCQHkd7yHk29OpiX5CxQOj+YGoVarBYTjCxqkoeHwYgViPJQLmXXlQznpv8exyDVxOmEIOXI+R27w2CAzJyIdJSCf9HW/65b2EAJWSDfWDjp2bjQaQeHYdtAhD1A1CWPY+Lher4ctM3wDaQghZ8IcJkTRcJULg4B3dIJFUmhbFA1ydA8//LBOnz4dJtENCkzfcmVZhfzaQUN1a9d3JP1ZST8s6a9LeoukrBRFUjq9q4mJF2K3z85+WTs7M2q3Z7W+/l6tr79fOzuHw+dugQuFgg4dOqSNjQ01Gg2NjY3FhJ04y72kpzwKhUKM2GPMPEnvcRljztwijF4H67tgeCzuZA9Czt/ICdIvFIusgTsDJ6u8rXhGh/woKqiM/gJFHdJ7G30HBGnIwyThNf93jsbJqWSMzpEab5qtHR0dVblcDh2jgc72NRoNPf300/rud7+rer2ufD6v6elpzc7OqlwuB0+VJJMoZoAw4PwIanTxvlhp3gdpQP0sz/XYQhrsVs96SmnoVYF26+vrev755/W5z31O6XRajzzySJgknt9oNLSxsaFz7zunzh/fuOD5TV1VSf9K0n+Q9MuSfmH/WYiivgqFZRUKyyqVvqv5+d/W6upHtbT0I6rVptTtDnf8A0KVSiXV63VVq9UwxniVYrEYU0x+WMsIuhgZGYmt1Bi0Jb5Kx8kSyBOehXDjPR2deMyLcfV40CF8FEUqlUrK5/Oq1+vhmQh4chWRowJ+2u12WFMLeuj3+5qcnNTo6KgmJiZiiAvDDNTmX/ekeERHXlK8os7ZYxCgk2EYjIOuAz+dmJi4AWejDPl8Xq1WS7/927+t1dXV4IEmJiaCQvX7/dgyJJhSLB7eTRoYgiNHjqhcLqtarYaJGB0dVbFYDITR2NhYEJxqtRogHoOKdXL2l0ICJpBCh9deey0Gt/CW1N8SF69Orqr/5jaG2/+qS/oNSR+R9ODNb48iKZdb1/z8v9TU1B9oZeXDun79T2tn57A6neGq/CiKwprE1dXVUM6HQuJZ8KSsg2Re/F68I7EecSLj6Kwx//dYy4simH9fF+qEnTSM7Twm5V88HT/cn2RSMVTcg8FwAglCiOMtXH5QQoe+nobxWJb2cxGv453J45PeQzZdeXEee103jTlplBcbELj/u3/379RsNvUX/sJf0IkTJ9RqtULAS6O2t7cDLE7msxqNhur1eohdi8WiKpWKRkdHtbW1FSxfOp1WuVwOioMFGhkZCZZwZGQkJmwwk8Qqu7u7qtVq2tjY0ObmplZWVtRut/VDP/RDOnPmTKDPKXgGlhxEdf+Jry1J39QtKSdXFEn5/IaOHPltzc5+TouLH9TKyo+o0ZhVux3fLG1zczOMH33BSzoXkM/nYxBrrxgJiJeMM508crKO1Jvvg+uCjTxJ8aWJzlzzXOJAYjoIGC8ZxINL8S1RPRziHvfwMNUo88jISECLtAvP6Ay5v5cxY6xQXEguaehtMT606U0rZzo9rD31ycACLS4u6rHHHtNtt90WJl0aLvny9AXPYEBarVZslz6sCHm/Xq8XU6paraZmsxk8J5M5NjYWO2QGpdrc3NT6+ro2NzfVaDSCII6NjWl2dlYnT55ULpdTuVwO0AMvvbS0pFarpZGRkUEfere2c/n/0Vcut65jx/615uc/p8XFD+ns2T+nnZ1MEOB0Oh367sUYXmuKsWs2m+GEMY/dEU5nwX0+kQsK4vGckFe+cgVCxsMj3x/KFd8LRZLECyENgo5MOsPt3AQX74eQYTyA9Cgt8bLnXTEUkGuD8R8aA9rHOIEM3SFBqHooABG113Wgcib308GSkAs8c+aMvvGNb6jZbGp+fj7AT4dGXt2DQlLvyPpJYqLV1VX1eoM9U1nRwuZbhUJBr7zyilZWVvRDP/RDSqfTqtVq2tra0srKipaXl9VsNoN1g24nrmDrCq+tBYLk8/kQg1SrVW1tbYV4Z3R0VIVXChpdqKs7uqm0OtItErf/R1wDuLuhY8f+tcbHn9Ezz/xF7e7eGYwX8+ZJdWIph2oYyvHx8VhelQtrT+rFCSLm2eEuix5QNt/KBC8C3IREYpWIP7vdbodCCIfPxMLSMB3DIn5XGt6F4sBC895WqxULB/CsXvE2HOsosLUgNcbY24ERwNjhRXEiGBwPF/e6brrY2icU78Pgf/SjH9X8/LxeeOEFnT17Nqz3HB0djS25ogFMCN602+2GrT9IlheLRY2Pj6tSqcTY2qWlJV29elWPP/64XnzxRR07diysD2QgpqamgpK5AnY6HTUajVCx4nDO4dDq6qqWlpYCXBofH9fx48dVqVS0tvADeuIjlzQVnVf7U22NjryuqakL6n+mp3JjU6lUT7omaWe/0fz/7xVFUqXyqt761r+jb3zjZ1Sr3R1jNt3QYmTxBNJwk2y8pBeUSMOcXrJyBkFPVo2htOl0OqYADu9oAx5wr7QCcDq5vEsaVl65gni5ICghmcekVpqY1Rll+pX0dBgV6r1J56DQtNUhMGw1yAyPzdjS9v2um+6+55DWIUaz2VQmk9HDDz+sBx54QNVqNbZzGeSATwTlYHivVGpQfsehsHTSYSrpjNHRUdVqNZ09e1bb29uan5/XvffeG7bldCvI+4FZQDp2WvPUAETD4uJi2Gal1xtUB01OTmp7e1tra2vq9Xq63J1VoXBM2Q9nlUp11e/vqnfbrhbmz0r9psrPXdOx8VekP+xr8ktXlNl9Y1/cvpTaaw6ykqYPmoHv/SqVVvUDP/Abeumlj+ill96vdnsmJsSuVJ77S447TKIXfHtJZTKNwD0oIcrqMSTxo8NP7sEB+Hel4UkA5CMdCfjKEl9JQ/+8msxRhJfsScOUlMt4o9EIyxidsKKvIDqgK4cgOYxmTFi84TEoRuNNl+/RaYdIWDlpuAs25VaZzGB9ZjJBDkRl/1bYLAaGwXdL7SQCVSenT5/WlStXNDIyoomJicDqQig5lKLYwa0tKQIgmtP5xLfEO5VKJVg91pvyLs/vZrNZnbt+96C9Cym9GEVK/2Sk537id/XF+/5AknTPC9J7vxwf23d+XTo2Xdb6fY8ovVgMSjEx8Zyy2S1JUhS1lUpt63u5okjKZNq6995P69ChZ/SlL/2Mer2T6hvdjBB6XMVYkAulBM+VEMIDL4ccOCnieXFP/vM5FyST5ykxBAi8nxEqKbCs3OvKyA+FKbDC0nCnP5QUkgxDwFI9V1aeAdICmtMm1wnuc5QSn5Ph3rtcsMVvOuZ0C4b3dIIHoWbQGCi31BQjOHxBQT3ZS3E23wNaoGhAVa+93dzc1Obmpqanp8N2KEBnH2QED5iBEEGjU6VEPHHo0KHAcHpiGqPD70Al/h4mIJXW73/oNb06Pxi7Zx6W/sVfiI/tketF/do3f1nFZ08onc4EWD8zk1IuJ3W7HfV655TLPamJiW9revoFpdO3zhxHkTQzc0Uf//j/S9/+9o/r4sX3K53OhXZKCvAOpWTsQBWMAcbI41BHVJ6k59+kgfYUhZclYhSZL9pGuMJcY6h5Fu3nsGTPcztU9AIEjI4Xmfi+S54jhSyEZHT21/ubSqViW7S4/O49L8OTs8vlcjg1Ya/rprA2WQVEYAtjSyqDz2B32+12qMQBowMrfXIRfsftNJ7BQgkYpIWFBY2MjIQ48sqVK4FtBDLjnT0ecSaP5125ckWtViscl8CughALCMzhw4eDh/ccGh7aE9Q7nR1tZw7weH3p7t13azp7rzbrm6rV6kHhJUrjInU6x7W7O69+/8Oamnped9zxGU1PP68o6uoAHsHmTxoZqettb/tXmpl5Va+99lHVaicURfETxJljX9HDnDNeKBOC53PH5RDWY0rmA7IFcg5IB5IBjkrDnCcpLoxpsvKLOQLF0WacCLDd5djlOuktXQYp4vDYlPf5iQJ8DimGPPM8DwNct/5EnpM8JGVTTBzkUKvVuiHp6wEvEAfms1gshobmcrmw7SDEDFudMIhenZFOpwOjSOUScAJ2t1KpaHJyMrB7TLBPAnFKu93W+vq6NjY2QrH43NycSqVSMEDECeTVUEQGGkHiXaSEGtuNGOpIXqPtUf3w+R9Ws9GM5YI50S2TycQOtE2nM1pZeYtqtYc0OfmcTp78lKanX7xlT5rN7urMma/r+PFndO7cO/TCC39WOzulGLz3WN1rkKXhqh3f+Nu9H2PhcNeZTEg9SWHvJRQVo8D7UTAPUcrlcphTjxWl4eoaaahgvr6VuUGePF0DMQiq43lRFMX2J6ItHiszTrSVueLvvDeplFw866AqoQOVs1araX19PXgPyp+AO0wgEC+TyYTtQ7C47NLNJNNY7oPdwytKw0oiSWHgW62WVldXVSwWdfjwYeVyOU1NTYXFsGwQdf369XBwEN8HmmxtbYX0TaPRCIZhZGRECwsLmpubC32g+ggvkky8OzzBQ3t1TDq1v3JG/Uid3eF6R3aLkxRiPSAgEzx4d1qrqw9reflOzc19S/fc879pZGTllr1oPt/SHXc8rvn5l/Stb/1F1Wr3Kp8vBSPKOSFe/ZIU3HQ6HYgjYr9kHjCpnKyEQcFRauQjlUrFYjs8r68eIoZkjD1GxJuxQqbdbocT3XASPMfRE2Sh1wB7mglUgWdG3jEwpJV4j/eZtnsNuTQkl5zP2e+66e57sJXU0TJJWHgnXGBL/fI9TVESBtir+Nmw2RPb7p22trYCHQ1OT6VSmpiYUC6XC0xrt9sNv3s1Sq/Xi23D2esNttyYnJzU1NSURkdH1Wq1Yoygb2DGlVyc7dCJMesqvulT8qrn6rqauaqKKoGQoLKpXC6H/ZqS+Twmv93O6OzZh7S4eEbHjj2u06cfV6m0ektKmkr1NT5+TT/wA7+hpaWHdOnSn1GtdpekbEAMW1tbSqUGG7oxn0mIyDi4V/KKGuaaaiRpqEQop+8SkYSm/l5pwI5yH8/jndyLwHsu1ZlkPLMXs/NMiEI/fxRZwDt7bh7vj3f2vLlDZHLtHgI4OjjouqUiBM9PJpdUOZOG8sBsQeI45GXAfWB80JN5KeLWlZWV8F1gGBaQTZ3y+bwajUaIRfFoHlfQn3K5rNnZ2bDZWK1Wi8EQL6DAGlKZQnsdErnQtbvtA5VTkr507Eu669JdIdaemJgIQumnZuFNkrFct9tTvV7Syy9/QktLj+nee39Xhw9/c5BvvckVRVI63db8/BOanf226vUFXbjwNp07d0a1WkmdzvCIAfaOoj/tdjuUQa6urgaDPTExoYWFhdh5OeT3CG2o7qnX66HSC0PM/OBBWRLmyucpMjcAKG4URUHmOC6EyjNSZoQRzB3K6yEKiMHZZI+58YqeIiJv7mQQxhRk6aSZcyH7XQcq517HofFyZ8W8EsQDXfA/lhbFSsZrXqTuFLXDKNIxVCI5wQDTSszmOSr+5X7uY8UKsQQT4YcZ0R8QAUKH8cFIgAaCYve6uu3abTo3eW4f7ZCeO/aczk2c0+2bt4fngkJgv52t9JwiY41X394+qe985+e1s/O/6bbbfl9RdHMFHQiJlMnsqlI5rwcfPK+77ipoff2Qrly5Sy+99DatrAz6WiqV1Gw2dfHiRb3++utaXFwM8JBlgTDuMzMzuvPOO3XixIlYG5mD7e3toDQTExNhfhF4D2mGSGF4XosLNWEWsBnWVFKsaAFEhfwlldJjRnKSDku5D2jscoAuOFfChQFJGuok/N3vuukeQkmYQUOlgWcl/+fWhVyTu2+EipUKWEGPOVESz5d5ZcXa2poefPDBGyh8hyWpVEqTk5NhAGHVPHZI0uZSfE9SnkuZGobBq2Tw6A5rfC3kPVfv0Zfu/JI6mb2hy25mV3941x/q+DePKxtlw+Q680sMhDA5RHehHdyf1YULf1ntdl7Hjv2hRkZqB03tnlc+v635+Yuam7uoe+75ml555VFtbLxD166d0KuvXtDVq1dVqVT04IMP6sSJE6FAfHt7W7VaTWtra7p06ZKefPJJvfjii7r77rt1/PjxIOgsmvdKLRADY868YYwYA7/ov9ftukf1PDoOxg0w7/I6X3cWtMdjYwyR8w1A5CiKQljmpJEvyt6LJMNw7HcdqJx+mlfS/XvAjidh20QEiM+cHKJxUOQE3b4rnMMYBgyGdWFhIbTFE82eY3VWjgn3BLhPqMMXFAzhwOp5raYLDF4DxhEUIEm3r9yuSrOi1bHVvQc3kp499az+TfPf6Me//ePK9DIBgjn7TZ9gxx3SAeOGqSHptdf+vM6ePaF3veufaWzs1uLQG5r2RgrmLW/5grrdL+ncuWNaWvqQbrvtnTp27EQsr4d3mZiY0KFDh/SWt7xF165d09NPP63nnntO1WpVDzzwQJAhUFa9XtfS0lIo9Oj1hscq0HdXPqAjcJL5wNNhcImLga9OBDnJ5dwChsBjTpQrWbjPe5H/4ZgNS/+SvzvK5F6Mix/0lLwOPFrXXb4H1F6LyID6GSpOpDBYfnIV3oeG4jGAlG4ZeZ6XSSU9usctvn7Q4SCDSSUHk4OSJZdUeWrB4YmTDt42vDDes9Au6Mee+DGluvsPcS/V0+P3Pa5//cC/1mpuNUy8x16SwiLopDEkxUBsjsHb2rpfv/d7f0Nf/vJHtbU1qZuEvwde6XRXp0+f11/9q/9S733vH6hcPqdOpxEzVF4F1O/3dezYMX3kIx/R/fffr2vXrunJJ5+UNFwAjxHc2NgI8WLSATCezIMLsRengL6ScJVQA3ljDpEzfpxtBqm5UiWXgHm6B8TnHjVZrMHlf0sioP2uW9pUOoqiGFOFAtAYlMpTCeTEsHieM+PZeD7f3hLFGApHOhQ7UGaFcXCr56QPg4GiM8G7u7thNwXiGy9boz9YWY4kSJIFtNWVlQXc5Oza7bbuvXSvTi+e1qsLr0r7eLB+qq8vPfglPXfbc/rg8x/UR659RIenDocF0xg4FMAZaGJxEMdwAUFB1eqEXnzxo7pw4R1697v/g44d+44ymd0/kSc9ceLf6fjxT6tafVDnzv11dTpngqd3go8c8UMPPaQjR47oqaee0hNPPKHHHntM5XI5VsTiywBdDhw+UtDiJ7eRYvNKM+SCuUO5WKOLPHkcmPSIvM+hJ8q9FwlEW7nfeRDPb+II6BM1AiCtva6bVgghxATqPIwO+/o4hFtSiC2TVSTOtPIdlCiwnVb7CNlz6NCh2HImlK5UKoWBcArbn8Hg0FYmAaPBe5xsoD2ugNTqSopZU7w6noQr1U7pz3zrz+gfTP4D1Q6KASNpfXxdv/vY7+pL9S/pg9c/qA9c+4BGdkZi7WS8ut1u2FrUV/J4qmLwb0ap1FE98cRf03e/+4ruvvs/6vjx55XLvbmlM1EkRVFHlcpTuueeT2pl5UO6evUntb09GSMMuXK5nI4ePapcLqdvf/vbevLJJ/XII48EpMR2MTMzM2FLG2dHIYs4u5N5xvuBeJg7xgbyEafhaSnk2ZXMWWHkFZ7ESVCQIvd6eAPn4RyBr3Rx1OPZiDddhODbCUZRFI76c7KERvT7/bCq2xfcSjduqut5TlIHkmKd8Y7k83ndfffdKpfLymazsWPkPAZ2QsEZQjyZbwLGPR4bYxjwpnht3wGOAXU2F0u71ySeWD+h97/0fn3mwc+olz6YRe1kOrpeua5/Mf4v9Nkjn9XdV+7WW198q+7cvFOp/lBQEE5HGFhwtuFw75PNFrSxcUaPPz6vcvk1zcw8r4ceekaTk1tvypNKUj6/pYWFf6tS6ds6f/6ntL7+bvX7QwbVD0IqlUqamprS+fPn9c1vflNvfetbAwfR7XZ17do1nThxIrYOGESAIcewetyIDPpaUif0kopHCOKK6cgKz0wbMIzIgyu8Vw15qOPxqaeEHHWlUqlQSx4dMAE3ZWuxVBQBQwZ4p7AWkmLeELiAAier+bE+yTInvJlT5g4vfbLcU/nxAbSHQYMN5R0ouDSs7cTrsIzNdw+gmMK9M+8BTTBRjE8UDSqBPvqdj6rdbutzb/3cTRVUkvpRX6ulVX3ljq/oG6e+oXsu3qMPvvBB3d24O9R3EnsDbTGeGBM8BfBpwJRmtL5+RhcuzOpb37pPDz/8os6cOadjxy4qkzn4xKu9riiSKpXXdO+9/5OuXfumnn76J7W7WwxxO9U35IvL5bLOnTunKIr08MMPB+O7ubmp5eVlzczM3KBcvtyq3x9uQu4pPrws8uRyRCjgBfWe3vM5Zcz4rsuiNIwx6RvQGWNNZsH5Fb7nKE2Kb5K233XTs1JSqVQoKHf4ieeQhl7KGdIkpnfo6Wyq76bGPU57Yy3d/ftCWawlE8T3WanAJAI7sHJO2ft2nbTD00Ael3pRAt/HSLhgYVnz+byyyuojz35E7V5bjz/6+C0p6ODhUjvb1rOnn9VLx17SPZfv0SNrj+i9a+/VWGcszEWn0wmrg+gfXsfnkhhsfLyiajWlJ5+c1tNPv11Hj57Tww9/TadPX1Eu970raTbb0rFj/1HSsl5++UdUr9+j8fHxYKiBddPT06pWq7p+/bouX76s22+/PVTrbG5uamJiIuSpnSdwpQEROFPMvPueT8SEnnvMZAbb1GxubgZl4pnuIZ3E9JgSlIdnJ9MAkpHiJKSTn66YUnyVzn7XgcpZKpVC1TxKQAM9qGaAnH1iYlASGuwxBQOHdWOwk5bGC67xfsSUno/Cw+3u7mprayuQBmxe5ekap7s9XqE97iGdmCB2ZUKdYXRCBOOCgO3u7urDT39YuWxOf/TQH926gr5x7eR29MypZ/TsyWf1h9U/1IfOfkjvuvYujWs89AVh87WD9MsZUUixwbF5ab300jH98R//gB59tK3HHqvpxImvKJ/f+J4gbxRJx48/q9nZSzp79m8rnX5PGHOqidrttu69915dvHhRa2tr4bBfDOvu7q5KpVKMre73+yG+RnaQSZAYvISf3EbBgTP+jAEGzRl/V1DQEM+D7JOGeXPCLTIIo6OjIVXoxA8E417w1YnSva5bYmulYSGBFD8kx5dl8W8ypnRm05P9LjieE2XAvAIH78t9vqSJ72M4+v3heaLuyQZCNCwjY4J9vZ4jAjy6fx+FTlaDOKNNEhzrjdJmO1l9+OkPK5PO6I/u/yPtZnb1vV69VE+XK5f1mw/9pj576rP62PmP6T0X36NitxgMIhdjhUEBIUgKO1Jcv35dtVpNIyMTGhl5VC+/nNWlSx/VoUOXdPz4v9fY2MtKpW7dmxYK67rjjl/V7u5vKYrerXq9HmLL9fV15XI53XPPPbp27ZoWFxd15MiRINT9/mAHRiA7RBCLFDxESebTnUHFADscBhE58ecG1zcU83fBrBIjg7BwVnhnDIzzELQpaeS5IPT2u256srWzTzzYc5AOM1yovSAa4QCbp9Pp2GlhKL7Xwu6VvuDZvj6PeAsllIYF1l5S55aXe4DMTlx5zs6Lux2K9/v9UETvkJ7VJOPj47GjDHlePp9XKkrpY9/5mO5bvk///NF/rmuVa+pHbyIRGUnXKtf0mw/+pr499W197NWP6c6tO5Xuxzcu9piIMIExJR4cHR3VI488Eo5vaDRmdPXqEa2uPqKpqW9pYuIpzc//sXK5+s2bFUnZ7JLS6b+kfv9fqtt9WN1uVzMzMwFJFItFnTp1SufPn9eLL76oe++9NxaOlMvlYACz2WwowE8WICSVzZWKsk1p4KEKhUJYbeTF+E4c+iJ9DmhCJqXh7n+Mr6O2JPfC/R57upLShzfN1tbrddXr9bD1iFsLjmgjT4kFcXoYYcXDEg+450Op3esWCoXwLCC0WyVXOI8HPU51yOIWC0/i7yS24H5njekH/drZ2YkZGmeV2VoTi42CMEYBXnZSuu3abfrFz/yinjjxhB6/53EtTizumws96OpHfT117Cl95/B3dP+5+/Wjn/tRjafGwxEXvV4veK5eb1AuhhdYWlrS+vq6zpw5o6mpqVBbDBHWamW0uPguXb/+Dl29+mHdddc/VqXyyi3B3VTqqvr9P698/gtqt2dUKpVULpeDMZ2ZmVE6ndZzzz2nmZkZHTlyJLSPNB2IJekZkS+v9CG9hFzyO2PP3FA5RM7ePZt/h8udBOlBJzrdebhcIT+kWjzDMRyj4Tabe10HKid4OinIBLrEf0CCKIpiDJZ3rtvtxmpusTB+8AwKgEV0JpT3830Mgk8UXs1zSAwOys6FgntFB/c75HFGmTZlMpmYAGE8uIiB8FK+ibPHubl2Tu968V164NUH9NSJp/TVB76qpckl9W5hZUnyamfa+vapb6v8QFkfePIDajfaMUPpyfR+v6+lpSWdO3dO4+Pj6vV6WlxcVD6fD0qEoSScqNXu1hNP/G3df/9v6tChLyuVat6Ckl5VKvUrymb/bmyLGMKQO+64Q1evXtVrr72mmZkZjY6OBiKNhQnU4TJnPuf0y2WQKh8W41NUz7+eKXBWmHlLoiVn+lF8jv1D7nAQGBLGGQ7AHQqyzHf/RIXv3lAvZdoPSzPwxAVASYQ+WY1BXOoxrVtLLqwUXtghjuePGDQvMZQUo/eZHPbPzWQGh+TMzc2FQfXCCwTCg3uqQRy6eBzEej8vrvAkuCtNppnRO777Dj187mF9967v6vG7H9e1ie8d7vbSPX3lnV9RLpfT+7/+fvV3h9/3eev1erp+/bra7bZmZ2cVRZG2traUyWTUaDQ0NzcXq2FG8HZ2drS09CtqNj+pw4d/Q8Xilw9U0CiS0unfVS53lzqdvx5jujHuDz74oL785S/rtdde03333ad+f7D/LMaPhczMFbLmHshREZ4tWZDBJm1sq+NoCCRH+7wkjznCqACXidmd1UfWcULIiRslxpM+vGlCyLdd4MHgat/8yt28xzQ8A4bNA3SP4xyW4oWSBQJ4L/eQSaY1mYohp+W7DdBeBq7X66larerChQt6/fXX9cADDwTLR7kgbU8yfU6bS8OCB+5zBtfzYPyffiEk5U5Z73n5PXrs0mN65vQz+twdn9PVsavfk5L20j19/q2fV7fb1Yef+rDyqXyM5k+lUlpfX1en09GZM2dUKpVihF2j0dDy8rJmZ2fDmTf+02z2tL4+pW73f9CRI39TIyNfvYmC9pXN/n2lUo9JWgjCTVsWFhZ06tQpnT17VnNzc5qZmQm5WxZSFIvFmMHsdDrhcCJkRoorqadA2DxtbGxM4+PjQba2trZCDt25B48heS7kTSqVCnKFAfaiG8ggdMFh8I1jE+37mXQL5XtURJCWYHAQPGJEh6vJyhWE1WNFx+RuQdx7unB7Tgj2jBIuvucxIgyfk1WuKCgm302n03r22We1s7OjD3/4wyH2wFMzodJwdb7HGP48ICEhgJMCjInDGS+QiKJIpU5JP3Tuh/T2K2/XE8ee0OfOfE5XylduGe720j198bEvKp1J6+Pf/ngsbOh0OlpfX9fs7KwWFhYCWykNVxo1Gg1dv35dc3NzYZsXLu7pdqe0tvb3lUr9vAqFLx2ooKlUVen0/6ydnf82xoIytvfcc0+A2YcOHQpz7FuawDTz/VqtFkvH+RyD9pLzz1hPT0+Hc3bYNcMXL7jhd+eAweX/KCEyTMrOv0t2wFFgkoDc77opW4vgIITUl7oyOlPqKYtkRYR7Ni9KGEzg0IJ4+kUa7pyA0NMpL7OTFHJjlH55RRDemOfCzAE7oPCfeuop3XfffTp+/HjIafE+oJILOu/3eJLf6bMbGmcKkwSHw/9+v6/sblY/ePYH9d7l9+orc1/RZ099VlfGrtwScdRL9/T5Rz6vE0sn9Pb1tyubzWpnZ0dXr15Vp9PRyZMnY+PrFTYU8ler1RjhBlpAsFKpWS0t/Yamp39eo6NfO1BBR0a+qImJB1WrfTCMIbFkPp/XyZMn9dRTT+n69es6evRoMIwgKWAkRp01tqCYKIoCM+4OgN8JZyh6oKyQk8BYL8wBVx4fejvwkjyXc39wFOPj42FepeFG3eiQh3UHkUHSTZaMAeGAB56Yp6PEXZLCNh4IM5bNLQ7P9UWweGAGSRoG/wysM6i+jaY0YNnW1tZiewc51EkSQfydU83Y87ZSqajf7+ub3/xmgN4ea3qszWR5st/7h0X16pIoGlSxsLyLz1je5DE+tH0ul1OpXdJHr3xU/+Mf/4/6a9/9azpWOybdAtLtpXr6nff+ji5MXVA2m9XW1pauX78eNsxGSBG2kZGRcDwGecatra0wV3gixrxWq2lzM6dq9R+q273jYEFL7Wp+/t9oZGQneGNnTtnulPI+5zEkhSovFBNiBgPsHo/5AM7i7T2swQCiODMzM7r77rv14IMPhjjcZRjnhHGl/Z6yIc9N0QSXx68ugweRQdItKKcLqbN95Ak9XsEbIdDkjjxGoyYR75Ise0picKyQDz5CzP4xTuxAJHh5nsd1hUJBhUIhbOoF4bC8vKxOp6P5+XmtrKxobW0t9Nf7xDNheiXFDIj305d0wSLi8d2oDYR3eEResVgMcS5GIFKk4k5RH7n8Ef2dZ/+OPn7148p14zuL33BF0sbYhp4886R22jtaXFxUsVhUqVSKrXd0EgRPCYHGERsIG0JOyeDAIE+p0fhb6vUmD2zOyMglHTv2L1UsDne7w1h1u12dPn1a1WpVV69evWF8QpdMsDHUlM8lZQd5SpIv9NVDMwr1Dx06pHvvvVd33323ZmZmQjjjsNTzoE74+PPx6m6cHc5ilA/yngcqpyeyPUDGkvR6w2VYrlxOdCTpY4eDOzs7sf15fP8YOum5KFdO1gJi+Yh794oLGTwmD4HkOVevXtXa2lqYxHa7HdvwK8lMO/tGH926urC7UhK3IEzEKBw3gbHI5XLB6vuzUP6x5pg++con9Zde/0vKdvc/tnzQYOmJhSf0/Orz6na7mpubC2PIvCCgFKgTK2NgNjc3YyeDI1wgpZ2dHW1svF2Nxl9V/4BThqOop9nZ31epdDGMH4q+s7MTtj55/fXXYwvekzLJWKMAGHmHjBgefphb7mM8kUc/kTuVGhSS3HHHHbrrrrs0NTUVxqjVaoUjKZ3MRDbQFQyz76DvMoThPyjmPFA5GQwUgYFy3C0NN25KkjoeDzpL6ykZhysOBfi+M3IMkFv9ZDmWM8v+HofbWP1Go6ELFy7o4sWLMbq+Wq3qhRdeCAoKCnAvHgoKLM70HxTSg35iJ9rnFpdx8PHw/vOMoCDtnj5+4eP66Vd++qYKul5a19PTT4cYK6mczBl/813UGY/19fWw342jCeSh2+1pc/NPq9M5pv4BkDuTqSmKvh7OuaGPGIW7775bzWZTFy5cCPOclAtPfSSVgz6xAyPjzlwly/6c+8BRpNPpsAnczMyMTp8+HU61850yPN3nJBDyiDJj9FxBGb+DUim3BGt5gA9kcnJQPJQBhpPvAGddoEnJ+Ap3hNrv43eegbVEgFA4AnegDgrrbQdmDqz9hi5duhQKJ5ikkZERffvb39ZnP/vZwAo6jOeC4XNk4QKFdeS7Xm9L+4HafrI0beV3h2axNEG7pw+++kH9wIUfODgGjaTXHnlNY5Wx4Z9sXPH8jkokxci/VquljY2N2IbXu7u7qlar2tjYUK1WU7U6oqtX/9/qdKYOEitVKv9JV69e1NWrV4PgMvezs7OamJjQxYsXQ4zoV/L/9MUJxFQqFVu0wfwwR8SRLL1jt0g8njsR5GJqakq33XabFhYWYocCYzQ97PKiB0eS3MvY+9Y+e10HsrVOs4O9e73h2SAeZ3ldrDTcs9TJDYfINM7hFIrthAtBN4OBhQXqOZXt+wehsJTb4d05ICeVSml1dTWkZaQhkYOSPvPMM+r1evrJn/zJGGR16+uEljSsxOF3+u+F8jwDxYTg8kokN0he1OGJ8X6/r/Z2Wz/+yo/rpamXdKlyad+5XJlf0cbkhuZqczEv7J6H52MomA+MAR5jY2Mj7GQAuYcQ5vNj2t7+sG6//V9rv4L5iYlXlcs9q6Wlu7S1taVKpaKZmZng2Y4fP65nn31WFy9e1L333nuDgWX8kUv3Pt4XD3Mckmezg/NcXaEw9Mi38y2gqlQqpUOHDml0dFRra2uq1WqxiiDmj7lMFp9AijpySi6g8OtAz5ncBQCcjnCgIMm8pTTc/BmomSw6YGDck/A9CBY8Zb1eD/lU6lshn/g+a06dTQbz483wkMQrCJcjgYmJCc3Ozob9cZ955hk988wz4TAmbz+CzdESyc8gC3zVAqtDsNhOdjnUQpiSoYILIcZvcmdSH3ntIwd6z8ZIQ+dmz8VWVGBIXNidZEERfH673a6WlpZ06dIlbW5uBqOHsWo0mnr99Y9oe3ti37ak07taWPiKUqlIzWZT165d08WLF4MRnp6e1ujoaPibx49c7i0ZKzzW1taW1tbWQuEJSsFcQHCy8TmMrjsX+uxyzHtHRkY0MzOj6enpkH9l3smxcw4QYwYC8RCI5+13HaicEDY0FDcMPMBzJtMCKKynRTweQIj5LgLA5wgJ72diyFHxPNbRUWzuA8nEOCTkuQyWM30QNEePHtXx48d1/Phx3XnnnZqZmdEzzzwTjqL3fqEgnjN11hVFRDnxDMAoL1ygX54bRfDcmEnxpWx4gfcsv0eH6of2n8xI+vIDX1YtFU/eg0h4Lu/iAqJ7bMzChHK5HLam4ZkDuDil69ffp35P0jclJRxoFEkLC99WuVwP71taWtKrr76qK1euaGdnRydPntTOzo7Onz8fK3xhPFAuj8slBX6kWq2qVqvFSkHdYELCUeWDDIBi/HJIyjszmYxKpZLGx8dDasjlnL18vUqtVqsFEhMI/aaLEFBAXD0vR7GSMVAmkwkeCi/nCXeEDe9IbSJC78QLbJekEA8wSAysp0sYeN7vlT3O6PlKjampqfCOTqejUqkUVuOz98uRI0cURZGuXbumO+64IxgLxoB+ey0x8Zuz2pnMcLt+DIVXNznx5bGy5+L4P/86ATGRmdADyw/oP5b/477zeXXmqp47/pzec+k94XssPPCxJFZyI0QKihUeGDevxHEFvXLlYzpc/7wK/9dN6VOS7o63ZWRkRUePPqte78NaWVlRtVqNbQc6NTWlUqkUyvqKxWLYo5d2JkMN/o58OUO+s7NzgwxzoWxJZliKp/hwEJJiddXJMSAs6HQ6ajSGJ85hcMfHx0PbkxkKv25aW4slZ/K8YiRpZYFZnoJhspPxjac1nKlMpVLBwjCAvoUlMAzChwGBacNKSQqEgMckvMMLDCCB8AJANWeTvfKJdjNxvirBx0kartlzCh/v69CMvC/PREkZf88xQ3xxP7tNnKyfHEDbfZBSL9XTl+//st5+7e3K9IeHUdEvfpKCjXD3+/1QUgdkA804Qz1AJ0d18cn36/ZXP6XoH0r6B5KMmIwiaXr6i9rY+IRSqUPhnJuNjY1gFMbGxnT+/HldunRJx48fVxRFN1QKuTel7Yybp9OYB1fiYVuGBS9O8jmi8+9i8GmPs8bAXMYIGZAG3hQHgUIfVIhwU1iL9lNrSiNdWRAaXxbkuNqPLGBgfYsJ4F8ulwtJ6PX19dj+P8Bq8n9YOlhXmGDeyz1ASknhaEBW3uP5i8WiyuVyUB7SLA5R8/l8LLZCyTyVkowvGR/awoTjJREaJo/YSBomypOpBBck/64kPbTykEbbowdNqZbGl7RYWYwpNwrqQsa7UVDK1HwRAB6WvyHcQNArekS9mbT0byU9rRti4krleeXzz2pyclInTpxQpVJRtzs4JY70Vr/f17lz527YIpOxYPxBAc4reF885+mXx5agA0mqVqva3NxUo9GIFTwkc6SkXSYmJgLEBe77mbToAU4Br7mXt+a6pcJ39xAIGLWEftaDC5xbEqyD53nwHKzBS6VSWl5e1uuvvx6El3eiHPwdofLksedj8YIYhG53eJQB7ZCG6y5TqVTwPqyW9xi1UqloZGQktrWJjwvvSaZC3DMjULSdH4fySe+F93QhIxzAeuOtU6mUKs2KTi2f0nNHntt3Tlv5ll6Zf0UnN06G+WXuQDn7pXQgPGg34QKfM760abN8QlemT+n4S69KPyXpi5KOuXx1dOjQt1Svf1idTkdTU1OqVqshr8qWmtevX9fS0pLm5ubUbDZjcaGPjRcSONHloYQbNZCDIytfVVWr1WJ5cjcGTiTitNxgc/KdG4lerxc71QC52e860HMyCEDGJJzFiiUFilgP7M/vURTF2LMoikI1zNbWls6fPx+YT0kBKjkBwCAQbONBfdPrcrmscrkc232bmNY9F+/m+wTtDLY08LYnT54Mp3szwZ7bTRYneF7UGU8MijPcnsjm2cmCCs/X+eVx7Pb2tjKdjObr8zetu33y5JPqp4cCy3zwTgTO55mxrNVqsTjJ+5wkzLL5Eb36sUfUL0k6J+n/KWl72I4oksbHn1Y6PSgPxOswXru7u5qenlY6ndb58+ffYIMboWTTMwj0gUXaHvb4vO3uDk4PaDQasYIVnwOcjy/QZ2x4J/MG28v2sRw3gaflFPaxsTFVKpVQXkpBykF5zgOV0+lfYgviC7bHdwuCpxwOfnxhLELsxFCvN9ia4urVq6pWq7E0QjIeQpDq9XrY0j+TGRz+A2PLxtduTHiWFwmgmAxuOp2OrXTAOxw9ejS2nyrfx1vTb0+fAMO97z6pKKDDRhcSfk+mr1zIHKF4En/h6wuKegcvW7lcuaxLlUuxIhGPy3iux7Seq0sKrMd8tBvjdfmOu1X902/U3H5K0la8LcXiy8pmPxe+D4PuAlypVLS4uBjqn+v1ura2tkLxAyENRQVem+yOAD4CPgEv6Rd9QD4xgOPj4+HYCEc57jQImzhV3bmWvUgoXzW113WgciIMTDwv80AWgfMtOZhkZ8CAB3hW8HoqldLKyoo2NjZiyXGwfDqdDoUKntpxz+zeOrk/rUPsZNUGk+Zb+xODRdGgKuTYsWOh7SgTZIhDJCcO8KZeROBQFmVmbGmnex+UmHZS0cJzGVMEOJ1Oa3NzUzuXdpRrHVwQ30639czhZ8IzvP0OTSXF+u0ezb0tbfW+h32IdsZ1/p4H1M9KelTSWLwtUdTTyMhX1Ou1AzzHazFG09PTkqRLly4FnoE8davVCvGwywXjggxJiiEskI6HO/TfOQU3XCAqX7fJHPI9jASnE+yVn3UZf9NFCAhOsgF7xVAMqBeh83KUAEXA0qB4169fjxE6FBUwWZBSnIlJzIfXBtoBmd3LQRAwcbTTq3LIh/oRE5VKRXfeeWdIGyQTyQ5t3cp6VQ2CC+pwYWbcgIVJCIbnQuBdkRCCZDsuX76s8dVx3ffd+w6aVimSXpx/UTuZndhYITReMMKFp3d0IQ1DH485pSFKyucLWpx7p7rz+xuMsbFvKJcbnCXjoRBjlc/nNTMzo+Xl5YCYGDPuQVk9bUZZpD+LNlKkgmJ71sGRCbLG35FrEJfHpB6bZrPZsF2JG+K9WOL9rpseAZhcA8ekeD6HDiRLk5zdRdikYbI+k8mE6h+gzMjISIykgKTBUntCnwFjaRPbTvi6Ol8LyoRhdJrNpur1esidFovFEPfcfvvtGh0dDRtM4Tnoj7OcKDukVdKTOqzxeNLjT/dc/NBPRxRMKp8xL6urq1pfX9fk5KQevf7ogUcPStK52XP69olvK5sf7sBAW+gTHseZXd8oi9jaFzlH0aA4ZGpqSjMzM4NtQice0MYv3iedktS4sS2ZzKZGRr4ahN7HmDbNzc2p1Wrp8uXLN4wx8uXsNmPksNPnwPvgyxj3QkFeF42B5HMcRJLFRc59HPfSrzfN1tIhBIo1b1g2BonfESQU06EruTJpuPdsJpOJpSwgPzgO3hchS3HhhNCRFNYbspUK9DcxsBNB/Pg9HneMj49rfn5e4+PjsdKxTqcTOyfGK028hIv2MdHu4TwuTcZCtIO+8ZwkUcEY8Cxqjy9evKh0Oq3Z2VkdaxzTbH1W18ev7zu33XRXv/XW39JuelcPvfaQMt2M1FfwMrzfhQgBpF2gFvoEFCXdhWdIpdLa+uD7Nf1/f1rRHnIaRT2Vy19Qq/UjwYAyBoRCExMTqlQqunz5sk6dOhWUxI1Uctzpg6M3kArtl4YpQ7yeKwwy6A7GiTOXR2m4G4dv8rVfXAni3O+6pcNzsUikPZwwIK5k8jzR7tDVFTJp+ev1eoDAWON8Ph9KnaIoCgwb9bAE/lguPB6DnclkVKlUAqVNX2q1WvCWSYWSFFg1SbG+e7wIWeSVSk7bS0Nv47CVy62zez9PgLti+CS7EaRtS0tL2tjY0Ozs7KDKqTihRy4/ctDUSpJauZb+2aP/TL/+4V/X1898XbvF4ZEFjIkL3l5QnPYxb54+QIAHAn5KSqf2LZAoFL6rTGY1VrDhKCyXy+n06dPa3t7W0tJSiNVIybmsOZsMfOU+R21OEuEM2PUBtILcNBqNsAyNefLlix5DurG/mYIeVL530yVjnr+hEIEqDQYEYcNDJusTfccEn8xer6djx46p1WppbW0tDDYpjc3NzXD6sSeY2ZVNiq/wIJYbGRnR2NhYoK2x6rSBmkqMAcowPj6u0dHRWNzqFD0BvhdPk4xPFsU7THUDkCQE8KiebnGlQBHohxQ3Gs1mU1euXFEmkwlndBZyBU109i88j13RoKzvX733X+l/+sj/pK/e9lU1c80Y4kFRiOVQAuJ1Z7BRWIQdJajX71SrdWzfZmSz11WpfCrU7RIyeDnk5ORk8J4eokhDI0eY43XYxNRuZDFukmKF/lzuaRuNhra2trS1tRVSMB6euLF1pHTToY+Gq5n2ug6EtY7VceukK6QhzExWbEi6QXk8lnKhHhkZ0V133aU//uM/DkpdrVZj1RNUBsGCOfxAEXwLQodbTDIwttPp6Pr161pZWdHc3JympqZim2LTLvd2XhHl1TRYVVABk+tlWw5jPUTwdIRPpiuiw0faxVhyz/r6uqrVqmZnZ2OlbdH3uH18L9XT5ZnL+udT/1yf3/i8PvHdT+jtS28PUNVRgbPXqdSggINwBwTgRMdgdVBKm5vvUbF4QVFf0mVJi5IelhRxjMN3wsbWW1tbgRl2Nv748eN64YUXtLi4qGPHjt0A89Pp4alrnoOMoii2tabPp/eH0Iu8ttdSY5BRKsI2Z9nxyskUTfJCLvbzqNItHAHo+TnPaTIJeJYwyb3h4bKutHQEZcVqdTod3XXXXTp37pwajUbIe5KrYlBZBVEsFgOBxCD6Gk48aLPZVK1W0/Lysq5fv65qtap6vR4WB3e7XT322GNaWFiIeX1pqFx4SW+3e0z67TG29x+447Aw6SmdueVzJzw8fUEfua/VaunChQvK5XKampoKytnr9VS4XFDxdFGt0db3dMxDL9XTpalL+sdv/8d69tKz+nNP/TmlusMtQiXF5pDxwVNlMplYHrjX62l9fV2VSkUbGz+umZl/r+z/tyr9d5LWJN2uQfumpJ2/t6vd0d3Ybg2sowQuHjlyRBcvXtQLL7yg+fn5ME9JogYGF5lFwSEXYXU9nEBu8L48G7QA8Uk6LRl/JyuC9lM8J90Iofa6bkoIkQpAITKZYcGuV/0QeNNgZ7Y8Yc6AOKUdRZHuu+8+Pf/887p+/XrMayKUWHBob4SBo8vX1ta0tbWlVqsVdoyDNCKGyWQGy3wWFhZ07NgxnTx5UtlsNraPEfd5bEk7GROspw92Mg3iiuVeM5meQAm9Uij5b9Lr8uzV1VVtb2/rxIkTofgik8loc3NT0y9P6yde+Am9euerevXhV7W6sHoLm9IMr53cjr566qtaz63rw099WLct3XbD+4HXGCnPhTty6nYHJ4zNzd2mev3tqqx9TtHVN15klYbtF9ravn87sL+UCzabzWCo8/m8jh07pu9+97taXFzU4cOHbwgbMKLEm55RiKIoMPC9Xi/EkV5M78Qb/3degb87dJcUk/2DPGLSy+53HaicThFTRofCeRxHLEg6AgtEgwmm8SbNZjOG1fGwWDsOgiXOzGYHe66urKyE+xFmBpld5UqlkmZnZ0MJH0XIwMJ+vx+K39kXyUktZ+ZoUxTFK1eclElabUk3WGb3hh6T8O4kacC4eOyJp8YIdDodLS8va3JyMiyl6vV6qtVqunbtmtJRWtNr0xr/0rju/fq9uvTQJb36yKvK5AZhyfnKedVz9QN3k+9HfX332Hd1ZfKK/soX/orOrJ4Jsa40PBEAgXXFpF/dblejo6NvoJZdXb36cyq992llR1fjaZV3S/17+qrVamFnRBBbs9lUpVIJ0Hl+fl4XLlzQhQsXAmJw2IlTYZw8HYicRtGgdtZ3bkRRGG9HeyiVzzX9Y0EADossgX+X+8PY9oe7S+x33RTWIkQohDT0qHgbTqz6xje+EduImGd4ftMnj889kT0xMRFbIgZ7SvJ7YmJCd9xxR4BPsHt4F1+xgtXzhDKekP1GHVo6c+ZWzb0+l6c7+BeY7zCXz9yTcNFv/1sS2ibL46JokG66fPmyNjY2dOzYsZCL293d1fr6etjkOAhRJ9L9z96vd7z+Dk1NT2l+bl4r5RV9Z/w7+sPTf6iL4xcP3E1+s7SpTz/4af3sV35WhX4hwGxHA8gDQonxlgZrM9fW1t7IUR7Tym0f0/zHfkvRv33jBROS/ltp5MR1TS39tkZHR3X48Jja7aPa3U2Fw2uZg7GxMS0sLITKsrGxsbDoHJlCWUFcpEtoI/fjSR0VeMmkr8LBGPO5E5ugx1QqFXMY++U4Uew3vZ4TgS4UCpqcnIxR3L4ifGdnR3/wB3+gtbU1zc3NaW5uLiiHV+HncrnAKFKczO+7u7taXFzU2tpagDLk7WZnZ7W8vKxer6fjx4/r5MmTYXMpx/4wfc7cYQhSqVQo3oZNTDKmnpwGwqMw7rWcUXUl5PJYZC+FTjK4CJCnLfxzNzBeVTU2NhY2gZYUNtZGWUgxAc9HRkZULg0WBCzsLGj68rQevfaovj73df3RmT/SpYlL+8anLx97WY+felwf+u6HYsLlS7Q8KR8E7I1qrygaHJZ09epV9Xrv0eTPflqFpzYGMefPS3qfNJa6oLGx/48kaW4uqxMnxnXu3Ht1/vyPa3t7W+Pj4yEWReHX1taUy+XCUZVJ2O3cAKV6xOap1GAjMKAtRfG+Ez3z7mjHHROyQw0A+oEc+QJxrltlc28ac7IVyMzMTBAEP5JNGqzcmJwcFDdPTU1pampKExMTmpycDOkMSqlItQBxEaLNzU1lMhmNjY1pdHQ0toaOInt2gNvc3AxwAK9Jex2SYr0drngFh+cigSJ4Pl9RwnN5JoqCh3aShzAgOQneRu6FiQQ+eczrMNe/l0qldPbsWTWbTR0/fjwwkghqku3luxQGoKyMRaFT0PvOvU+PXn5UTxx5Qn94+x/qSuWK+qnErndRX1+++8t66KWHNKaxIHDuJbwYRYrvaI4xHhBy2+p+KC19QdJhSWTeepLe+Eoq29bIyKruuef3NDPzgi5f/q+0s/NWbW/vBAXa2dnR5cuXNT09rUajoWq1qvHx8fBO5gwD6FU+GFj3oOTOqSzz+DOZO2VuqS7iYu5RYs/bJi+yCvtdN1VOlIpJ9RInOlkqlfRTP/VTQYg9TnPyhkZ7qRRWhhOHU6mUDh8+HGJWGFy8w7Vr13T06FFJwxKzXC4XW5tH7Itl89X6eE1iDBSD9nAPSu5wij77gmImn2d5TEn/QBFO8LiQOOxP5hU9DSUNvOPVq1fDOZoc0b6xsRH6ivL7srUoilQul2PxmMP1bCurd7zyDj1w7gE9fexpfe6ez+lq5WrMk66Pr+tffPhf6JNf/KSmM9Mh3vIKHFhNZIUxkRRKM3O572pkZFU6YcJ2SdLfk/RHGrzzr0n6K1KUlg4delkjI7+iP/7jv66VlROBF+j1erp69arm5uZ09OhRra+vx+pppWFYxfItDzF8Ttnu0h0CMNg3uqPQAkX0jebc6CVLBv3ycX/TFUJUZvi6Nnfn7jWAL34aGQE3SWEvIAc6QjZtbm4GC1QulzU3NxcGApaVo+BQVhQ2l8sFj+u5LRQhWUoHnPbSQE+FeNWH5/l8Unl+sgKIyeHZKJX33Qkjt+5O64MsvHwQYdze3tbMzEzYJBrFBPI5JGYsxsfHNT4+HgwacyYNa4NTqZRGdkf0rtfepZ/7/M/p3S++W6meiUgkvTb/mr5+x9dVGC2EI/W8wNsNl6eUIKoGx1xsJIRV6v8bSb8p6WVJL0n6f0j6Rwqbg5XLK3rHO/5nlcuvBgUrlUrK5/N67rnnQly5trYWK+L3WN1hpxtBFAXExO6IyIlvRYI+OInoq4qk4X7Gbqj3VL4/CVuLUnpqwGMLrFKSUseaMFF0Znd3N9Dk0vA4Qd8ziAHl2AAsMbGCL/khHmangn6/H3ZcI6GMMCL4SQYUpcJr82wgsUNOJ8UQRk86A5FROq+YccLJoTHf8/QSbcE48RmHBx06dEgzMzMaGxvT1taW6vW6crlcWF/LO/husVjUxES8Ysh3O6CNDtUqqxX9+Nd/XMWdoj7/4OcDYdRP9fXp+z6te6/cq9s7t0uKL8Vizh3eplIpVavVwCM8+mj8TM96/aQ2fvytmj30BeU/uaaoKqmlgYJGkv7q4N9SaVWPPPIpPfPML6tSORIY/ZdeekkXL17UHXfcoc3NTaVSKc3Pz9+wrMvnm7bx9ySn4PeRgvFVSAFxvFHSClLxOefnoPhyP8JIuolyevzlhb4Im6+sYL0l1sjhIErmHWdCU6lUOG0Y4QcSUAESRVHY/AnDwILWsbExTU1NKZPJhOIFPA4TQxuxfp4WctKAd+3u7qpUKgUI60UTnruFnPGJdwVHSIHVwD9gJ8/wEkJXeieB+v3BUfGdTifE/xzp5ykYz8fRHuIw95bpdDqcg0J7iEmpjmk0Gvr4tz6u12df17kj58L8NwtNPTf6nEovlW6Azv3+cEc/2sC78Ky9XryedHX1QV269LO6ctsnNPu3/r2Of+czyn6mKW1I+t8k/YxCFdHhwy8on/83ymT+oc6eXVK1WlWhUNCLL76oU6dOBf6iUChoeno6FuslIWbyd2f5kX1khqMY6IMX5XgKhyyCGz1HXUn9SpKJft3STghJuOewyzE3i1+l4T47YGoU0QPmTCYT24DXE/G8e2pqSkeOHNHRo0cDLGG1OcLk1pEAHCHhmRgJJ2689MrL/7yPTJRbVUcRQEKEHjTgVUbuRTzeA1Hwfy8J81hNUoCFrMhPpVJBWblok9c9E5oQDnhumXwzCMTXuuK109207nj1jhu2Pjl3/Jyi9LBwHGhHCOO8AgjGS/y4+n2p0SgMFjXszOq1u/+Cnvy//aKWfvshdd+dkX5Msbg3ivqamvp9RdG/1czMjIrFomZnZwMTjCH1w5fcc+1VRMLF+KCM5MOZOzYAQ2acEMVA5fP5kNLxXRf2uw767KaH5yatuncEqOrVNdxP6ZSXO7nwARPYaoKJBMY6e0ZnmdRGoxGWFeF5US5YXpLCbNyF908mkd0T+lpVCvA5et6T10lo417Xvaizulhcr0RJFiAkDRPL1FKpQQ1tNpvV3NycCoWBMK+vr8diGt/9UBoYSNIPzJPHphhMYjQ8NgaZBcwPfOcBfeOhb2izshnmf3lqWZlsRv1OHAYyt24E+Hu321U229LoaN3kKKPz59+iZrMWVhNt1R/R87m3auF/+ayOn/g9FbUYy/BEUU+l0q9pZ+chlctlzc/Ph3NvTp06pW53UKy+vLwc9rxlvpNXUqaZLww2Y0v+M7nBNQYa9n13dzfIPKTSQaTPm1ZO8DNUPNv8OZFQq9VCJ1GYra0tvfLKK5Kke++9V6lUKpTTOVnDTmsOJchVYs2xvNLAcm1tbQWFAC55oO+CPzIyEgqXWWqG0Hi+EiIK6OGrHfz50vDMGJ8090goohSPLRkjjAdejna7dXfh5nlbW1uxXPPKykqsgAOB8bFhsYAfVFyv18M7eDZjw9/w3Ch0JVW54SzQfDuvVJRSKjtcIshzHd52Oh0Vi8VwTEGlsqmZmcXYs9LpdEiNQMSk03ldvvrDWt+8Tw888Gsqla7EvpNKbSqT+V+UTn9SmUxGU1NTWl5e1srKiqamppRKpdRsNrW6uqpDhw7FoPdeF0aLeaEvyANj7IU4HjqhD56pwBntt/IEI7jfddOtMYFfFBMADaU48UOZHbnIa9eu6fr168pkMpqbm5M0LJan4Zxx6Du6u6VByPBolUpFly5dig1Ku90O9xNXerzl8TLvoP08h8FLp9MxxhkL6hbSY2h/j8M1j19c6ZxESypgkkQg/5bL5bS4OBBmlJOlS8BXngmcB1J3Oh2trq6GWN7JK5AFKzRoO94UI5dOp7U4s6hqqRqTjYdefUj9bl9d3dgf987MH/nQbPZGkSsWiyGGZryI9Wu1Y3rmmV/SW97y6yqXr4bvRJFUKn1O29uf0M5OJsjQa6+9pqmpqcDs4/0P2quHOcM4+zwih17vS/+QiyTyQWemp6fV6w03BfCLZxx03VQ5S6WSOp3BjmcIi9P80tCi4B2LxaLe9ra3qd1ua3JyMlY83ukMtwcBJrgXYTCIV1EkhHV6elpnz57VkSNHVCwWwzN8/x8sOc9zBpN2eH0ugkgqhlper5XEoyQFD+vnQgksZgwd/np76ZsvYQL2JqF/pVLR6OhoIHE8NgVyScPic9qQhNs7OzvB+CBEPIs+EIt3u10VCgW9evhVbedsT8u+VOgXYgLK2AJlPYdIOo3+xWVsmJpKel/GZ3PziK5cebfuvPNfx1jeTKamQuHL2t5+p/r9vkqlklZXV3X16tWwpWavN6g39tPckjK+3+VjR4jG5uIgIOTCU1SMp2/f8r28N/TvoA8RPIQRggVLyKpzhMvZK1/G5UQQGx8xWZOTk+H5Hi+yoBovQmdnZmbU6/X0jW98Q+9+97uD8kDiMKgovTQkadyoIAhO7BDso4wIGkLoMD/JtLox8DgrOZagBim+CwLWGqUHIm1sbKjVaunw4cMaHR3V4uJi6BvzAIJBoGFQ8YzMicfYkgJnwBh4jhSvuZHZ0Dcf/GaMlCm1Srrj4h0xYoz3+w4FkoKR47N8Pg7xarVD2tkph7Z4uDFEPzmtrLxfp079QdgIbNDGnnZ2rqharYb+tNttvf7664HoKpVKgb9gHA6S9734FSle8YMSMk9+MBWGEjlzQil5eYi413XT8zmbzWaAhqRKIDeAIZ6od6vpsZ1DJaArFjy5By2FzDCNeFyKFs6cOaOvf/3rgQBA6XylCwrqNDaDz7OAqR6zOrvn0A7ldC8/IDiGsYxbe/dIxHv8eJqJe10wEIYoigIRNDo6qkajoZWVlUBsUdPpkBUj47lG92wOvZL3+O+8/1unvqXViVWTKOn2a7drfHc8vJPnOTqIokFFkpcUFotFnTnzrPzczkbjkLa3S0qnh+kdiCGfs2bziNbWHtD8/NdiMrqw8IqeeOIxZbMjgW1eWlpSr9fTwsJCQHXIE8cuuoL4xfs8xPBKJ/rB/HqRghTfvO5m0DWKoqADe10HKidLtvA8eBiP3RyC+cRyr3sbaQi7+EwaJv3ZeiJZgUEny+Vy+P7s7KwWFxd1++23xwYzyRB6e1AWjysZUI8jvYIJxWQi+ZzJ8KIEh7XJPJ+TRA5Hk+WBrpzsIJjL5bSysqKlpSXVarUbvsOVTqdDFRAkD2PvVtpJH95FH7xPnXRHL8y/EFtWlu1k9QPP/YDSvXQwav79Xm+wQsPjXwxSLtdXuXw9Bk09LCB+Rt4Yr0F78qrVjmluTrHvT0xsaHd3W93u8MyblZUVXb16NbZrBrIwPj4ePF1ScXwOaUNyhYoTji47Pn4evx50+XP2/PygL5OKYEOs5F4xWCSvJEpCAvcQ0nCnM6/8kYZbosCsEptR3AD8qlQq4dwM4LHnW4eTGR84PCKQmQvBStaB+qp/j634mxsCFyZXYI8fPcWx1+XPYcXN5cuXVa/XlUqltLW1pdXV1eCdaafnMkdHR2MCxOfMCTk+3zHPPaYzrZK0ll3TucPD4gP1pfsv3a/TG6eD8LlxgKHFyIJmgIOFQk+zs2ftfqndHomVOxK+0CeUv9Pp6Pr1R9TrxYmdkZGGKpWVsFtGqVTS3Nyc2u128NqsNllbW9PGxobW19dj6yiTTLl0Ixno/IIXIiCj/v39fpLXfvEo103znB4Q8yCSzigTsBRP4LEXXgShI50C3ENggKYoOoXs/E4+bnp6WisrK5qYmNDMzEzsqL5SqXSDkUjWWXoCmknx+BFP4wLuntPRAf3y53neygkZV0y3rEw2KMVj+s3NzRCTb25uBg8EY8izUDjOh/Eco8dueA9XcFdGDz/a7bZevvNltQqt0O58J6+PvvJRFTKF2HcxbBgo4KUbtmw2q3K5plRqWDQhSZcvvzsWbjAPLjOM19raEW1uHtHU1Pnw/ZGRbU1Obmp1dS6goiNHjoSCmCiKgrK3221tbW2FFMfs7GwsB+lKCvdBiJAs3QO9JI3693rtRVJxHaicnjNj4nwVvC/BIX9JRb80XKgKM+o7otFh9y7SkLyhqsRhMTkztuVAsCEmGCCviJGGKwZoa1LZXNCJj5zZdYVOkk7+fPdmnmbhcjZPUliFz7gSw2OoKOhnXGiPQ0HSXJOTk6FyiDHxdzmU9Tpoxs5JsN3dXXX7XV0evxwjgj724sd0f+t+tXPDrSFRIsbS4y+8GYsi5uaeUi43PJVOitTpZGJC73DWlR/kdOHCozHllKROpxHjO/L5vA4fPqzLly+r0WhoZmYmljGgjnt5eVkzMzOxyiVnmZOcABDXoS2G8c1eB8Hfmxa+Ay8YIOAZlhM46nGir2ZB4GF3EYwklAJ7O9mBNfVJxkuSNyPI9yMC8b4uyD7hHlskY9ukQfJ0g9dO7nXRJ76XZHMxOp56ogrICYd0Oh085e7ublBiV37eR5zlVVTOztJvj4c9dcLn3v5CoaDtzrbOHR1C2lQvpYXdBWXSGXVT3WBIvMwRWUguZhim4HqxeLFen9Pq6qlYbpsMAAiM2JV37ewkDluR9Ja3vKTXX39LMH79fl+Tk5Pa3NzU0tKS5ufnA1vLmBKyra2tqVKpBLaY8fcxxFN6mg5OwpHi93Ilx36v60DlpBzMCwGwMI7JYddSqVQYWO7Dazrc5LMksUHH+RvC6xUoUnwbCiw1A0dbKYSXFAbbB4U2IPSew3SCCEPhkBBW0Y+NkOK5q6Tl5130yWESbfSwodVqaWxsLBByxNe7u7sBobBulXWKGATgMUaV6iHGgLnCazP+Dn+7O7YRdl86vHlYj1x/JHgORxLICHFes9kMbcVo5vMZzc9/JyGgOWUyI+p246tAeC5tdVJmY+N2bW+PqlAYbkBULA63wqFP3W43LMJeXFwMR73Td2f/m81meA/30G6fFwyQ1zMnw5VbvTyc2O+66WJrX8UgKcALt1LAAjrnFTl4XDydw6dcLhdqcL0yxZdv8V4XLMgkt3IOTfEOVIm4gXHv6F7OUwtJgsDJE2cfXejd6LgnlhRiHCfSkp6S7+RyueBxWODOWLB5GoYJ8oRxdPTiRgGlJYRwz+/9dVi/XlxXq/jGIoZOVj/xxZ9QoVtQT8P2k1Zot9taX18P8RyGlfNTW7stXapXlSsOK3wk6dq1x9TvpyXFobaPpc9Vv99Xq1VRt5uT7w6WyWRULpdCrM62N8ViUePj4+EcmUqlcoPR9Px6MExmTGkX78eoorwYbunWqn6Slyt68rrpkjEE2C0kJzOR4PWSJveCMGLsz4LHpUO+/pCOuxfDmpLrdDjnFTsIrENf7vN7nCyRhjEgsZN7ezdMCDhKhpDTPo8TmXApXvyQhJkYKL6P8kiDwn62dAGVJPtIKMHOgi5wPM/jMM8bJmNk/x6oZ3l8WfWxQR3u0ctHdXL1pNqleLkjFVpLS4OlWw5v+X10dFRbU1v63z/2j/STv1nToT8v6YjeGJ+ystmcut34zvZJg+weazC3cY+TTqdVKpWDbPqC6EOHDmlrayvsEt/r9QKj3ev1wul2rJrx8EZSbD4Jsag1D8jiT3AdpJwHgmWHpB63gN89fvSAnrglm82GVAjwt9VqxSr7PbiGVEqlhhU7eEgE0j2SNBQ2ZyHpsAuzExXOIroBcmbaqXOnzx0C0hZ+95U3xIpeL8pzWUTAvj7ScDcCls9x7ovDR5QONtQrUzxv6YgAq7+XwfK2M8aeWhoMlnTvK/cqnx6utJAUYPa1a9fC6W7ScBE8W5TOzMwolU3pzhcbOv3LPenvS3pDHl2mPK3g4Y//DPo0omvX7ozJKXEtCIxtWCknnZqa0ubmpra2toI8OnohznWvx3hDoFE/zrgwz8jTm7ncAe113XIqhcl2Asfdup9hKA0XOHOKmKSgrAhMspgYi+sVLp5T8kEgJcOKEphdPBDwAw+Lp0rCPYSC/xO3cq8LDe3G0zobSluZvGazGZSGZ6HwrkiQanyfogMKCRhTxhtDRLzJWLt3d/bbIa4bG4/vkwaq1+up0qio2CqqVWypWC2GtZEw9NVqVVtbW6Eoot/vB4NDm8fHxwenUu8uKuppcAzDP5A0J+m/GswjK19QVNrJ2CfnPJPJqtmcvEFOqXtdX18Pp11zjY+Pq9lsanFxMZyF4zElxpTF+oQvvBPlhUHPZDJhb6uB1y7F8uKuOwddjhL2um66ZMzjP0gQHgomR7hojG8l4mweSuINQ5G4D6aXAB2BcNbMIaKvNnCD4RPrzCseIkkyoZhewOxxhtdOopR7scHEPRA2oAnPofK9pPIjBGzahUL7RY6YeWAOvEbWoTsK6OPuBIsrKEUlqVRK8415lWtltUZaevKhJ3X8D45rrj4X1ip6TW+7PTjzZnJyMjDz3v/Bgvg3OrAr6W9JnbcVVM2cibXXDVgSqtN23yzOL+LA6elpra2txZAMqaYrV65ocXFRJ06cUK/XCwrmdeEoG/OL9+W+QdzbCu1iXF1Bv5frILb2phVCWGWHdk4iOJni0CnJBjqJIw1PduIZCJbHSk6GIKyeL8WD8c5kHo/vuld0OOLkTTLecS+DV8GKItDeZiaNmMSXl1EG6KkQvCNQvt/vx7wIyIEx5H6+78KLkmPEGAMviJCGLC33ept8fLvdrvLpvO45e4/Ul86fPq//9af+V3117qt6ZesVraysBLiOF5qeng7rb3O5wYFT+Xxe7XZbxV5RY9uWAmlJvX+VU6t5JPTD4z6UhXlwBd2LdOn1+uGU6lwuF/ZOnpsbGJN6vR5yrZcuXVKtVouhC4wNCs1eVx5WkddHFjCI1AITwiR5iIMu3rffdaByEv/5bgcoFQITtrMwxpLfaSxekc8cZrpCe6NdaDzWcyIoCdeSzCpxarIowFlBJ0Uk3VDh5LCetiQhLPCdNJKfWube2vPBDpUZE4/FnUWmTwgv91JO6Uy3FyDwrDDZhiZ8vPwejFi/19f7X3i/Tl86LUlaO7SmT/3kp/TZj39WzXQztHWwO/thFQqFsG7S+97r9VRpVnR8/URcuD4bKV0dEj5OwEnx5LwzzYyhX/1+P2wgBoE2OjqqSqWimZkZSYMwY2pqSul0WufPn1ev1wtwHHaZk+iI+92Iunwm5YocPj8uv/tdzBGhy17XTWEtJXUerOPiuafT6WhkZCRMsheOJ0+uhvgJQmBK5ouemTT/1z0cFgcWFm/iLGiyesjjK57JvXyG8rlxcJLIlZJJx1t6opsxwkO5tZWGFVa8o9frhVQEu4QTUiTRQSaTCbviowDc48jCDYF7eN6Hh3Z0wHhns1kd6h/Sf/GF/0JPzT2l1s7AM8y/Oq9sO6vCyKBGuVKpKJ1Oq1arhY3RHAqCgEaaw3pmSUpXmyrtvKx67i0xnsLnxC8fzxtL3vrBcw12W6ioUqkEUpGF3NPT01pYWNC1a9dCfTYGX1LYaK5UKqleryuKojBfbkydzJQUUASrVXylyn6wlXl702elADGYWBqFwHpCFwGlwc6q+rkdQE4E0wWn1+uFQ2FcQfx5Xs7HwHpsSLtQepL2SZKEgXZvTyUO27FwOWOanJRutxuKNRCuZF2pFN9bB49MWyDFKGyHaSQG9lgQ4wa5hqHwpH8yjmMMHLZGURSOUgDl4BHcyE3tTukHz/3gMB2mnlLFwTEGlUpF/X4/nPZMJZV7ZxT0/lcfVD/6mljgkq62lX38ono//EAwuPTVlypKwxpl0ktRlGQ4h4u4OfCq0WiEfW2Zj36/r8OHD2tzc1MXL14Me1RJw+1n2AYHI8g2L0mkB4ryyjZOqwuhgR3QjBw5qfonUk4ax0UeyCuAEPDt7W2Njo4Gq04Dk7ASRUFgnSzi5C/vgBeWe/yJh+AZDCbekgl12OhwEeHxgXL46t6bScESshImlUrFttpHuD0l47CUvyMQCGWhUNCVK1e0srKi2dnZMLY+DqlUKpyaRv/wsJ1OR9VqNeznBCPpENwZWhQGYaEKidU6XkNKv5MxFwoNNKO/LILgOVyX7pnVtWMlHbk43NyrHL0u9YengTN/zjUk4/5UqqFDh56IyWk2OzjV2yElVVbk4jGevV5Phw4d0rlz53TlyhWdOXMmpmhRFAWmvdvtamJiIob0MIKeUkHGmSfmh89RTJBVkqzb7zow5vSdD7BmDJJbYCAssM+9HsQQyg3b6MwqE5IsJPc8KhPkcQCCibI4fGWQuI/B3MtrOmSSFCygwxeUle0O+/1+jMFFOUEbeFM3ME6KeTtXV1e1uLgYTnVOMtP0o1Qqhc2m/HBXz1HW6/Xgdek3cdFeHAAFJQgLsSz/onxU/AD3YOtZc8pzMZj+nl6vp7HUnF5429FY+cDMl15S1N0K8+C5Ug+bmK+BjOwonfbi+cH909PTYRd8P4CoXq+Hs3VAZlE0OM1ubW1Ni4uLoY1+2lir1VK1WtXS0lLIcYIQfJmYE4dObNJ2Sjb511FTKpUKNdHfs3ICpyB+8vl8rDiYhgEvGWA+c7IIL4OFRHhRRk/XkET2v2P9PAmPUCIMvqqD73jlCe1ECT0mc8Fwj+FpIvc4yePd/F7aRjs8pUK8m0qlQnnh8vJy2NsHoZeGHmxQnjbcysNzcF7kwYTzLs/DkuKA9PA59NjYuQAv6OAZWHvmDkNGu3k/hNGQ8Zee+cij2rSN50uZZU2Mn4vNIWPpTCrKg3L2evGYlLEslUoaHx8PR3Kw0Rflhchzv98PpXwvvfSSVldXQ1sJbaSBsdra2tL6+nos60AcS+kkDsfjd0dQTuQ5Ispms7GDl5LXgbAWN4w1RGjd4zCgjqchJDzOYQK8mIA4AwUk1oOeZkCc8KBzfiIY8NXb5jEjMMM9K4LuVoxn4+2BQrCwPB+aPZkG8DjPoa4jCvfc9Xo9nH3Cc2iPx9esusES026U1wkuj408v8qYIfy8B8jmOWgMaxASS214nC8pFJSn0+lwArXPM8Yxm81qOvcWvXrP7+nRr72xD9Dr0thLL2ll4cHQNx9TYCpwd9Cftm7Y4fqNyw0OedfBqWYbqtfr2tjYCIxtJpPRwsKCXn/9db388su67777grwyH2z21mw2w/YmbnwZE1ha9AOFZa49j5okWQ/Kjd7ScQzOwkrxrTkQYgJnPvPJ9PjLLTsTjDdwD9tqtWJnP7rlYRBceGivTy73eH7Q4ThC7Z6WZ7mB8OdgJakWQRjoHwrla/z4G/d3OoPdDBcXF0POrdvtBoXHQ0vDbSMxhngk91QU4QObHdYzD5728vJGNzIeDyUJtyTrDJSFVPHQJYqiAC2B36Ojo+qM5HXxzFhQzqgtTY2+ppdthY6jE9rlSCKX27sWFYfAHIyNjYWUH+HP1tZWDPaOj4/ryJEjunjxolZXV3X69OkYwZZKpULtcrPZ1NjYWDimgzFE0XEgTkAxH+iA52oJIeAJ9rpuuvuekzfOAqKwvtLBvSgWAnrZPRsTLCkQTF61grI7M8xEScNKETwtAo0gJePevYJub4sLMkrJe/luvV4PcIb7vT14bCbD11963NrtdsOWI41GI5AX7XZb5XI5xCYYqeTWorwjmx0eoEOpn8fytIP+k+phrpJMNULjf0PgnZVHEVqtVozVxRtwH/0F7hNbnVy5V31dDWu4J8vnNFJYUb0+FRtT5pV+DmXrxkhsbW06hFtOKDnhxth7mi+Xy+nIkSOq1+uxUkRkcHt7O0BmjAYF+EOYvRsbb+8344rB92wChNra2tq++nfTVaJMsKdP/EJxiH34P1YEggQhZtCH7FsqUNUe/zB4DFa32w1MrnTjCWge83rM6ELmK0S4P8mKUi1CXDk6OqqRkZEYxKff3pfkmDnRlU6ng3e5fv261tbWAuPHJHc6He2O7UoFhYL4w4cPB8GgfYxrMueHwHlFC3ER3sLhqAuSG78kNCeWd/axVqsFmIxhxaAwpvzNoXev19Pv/Jm+2obkcr/TVC7XveFcTd5FSEQb94K0Fy/eEZt3vst76bt7MPrU7/e1sLAgSbpy5UpYPYUiw/ru7OwEsm1nZyecVoBSO59Bm5OVTsmF2e12W8vLy/vq3oHK6XEQcMkrdlAmD+RZu4lVpYFOyACPgBfuvTzwxoMSZ7kSeaojSR44XPU4M9mv5OcIAywazOj29nYI/t1DumIiFE6z4zXS6bTq9bquXbum9fX1GARrjjfV6Q480hMffUKrx1aVyWTCHkkwrcRwnq7x9FBy0bXHpihRMo3keWvP32E0MDJejbS5uRlYe+be011eJ51k9Lvdro7V71GqPxS71JWuDk+8HDb0Jp73TaDxnHsZwk4nq35/PIQKS0tLITXECdjMr88NyA5vxs4J169fD89mjmCqW62WlpaWYsqKAXAmnvF1WeP3ZGHCQeV7t7yHEMLm8CBZjePr+OgY0MuVwGFHt9sNMRUGALqbTiGMrhz8HY/oazV9gTX94F8XXv+MSXCG2pXYlRnPxIDTD/rg49DtdrW+vq719fUwYQjF9si2vvaTX9Njf+8x5XI53XHuDh1aORSLI93Def+ddGKFBVAJRhYl4zAnZxAdpnI56ebMI/OAkCZJKoxGMn1C+0ulUpiz+ea8IqU0OGNeip7qq7S2pCgTN5a9Xi/U6uLNvACE6+rVO7W0dJtSqfhRjBQTsBuEzxP9dCg8OTmp7e1tLS0tqVQqhY3V+J6/FyTE3/CwyARI0nXHQwzmsFgsan5+fl/9u6lyJgUUS4rXg8lkMpgkL2zncwTTg2gsGpYYpWEQ/Zk8g076v1h899geL3sy259Bv2BliYExTPSPdyBAHl8yeRBGDpFXVla0vr4e3gfM6/f7SuVTijKRRkdGlR/La2F1Qe1MW7v53dhO6QgDCuvWv1gsBiF2KI+B5H6Ex8k9xoviA+5Hcd2gwmJKgxUYzK3Puwsg8ZgvpSL2dPXq9SMtNo+qMxI/QoPvMJfA1KRyXr9+h/r94VmoFGAAZ1lWODY2FiO9INRAMYVCQSdOnAjecXJyMlZc4EaIHR6cCWfJmZdwuiFwZ8M4pdPp2DatN+jfvp9oeOY9AwOkhbnzwmtJQbidtPA0gisEz0fxgE40mvsRdiAaSuPKjZAQ37l3ThJJDsf53jCeufEkY+5hpQJCj5Fy6EiMDUGztLSkjY2N0DagFGNU3CnqBz/9gxrJj6hcLodtSIjzOOEM1tFjZGI+n3AUhTlBYOECHLI6E4lC8lz3GMwZ3yPnmgwJPCXAM5MIKZ1OK1fJSZbr7Pb7er62FAyKG0zvF2PsHsvbBmT13Kwb0STsdKNMfrPb7ero0aPqdDq6du1azCl4CpCiBOQQJ9JoNMK+VV45tq/yGQLb8/ODvkzjXajplCulew7vvHs6FywGkOdj+Yllnd10QsKZYgwGCsRk+MRxP59BiKC83O/vLpfLgSGkne5lnKTy9IVD/rW1NS0vLwdGk3udTMrlchrNjOpo72iMZeZdxDTAbGcaeSfoIDm29AeYzhab7HSPkPMcjIYzt06cAGel4bpFBNLzxy7wTg7GYr3DHenhoYxlOtIHPz2sDJOGhhYD44rOPS6jzhUkER7jQv88nSEpGFSqncbGxnTo0CFdv349nO7mzok2caask1U4LE8B3ky/kv3x66arUhBkJ0L4mzT0ll4x49S/w04sb7KgIYqisHJAGu641u0OS8fcAwKbEX6fNH8mFpSB8nu8GAAhc4/K+yj4hlUmhkmycQjG6upqOHaPQgfPR3JvPp8PJzOvra3FxtlDAFIB9Mcp+mRu1dGD/7hRcghPWzz3CVyl/V5yhlI7EdRut0PsiSKCDpKH+KRSKXW6U2o0F1TWxYGASoqimlKpvqShJ242mzFYiPca/Dce//ncOrpiLL0tyI7LBOMIdJ+enla1WtWlS5fC8RbIMOPW7/dDeg0ngsHsdgc7WrDnsCOBvXRsv+umntNhHH/DMjI4XgWDJeUYc/dKwDP3tDSQgQSauDKivD4BDld4r7O0Dv88TvNqDX7cIzGYTNTq6mpY5YDisll2kgC4evWqVldXYyQZ/XSFKpVK4UBXrLI0LCd0D0SbeZ9basYWZca7usGhfWHCU/H9kvwHNMTnxKi1Wi0gHL7v6If2JImmJEMvSZ3OmLa3D8XkbPalp9TeWAv3Mu/0lxBpMJfxCqFicV25XCYYjWSKiawAsR1GmNSNk5U+Z7fddpvy+bzOnTsXZNcJNeSGONMryKTBJm31en1PEosL2dzvuukeQtJwO0w664XmeBX+RSkZCDxG0qrzLIcZBP58TrEy6/I8HeCQ1b2CFIcL7jWIERAwF24EiWQ+MQRnbLRaLVUqldhx90BI9kat1+uxlQe+XQtQlpwpzCvsJ0JEIYLHkHjMg/pMv114uPAQtHcvUoz8qRtdjLAjHoQbwfJ4nXnE46Govr/vnoJ6pautlU1lU/mwsglvDMIa5o2HxzdK0sLCt5XP/5jy+ZEbOIhk+gkjTu02ns2NGOOWy+V08uRJvf7667p27Zpuu+22gCYwhizb293d1eTkZDgMGqMNCZeMk33sD4K+Byon+BoFSqVSgQRyKIjSwD4ly+ncm/G7NISvvozMY1wvi8JyedxJHhTB5ZkeZ3qKBPjniWr6yM7plCFiWFjNkclkwsZR1FwSF7FShXpKJoX393o9lcvlEK/5RDEeGC3Y11arFWOoUSpPb7hyeoyWRA6Ofvi+J/hRJASFsarX62q1WsFo8JkLv3t6D1X8sKSkMievvqR2p6Pa+ro2NjbCJlsYJ188HkW7Mt1UqbSie+75tM6e/YsxUgkDB1rjs83NTfX7fdVqNW1ubsZKMovFYpAnVgG124ONsq9fvx52VUDOuA/lnpiYCOV4oCvG3eVSiu+HvN91oHJiPcmTcaQak7G7uxsTVOh+Xu5Qkk74ZPIOh6YIEJUoeDJiPT735WVe+ucDwX1euM897PuCwvV6g4Nt7r77bt19993hnUwY70+n0+GQHGlYOVUoFGKsJAKZy+U0Pj4exg54mJws3zjNGVRniumvGycMHOsM8bIOL90gSfEzYlyRpeGWIVtbW9ra2lK/39fU1FQg7XyckwrppA394/98N5vZUSa9eYOsde25HIZL2/CIg7HdlcPaVKqvM2e+rGvXPqadnflYCAZ5h2Fn3Dc2NkJukgIPjGC5XI7VGy8sLKjb7Wp5eVmpVEozMzPB+fC9TmewnnZ6elqjo6Mhz86PE1V+Ocex13WgcvqkZ7PZ2IleHDTkBAmKyqR7/g0P02gMdupGaarVarCOTj64QAHTgMh4Wo+7yKHh/ZyVBColq3zYy5T9V5977jl95jOf0eLiot73vvcFuELsjLVkQB0NINg+EYVCQRMTE4HQoT8OUfGw9NG9OWQDIQNxjaeLPJ1Bv3iHFC/UcGaa+2hHMk/JtiNjY2MaGRmJGU1kg3FxmO9z5AIZ0iHXVjT+oh0rqAG9k06n1UlwHE7WDJUto62twyqVtsL38/mGstlt7e7Gc/JJaO/LDCkKwZnU6/VgiFFiyK2FhQU1Gg1dvHhR6XRa09PTwfCQ82w2m2HrE/eIyTGgLUEB36xyOrXPxOCpyO+xAh8BYkEqcMLzo0lPirKgUKlUKqwmwGvm83mNj4+HGlOgBB1H2RhQBjxZxwhDCrPGJk5YT1hHj48QkJmZmVjKhNgUQ4KHQ7ClQX3s1NRUbBz6/X5ogys5xdQIunsk4lO8O/2mIAGvh6Wnf/yO8rmQI8B8z2M82jAQ+uHp4658/l3vN0rl6Im2hAqisY76RyUNdUvVO+7Q+OHDahvZhAHxHfBow8bGKS0svBSTVWedk+w0bXE5yWQy4dgGjr9wPgQSCFk4deqUXn75ZV25ciUoMB6WopxOp6ONjQ2NjY0FgycN9zCmHbzfkdNe14HK2Wg0NDo6GmAgBcRbW1t69dVXtbKyonQ6rYmJiRsWuY6Pj4fGU/PpAikN99rx2JA60M3NzfB8BhtPgofxxDjfR4g9Lwhpw5mX9Xpd1Wo1wEXKB48dO6a3ve1tmp+fj6Vp3KgwoMTetB1lSqUG++tMTEzEUg4ILc/0IwE8ToKlRWk8tvFKHIwhyuQkDN6R76DQPBsj54LtXn9rayswkMTRnrP28MSVkXHiOwhkpzPcDziqpKUjkfTdoZxtHzumwvi4Chqm3BhLCDiM0KA/GfX7wxOuo6in+fmndfbsqVj8ztw7ynDegyKMer2uXC6nUqmkbrcbZCjJe5w8eVLnzp3ThQsXdObMmbA3kxsu8tO0u9/vx05bc86Fd70p5URgPH/Z6XT00ksv6cKFC6pUKrr33nt11113hbjEaXhiMCaLOBGG1AeMzqFwCCRr3iBisI6rq6ux7fWdHUUxiWWdpQT6LiwsaHJyMhyg6pCNmNKFFuHPZrNhdYKzp1hEFNMZajy8NIyxfZJckDAorCqh3YVCIRxa7Hk3Z3S9ID6ZnklCWVc297qsJGH3dldajGKz2dT4+HiYKzcQyaowh9qpVEqt1pw2Nu7UtJ4P71+KLkjZvlLdocHAMDCfzOlgJcc7ddddf6DBwuuBkubzKzFDjRzybuYXTycNT9FDSbLZ7A1n0xAidTodVSoVHT58WK+//rouXLigO++8MxaCocyNRiPIvaSATJKI4k+knJ7k7na7Gh0djZEC7XZbV65c0fXr15VOpzU2NhbWWHoViifJgcQopQsJz0bIvHSQQWMzrG63GzYOxksBS/kpl8shJuNz+uMwm/jX2wLrzPeBv1EUhcJyxiCKIpVKJZVKpaBQ3CsNdzHE6LjHo988BwFD0J1kyeXihwK7cjjb7bDVDQQKCAcAioEx7/WGRfK+wsXbCtcwZE+HG4gjkCiyM5Ie93a78cNm7/rKWX3zJ7a0XRwJCuGemf4jA7u7x7S1dVyTk6/HnsP4OivtLDQG3FODeDSMisfKGGU85Pb2tubn59VoNPTaa69pbGxMp0+fjhlo9AYUNzIyEvZ+wsA4oQQH8z0rp0MylCmXy+nBBx/UkSNHghVnzaCffeJMo1vWbne4z6xDXToI3GNi+J0SsOvXrwcPUqlUdNddd8X2I6IoGyXwpVO0A6WH6STw53PIJCYaY8QODfSDAXblwULTHmftEHCHms6s0l+MohtHjx2dcPAKJNqDoDgqoS94fAQJr4nHBup7ez0G9vnjwqh4ntHTM26QcrmcUmdS0lMmaO1dXdu5pIncmdDmpMDTxwFMzGh5+W5NTLweoK3UUSo1RA0omffRDQ1/Q05xKBCcjDXeDh6l3W7r9OnTqtVqOn/+vMbHxzU3NxdL4zAeyB198lQg97pD+J6UM8k6uaIeOnQoTAYVLyR1feK5nxiJTaYddkDWeHULSoF1ZjfvCxcuhATx2972thi2p408340DgotX4x6H2LTZL57Le3wBsadL3DpCUuFJUFqUwuNK2kSynzF3gcIAuCBhAJx9TgodRsBJDk/TOMuL4krDM25QLEnBqzpUZfz43SE733NDwPc23n5K07/9ZCjCK1abOvOdc9p8312xseEZGBSXp8uXP6ITJ76iQmFTknTo0LM6d66l3d1h3a97bCfhMEQ+rryPMlIIHTcWPhYPPPCAnnnmGb300ksqFouanp6OpU1SqeHCAncGhBmMh+/VlLxuelZKr9cLAsYEE3dgzegAjfAcHNbYCRMa5vEKcR5KiMKgDAg6QfXU1FQoycLKuyCgVEwOP1DfxNMenzCBeKNmsxl2X+h2B3W+tMUnj310UFSMgi8f4nJaHU9EHOSC4qjFvRZ928vyukJ56smZaze29B1l5yBjlID3MobJEkm8DV7ZU03+HvrCe1Z2TkuFoZxlutL54vNKZYdtp29ewcWzB8J9Qjs7U+EZ+fyGxsaeVzqdDjlhxsQ5DIfbtA+lazQaqlarMWNPHEmfnaS75557JEnPPfdc2OLEw6iRkcFqI0I0rxFnXA5KpdzSqhQnWsh7orxgcvckTpYAzVA2LAk/wE72T/UKIY9ToNShqZvNZhBeBIg2eXzrk4BwIXQoJzEqSoGCMtHEWJ6LRaEpySMGo8+kidyoJauZUCyHOwNByweDg3HCQEjx/Zw87uRyxXTLjFC6gHlchUD5M7ifsSQedAVlHpIpDBSCdzDm9TtuV+tYvL72kc9dUrY7PEIC49XrDQvKMTQDhjSr69ffbn3uaGzsCVWrm+FAKd8y1OU4iiJtbW1pc3MzhCm9Xi/UUm9sbGh9fV2bm5ux7/i8SwNe4o477lCtVtOzzz4bdkZgLOg728Ww95I7JZ+370k5uZyA6Pf7YRU87C1VRL6ZsUNghzhMPBYDrwrDy4Ui+Ya8/N1TDa7AwBngnBd/Q7j4xmG0iX4h/EwA7+p2u2F1ikObXq8X8pAIJrGn50v519uTVCafVLwRpIIzge5VnEF2WEtbnAzxAmtHGe4l/fgACDFXTK9s8lABdAEiSQoeis6YNdoFrf3A3TEZO/30VY2/8moYJ8bNZQcFR142Nh5UpzP0knNzL6rVqmllZUUbGxtB0TDkKCiGi7HBoznsJG3XaDRiSA7uAyM7MzOj06dP6/Lly3ryySfDUrLk/MLLMFfJvOpe14ExZ7PZDILDRlfUmOJ1PMbhwso4hKMxCCdMZBRFocAAi8kEocQwlOl0Opxk5fGap01QaGfGeDYC7ysvHGo7LHUoCaSF9k6lUoE0KpVKMfqeHe6cjXa20wkfBM49M30rFAra2tpSo9GI5RxBJgiTQ09/JhPPu0AUGCAu8rUk1p10chjtjLMrHoYT5OHpCydhXEmjKK3Vhx7RQukrStXfqNvudDT5ta9p4+67Q39cRlzQkalq9aja7YIyme03PttVLtfSzk4hjLfHyV4Awo4JPiccfkT/GTcMu0NtntNut3XixAmtrq7qtdde0+zsrG677bZg2KkxduLTQ7skx3HLyum70MHiecWQQxcGlA44nKOB7vGciHC4lYTI5Jh8JQyTxH0+cB5PeiUTu9V78p72UY6IIO9lXEAFzgJ6xVE6PSycxvL7c5wMcgKKyqRCoRAMFySPn5zljJ8ztii17wTgpJIjA0czfOZzTP99hwX2p3XI7sQViupQ1N/vHh7hzGQyWp18ULsnx1R4fmMgqJImv/MdXWg2lXlDzhgnN0KOHJLX6OiqTpxY05Ur94aUD7IKIcMWne6heZYXiSTDBj5n/vkeY3v//fdre3tbL7/8subm5lQqlSQNa71dBhkzQp03pZwUDgNzsBR0LJVKhWVUeDLftpFOIjQMtOfrEHSHRHQqSXp4p8rlcoB8LtB4TLaL6PV6YVtKvpskNZKlhW5RgexOJGUymbB6BIiE0iSZUGlYLODQDq/uzB2GBMUvl8shzmeFCO/3MMCJOGA1fXQYTf89Dmu328EzM47OpmNo6U+SNfY5Ybz9HX6Eh79/J1/SyoNv1dHn/2OQt9LFC+rVF6XCyT0VkD7wWRTFi8mjSMrncyHnTB+JK/v9fjCCydAKxMR7nNRiLJI1t8n024MPPqinn35azz//vN7//veHdvp4IF/u2Pa7Dow52daCGBNWkvQGREqhUAiw1xXambq9mD73mAwk5IRDU18+xZHmCJTHOHzOFhlu3WMlZNGwnIvPpaE3RuDxvBQdOHvqcTfjkPSODjX5u8eGTDxK4eWGQEkv7Gi32zFWu9FohEN7/e/UDTvR4fPhqKXXG+4S4DAL0i2ZUnHIms1mw+ocanDxNhB9DgGZ+8HYRLr+nkfVT1s41Gkr8/xXbjAEjo5AOgMYvXe85vE0bae/GDZ4AUgmlyU3OihtrVYLJZ/IqldddbuDE8nuuusuVatVvfjiizE06MsQfaeQg9jaAz0n8RSNxjOS82PCfNC8KN3jDbyQDxyLUJ0y5x5ybZ7zAvuTeiBhDFlEPIawuBekH57/Qjjx5A6TJcW8BCQXRA+fOcx02O6ki8NboBDPY2yiKAoQEwhK24gFfXz8nYyzM5TO1CahLWw0nALwzwWUcfWSM/d+KLGfp8pYuxKyeMFrUEEp9cKcenMZpa++sfyvJ5Wf/YZW3v+T6vXitbHODzBPqVRBjcacCoWNILOuyF6G5/G63+vz5CgAY4ZMY/jcUEGWoWi9Xk/Hjh1Tq9XS8vKyrly5ohMnTtzgkNwwJzmAW1ZOXDmJfCY2SdN7VQqdpIoHmOVsW5IwcAEivnM45iRAEqYxKF4ID3WdSqViq0ckhWIHj41QPgbN1+O1Wq3QBqAMBgqBc4FwIZKGLKlDzKSSomgcQYdyoxjJPKV7YLwUtbjMladHEGhinG53WNxdKpUCmeawHGbaF4f7v/THS/1cuL2dDiPd6FbLs9q443ZNXx1Wwd/zVE3faPXUzA4rp1B6ZAfHkE4XtLFxh6amhitUnKhy6JiMh/1Z/l33hMxnNpsNCznQB/russkYHj9+XOfOndPZs2c1OTmpSqUS3uE65PK813XTPCdwle1HnKr3PJlbG2InlBJlcMF0Cy0NWTjIH56ZXFGPd2DxMrk57kFxfHsIJ3b2Su/4ADt7hhAn245iJifGc44ospMG3OvwCYofss2XhknDAnD3kp6od7gHmcPWlSgn48H4MobMJW1Itt3JCjeiLlh+1IEXaDC3yATjiMIM2hKp8afnYjK326tqZfuFmII7IUd4kVQsLucxkCtILNo1hMXxd+BIgJp4fk+nLS8v6/Lly7p69Wowihg9R0iHDx9Wt9vV2bNn1ev1gudmfD3u3++66abSDtFosJeNuVX3ogCP31w53VNJwwQ8KQ4sfzKnx/8JysfGxgIUlYZWnaoMhxHOFHphghTf8Nq9E94Dj0xf8VAYJs+xJj2M95O2uMd3koq40gXaiZYkE+sQ19+JIjui6PV6ofLK00P03fcN6vWGNaHuYTy/iYBJCtU4TuAhpMBrxgKv4R7q8uZjOpL6vNJv6Fl5o627nljV0seGRInnoZOw0OeSe30OeQ8GgcJ+N0zIH1AY2E+fMPI7O4Mj7be2tgIiO3bsWEAO0jA/z9GIy8vLeu2113TffffFQkQfp/2um27w5Q9y6OLVLm65k8l9z40xaB7/ZTKZsNjZlYN/mUiE/9SpU7E1m0Am33gLZcHLOTnCxWTxjKRlZaCdsOp2Bytz3Lp6HOTLxJIkiqefJMWE32EfS9VcyBzKMaEYPB8j4mGvskoaPJZFJQ0IgopnIg/onp77eR7z4gwkY4wielUW32NuUqmUXjg1rtNzJc1dGxxHH0lajF5SLvun1WnHj6NwlpN/k0n8ZFUYY81c1Wq1MDZ4RQg15KBUKmlsbCy8E0PW7Q4OXJqdnQ0kIduaViqVMB+MwczMjJrNpi5cuKByuayTJ08GnXCEtt910zvwEMkAmdUoHj9h4TyGcYvnUJa4FUjly7kQAq/TxerPzs7qyJEjMTiMYrD+070Vnh3LTTv2Sm+4V0M5nd301AeTTV95vqMM2uHGjf+7gDmdzg5uTgwliSo3AN427zc/xLHAfQ8TmEva4ijJlQKInIz5gHP8JOebuJf2Jr397u6u1BzVlz4QP3r90T+6oEwUPz09nU6HvDLz4EXkXI6kvC+0fXd3VxsbG1peXg5bV6ZSg6KP8fFxjY2NxfZLarVa2tjYCATk6OioJicndezYMVUqFfX7g4X8VAaBtDCghw8fVq/X09NPP60LFy7E2oXxf1PKyUMcoyNMXhXjFtZha/IeYhkEjuoSBARlcEII6+jC58pAXMzz8Ra0xb2N79bubYLcQNF8MrmXXC/beXhJFxPhSu4pIfcqGC+MicP/KBquFcXLOEREIZKK739HObvdYaG+l93RpiiKwqoJR0WkRrg36X083kNJ/Bl4pSQj7t9HgLe3t5UrFNQ9cpt6JqOpbnwxgkNBDABjmPScpEjoD+313Ro4MQzDQd56YmJCc3Nz4bh6CL9arabFxcVADvqRCxCP1IoT8nl9wMTEhB5//HH93b/7d7WxsRGbO9DOXteBsNaZK1+j6L9DZAABeKl7UmCaQxMmCsiQZGfxrHhXD/KdFIE1Jv/U7/eDgLlXRelRNC8Nc3YVi+YeC5RA3rfdbocNoWhXcqsOf+5ecZLH8wh2LpdTtVoN1Urc7zlCf5az0O49W62W6vW68vl8WIjOO3zLDOaA6h+vguK90jAUkRTqVPv9fuh/0vgkobt7UzwSocmhiUO6eu/dqpe+obHacH77vZ6Uiu/mjyw5GeZyIekN4znMGXucSr/oG0gCIgjyk3FHGbvdbtgIjh3gfc6YY04E4HOcyqlTp7S2tqYvfOELuvPOO/UzP/MzIZYlP7zXdaDnZCA8RqCTCIILjwe4/J2Yh8lLMpWuFEkW0uGsDwIejI7htZxl882a3bITY7gHYPvNJPx2ZWKvITcAeBaMD/ei3G71PeZ1uOVIZGxsLBgqV2YnhnzsGG9/Hl6B7TUQEGm4aCCTGWxbsrW1pXq9rs3NTW1ubu657Ck5txhqvL17Lie9kqjF+8/GasViUVEq0ns+31HZTl+PMvGCDR9bZ4MHxjZOqDBHHgZIw+L2QqGgsbExzc7Oanx8PHhZFNerrfC44+PjmpiYUL/fDwflIke8C+8LmiDt0u/3NT09rZ//+Z/XwsKC/tE/+keBwfW5+Z6VE4XynJVbMhc+hyEeX3hM6vf6KnEP4B2iQeezlw2QIbnBtENTJylQdibX20b/kkKOpYfBdKvrkM1jYi4nl0AFxC3EFw7vk1afYnqvpkqWkfkYudLjjZrNprLZrCYnJ2MLxFmriFL5NiXtdjssgmc1hns/N2oO+b3oQhrmYx0Ku3GijWzfUSwW1eu21Vj+49ixgLWPnlKUiae63JtDwA1+bmQ7HR0xbzzL4+nkj4clkJTZbFajo6OampoKY+nGEseA08AbwuiOjY2p1+vpoYce0i/90i9peXlZf+/v/T3V6wMC7CBYe6ByegrCrRgCgTUC0uElmRivuuG70jAhT/xBbCQp5BARSn7IpzExPIuYGIFwOEhsk4yVvWCBgfZKG4q9aTseKAkh+T/CieC4MaO/xNSOFmg/xgGD4ispeLbDRTd0/N9huJcT8vdarRbb3RD4xTvhFfzZxM1svs18O+nnhRWOfBwJIcScOI3B7fV62u7Vdee3N4fjIWkrukftBFPryuUrRXorPc7hjd3DmDMfxKr+uxsRN4QuX16wwjgx3j6PhG3E95/73Of0X//X/7XW1tbCaq4f+ZEf0U/8xE/o937v9/Q7v/M76nQ6b/58ThTKLc4NDzDL6UwlDWaSGQQ/FJbBwiMCCfihdM2hE8YAyOzPIHbCmnlMxiQRwDs0c2UipsL6YrXZoY+J8fHwfjJJPh7AXjcoDgH5wfOwGj8Zp7pCYRyAUNRAcw8/JM85LoJzQOmPpPAMH6fd3d1QbO/7FCUJOtrpaAqlcIQiDTwraxoZi5HXXlJxfSXMb7tc1urdd8eKFrgc6Qza3NH0H3xHujKUR9/5nnbTLlfOJCvtIYdXxGEMOGaBvsNOU3NeLpe1tLSklZUVjYyM6Id/+If12muv6W/+zb8ZSKBisai/8Tf+hu688079/b//9/Xiiy+++ZjThRAvh/VzYsOtmaRQt8kEOLGTTg/PA0FIsOZMqBcMMHjARWdAPZ8HFE1CRoc4PBMBd+Xg/8nd7RBeZyKTW3MkUwW+/g9l9JjdYZR7X2JbvDpGyOGnK7OXj3ms7mkOSCwW+yJUzqjjSWkPfeC50nDjNebTFdLZen9/Ei3QXjfitz19TiPNocy9cn9JW1PlG9ISnlYLMHVzV/l/vi79rwonNDjMB8l5GanDbo/B3Ysiz44cOp1OYHDdcRSLxRCTfvGLX9R/+V/+l7p69aqOHz+uT37yk/r93/99/eqv/qrW1ganqB07dky//uu/rl6vp69//es3sM23rJyO1d31IzTUaUqKERAICl7OYW3yme6NnJWFGIEt5fJBlBTyb3wHQXdvnsxDujXGIDCJ7pUkhVPBeFYStroBcdgpDSERwu7e2O93JS6VSqE9vNPjZ4+N3Dvxfe7FS6CkvprI4TIGwZXH58PDAs8t0z+HmD4ujA1svBdtIENFrevM2a8M2x9FSt3zIWXypcCmexkk7+A5O60d9auS/pmkqwpzT2gEw07bPU505OQpP+bC39vv94OhozKOPoyMjKjZbKpYLOrHfuzHdPHiRf3yL/+y1tfX9YlPfEL33nuvPvWpT+mXfumXdP36dUVRpLe//e36hV/4BX3ta18LY7rXdVNCKFme5/WTCDKWza0Nk43X42+eS+Ti+x4/Mlgu7Ey8l10hwC7oDiPxqAi7Eyk+6Vhaz1kSQ9B3BNvpeUcXvpGZ/46iOHO5XzxFqRj7mTIOjIXHQ45gUNxkmoHvuff2d3sI4YpJDrrf7wev6UbEjRpbuCRZWvJ8hAUe76ZSKVW+dVbF7w5XlHRzOS0/8g6lohtjckcJjNXk1HeVy9aka5K+JfX7kaQ4kqCPSePhYwRhh1ISGtAfX+Pa7XbDEYLI5m/8xm/oiSee0NGjR/XTP/3T+sxnPqNf/MVfVL/f10//9E/r/e9/v+bm5vTJT35Sjz/+uLrdrj7wgQ9odXVVly9ffnPKySAQ5CIcfMbudFghWEGS9HQacgDLzWAxUXhhFyIUhGciBNvb29ra2gqVNN7OJJkhDb0oBsahGRDLN4CmWglvQy7PJ5XneSWJ5/rcUieVGUPgBJuzhNIAQlIokEzHeByKZ0yWrLk3pU3c4+92RpOxcPjIj/fNv++5RDdu9CeZrybdkM1mlct0dOern1OKOZR07f77tVYqxc6vcaQU5yTSKl64rlSzMyCE/pO0vT2vev3eGHJLknb8DSO1vb2ter0eUm+uyPzu8WgqlQrHeZCCe+9736v//r//7/XSSy/pp37qp3T77bfr61//uv7W3/pbuueee9RoNPRzP/dz+vN//s/r7/ydv6Nf/dVf1fb2tmZnZ/Xss8/uq3+3fJCReyyS5E6o8JlDXw/gk97DJ9crZJhgBpVa0+R9tM8hHJ8BpZgUj1H9u5lMJnhUSWGCQAs8g3Y7nMTguNd16IbQIthuPJL/dxQBcYA38vWxDhmdQQQComC0ge+xTy/9cgbdmXFpuKM7yMKZV8bIPZKnfHq9wQFR7XY7RnTwfVBROp1W8dKaJr92cShruZxe/uhH1ZbUe+MdPJvLUYL6Xc29+CWp/saHkdTvZ9Xr5UQA6vPkni+bzQaFxKP3er3YgUPScFcIZMpTcdVqVeVyWVtbW/rABz6gy5cv6xd+4Rf0D//hP9Sf+3N/TouLi6pUKvqVX/kVLS0taXNzU3/qT/0pPfTQQ/qt3/ot/dzP/ZzOnj2rU6dO7at/N1VOICFsoTOQWFC8kMMOhN+V1T0EVgnlZyAcrgJJ2XKCd1GV5FCLz9ySA6khBIDgyYmHuGi32zHiA0iGQjA5Thpg0TEebpVJPTCWxCt+OcHjRBAenTge4cIY8ex0enCMPQIkxQ9mBWGUy+UbIO0NCvPGu3g38wnk4/t42r3WenqVkX+GcEdRpE63o+jFP1SuOmSCthYWtHHmTIyTcPTB+xmH9WhV9c9sqCJJaan/CWl19X1qt6V+fzc2zv4998ROEOEYONTIeRZfOrexsRHuRS82Nzf1F//iX9T29rZ+4Rd+QT/90z+tL3/5y/qn//Sf6r3vfa9ef/11HTlyRFEU6dixY/rlX/5lvfbaa/ryl7+s2267bV/9u6lyIjguSA4BmSw6wcQD9ZJ5JrfC/J+6Vl87hzfyw2CcEGAvGBTO2+gMrrO/xKK8j7ak0+nYhsrEmmwu5vWjTiI4OeZFBoydK5OPVVKBEF6gLBCw2WxqamoqPMP76nGYx5f0wWN9vKMjkna7HTbo5vmMn5c/utJhYKiESXpyDIsXjnicjdddX76otz3+VCg86Kaklz/4TkVWK8vz3Ht7OirauqT05vLgAZHUmSppcfHD6nbjW2q6LPD/dDqtqampMOYul8yvk2eMAbEz6aVOpxNOv67VavrZn/1Z/ZN/8k/0a7/2a9rc3NTGxobe9a536V3velfMCKRSKd1555264447dP78+TennA4RPcDGMhJb0ZFud7CkisoKJhuCAUWAPOn3h4fh+Bb4uVxO9Xo9JthRNCwS9piH1IhPAB6a0iw3KJ4yYBJQBCYVI+NW24U3KbhMKH1EQJOkl3t0rDjxGO2WhptKA7PxXDwjiqJgOJJMtufxgLPMmb/XiQ7GXVJsXWPScDhB5nlYFAl2FIHm79JwS4+lpSWV89/RsdyQCGqOF/X6u+/XaCI3yhhjFFDwfD6vldYzmrs+lNV2e1KdTim0yf+l/44sHDUhY76HMuPj89zvD/ZsRp43NjZCeoW62k9+8pN6+OGH9fzzz2t+fj44BG8Tv/vc7nXd9JQxL+eiNMyrJPgcC+VVJEAZhxQ+ACgYcJTPGUCsM5CLCUYgGDApnpPlx89fcY8URcP9etxISPFj4aJouE0L70SQk2QPwuoxJP8HQsFmk9B3Jhfh8aVdFER4btAtedJzOoFDnO35Y1caX6nCODqH4F4KhEJ+17kFcssU0nO5kkkKZF6z2dTCpVFFF4b3XjtWVKE3GZQZL0afMMa0dXd3VyuZK7H3VauPqJMrK52ObwvKlQyBfN8pnwPmmjl2eI0sc57K7u6ulpaWlEqlwjm2zWZT73znO/WBD3wgxvj65e066DpQOd0juYa75wJ7l8tltdvtsGWIFN+HBQFGmBBQ6HiE26Glky7NZlP5fD5YcLwjxQ0OY93atdvtsNyNSQIJMEHkVBFA+kj6xxXEGVLuc2V1hOEkBMrBu1Op4TI1FAooDavJ5t0gFH+mx77839lFJ+kgvlBuh/pOJIFEpKGVRzD99DcElrn397riYvyAyewgMbm5KbWGcrZ97C3q5UYkSxnRb57n+yH1+3391L8wAe9LvY2edGhYQpnLbSkVbWmi/iWl23Wt7TymZvm4ut3xGzgD5IIxcU/pCk3bOP9kbW1Nnc7gNGtfjuhG1OUheSEP+1033abECRQgIVAWQcHTuIUmGcxAAKu4z72lew5+dwKJPWPdKHgcxb9Mqsd+rFn0AfY4l3QICuLQ2IXVk8XuxRwaorhMjFtIjweJUb3YwWNF8oPEwhBTySIKjIYrr0NWxskZZubPyTfGik293No7ecT/MaaEIjCz/mwUnjWOIyMjYcf848dfGI5LJNUXyirmi7HCejf+rAhysq20bcRaVyr/zWdU/umvSqOjGj9/XofPflr5zKKir7elZl8Ls/9GrbuOaPkdP67an/qYdtsjMccgxVdG+dgg165UY2NjSqVS2tzcDCt8WJRBvx2N7XX52O913XSbEmhmBB76PTwgM1yITfzI7zTA40kpfsw5EMshppMwKAVGwK28QyuIDzw8XsiVxJWW57AWlHs8RYGRcaV12MxzkrCH+xw1+PNpt3s9J9QgjorFYkglMQbeFifZ+NxzoZ7q4R2ME2PhsJwx5bvScKlV0nA5YnIikJgbZEL8TdyYzWY11l8OZFAvk9bqw48FWXLDALQmHHAv9tUfmtdPfWMop6Vzl/SWX/mV8P/k/gLRxbZGL57Xia/+j9p6yzd1bu6/kZSL9cvJIJwJ88t7fUzYuXB9fV2tVkurq6uamZnRyMjIDeiGtievg5TzpoXvCLEv8fIJxAI7eygN6xZdGIFwlFa5l/HcqL8D7wKMoraXzaPJu4L3aYt7Zs/HMQm811do8H+HxK54WFOEMOmppCHZkDQKSWLBJ4bvuCCm0+kQd9IPlMirqXgPHoUx9/ytNERBGD7awdxAoHluV1JM2V2QIf2IQ5NbhuCR/NQyUEzvaxYL9iMVWwOvythy1OPIyIgqlUpY4OxGaeLwA+rdMXxflPiRJI1KOmM/U1LUkMZ/5is61vjbkuIhEvNGP52ZdxnB4IFMJiYmgpFfXV3V5uam6vX6Damg5AXy2u+6KSGEoriVd6jrKQW8m3sGF2zuQQhRqn6/H5hJBAFYhlB4TpH3JL2PM7QU1RPb0Qf37g5ZGHyHdskiBBTM4xMG2eG0x8xcHj8noZ9X4TDxQO7NzU01m81YvtXJKAyeowJnHPk/UJrvMh6u+CgA/aNPzjK7kiQLyZkDPAtw141tNptV+tm4T0inhkdauKdxVOIC3u/3tVY6qSf+mz+r0194UlN/cEHReh+JV/X229X7cEFrD7xb7fvmFUnKF65ofvXfq/B/uaLoojTxhW9o/Eef08rKA7HnM4ZuLF1mCKHcYUkKaSmesbm5qVwut+cZrX7t93fpFrYpQbuT8ZNv8+HlfQglCtVsNlWr1QIpAIvIZ8SM/k6ew2pyzycS63ockM/nQy0qxcjEpG4Fk0LL4LN+kvWMHl+iAExY0ot5oO+snL8D5QdlOMvqhoK20m7il1qtpsnJyWCt+S6GAGOZTAW4UXV0A/QnZGE7DuYWmLq9vX0DyYEx9fJH+sV7WKnhRosrnU4rSt0I2Hiey4AbRMZxqOQFvVT/Ab363vfoxPs6Gmm/kTaLIrUWFtQDum+8cb7ndkbV8mO672f/stK/2FL60zsa/4lvaal3X6yem8vbTJ8xhO4QGGuH88wfC9ynpqZiRsd/xwHtdd3S1pikMvBwjUYjHO5D7ontG3yZjhcMMzGTk5MqFosqlUqan58PeSJfjkZeDevELmlAH6AohsPX1vHvXnQ8AucwsN/vhxiBgcZ7smkT24fsBUtRNhdiHzsncLyAICl8TDCKS75xdHQ01HA6YkBgGQNSQ8laWE858fyYoryhQMSFGEsgPEiEz4CrtIe5Yr59ew5vh5NmSW+RSg8LD5wY83LNZDqC5zabu9o4ckSamIh9p98d1lZjwHZ25lT/xGmN/+PnpQ2pUv6O2ObE5zY5h24Y+MyRE7KQRI47Ozuq1WphhwS+f5C39OumbC2W+tq1a/rKV76iarUai43oWBRFKpVKKpfLYYCZaD5Pp9NhAS+QxZWI37mw7DzDz8Ikpm02m9rc3IwNarFY1OTkpMbGxsI2+v1+P8S5Hs964M8zfGc9PJLDdASasaGvDmed6UWQPR51uM9YOwxF2SGFGGv3wIwfKR/66ceb4439/W40eY4jASqLPHTA2nMPBhVjOjo6Gog9lNh3nvDvu1BLUiq6kd3GC6EcjuKcoGJBhBsR5tOh8mC+0moePazx+eelV6VohXmPy73Hn27ckqSmy4XfQxtotxfL7NXPN6WcWGEE8PDhwzp8+HCYNBbwsorD93nxqhfYR6wzFLmnAiBzPC6lk9VqVa+++qp2dnZ06NChsOaxUCiEYDzJqrE798bGRkwAaXO5XI7tCkBgTp+9AJ9Tzth5D+FkkrDYTgR57JmMl/gexggj5nCZYohOpxNODYNFdKPiMbp7fRcSkIvH617V5aRIUgm5EHiHsbSJ8XdE4ogh5tH6e6QPohuPlfDwxolF/gZbSibAFzVIw1pmJ9Akqd/uS21JK1L0TSlzV0b9ftw7u6H192PMkv2kjw5rmSdQXDJ1xpVEBDH92/cTDdk8vNajjz4qKZ6Q9YNWk+wlg8xEu9LCJlLFw9IdL52DJFpaWgreYWpqSpVKJVScAHmBZhAR2WxWU1NTKhQKYWlbvV4PS9CuX78eW6V/6NAhHT58WNPT0xobG9PGxkaYKElBQXy3No9995o4hJu/JyuLUBaH/j7u/f5g8XW/3w91tr6nrzOsXmjA716tlYx3koLlKMg3sUpCchd2ak2dxXSB89iMdtW6q9rUFc2/0deoJCkXV2Yn0/xZIDbaGUVRQBYIv3s9lr+5t+4/35eeZaTjRiGK4gUzKJz/jXHgB/kB0ruB9RBkr5QJTmC/65aUk8mB1kcomVQUYhADNMNWG1xJmt/JBDwU6ZHkBDebTZ09e1bValV33nlnOFKtXC6Hyeh2BwtgO53BRlaUiV2/fj0mmHi8VGqwt2uj0dDKyorW1taCR7zzzjv1oz/6o8FLuZAh/BQHIJx4R3+XQ5ykAmJJEUSEzeERY8wqnFqtFlIKCJzvbuDC7KQWHs7jU1936XOEQUXQk8SHs/XExl5H7B7PSRzmvN/va/ZqU2eeXw/v3bjzNjUPTau/G9/KJdk+noks7gUvkTMvsvD5k6TG1oz67UG6JZOqK5drqNerhOc5JHUnxDz533E6SR6BsXfy0jmLpEztd9005pSG9DIegMlzKEgDPf3gcIMfV1AKiDudTlAuBAEKu1wuh3zawsJCKDzmvcAXEsIMGAPk8KPZbIZ4l7bV63VtbW0FLz0zMxNK6OiT72rggsqaQIyNH93uRQwee3taAdbb40PIBCZQGlQ5bW1tBQ8A5PIUD3ND2MC9EDa0CwTinjRp8VOpVKhqwZOiBIVCQdvb2zectEY70ul0eIcbJvqZTqWUshiv0x1Xp5NTFMV3wvN4OZnPTQq4x+teckr/HBWsVR/VmexvS7t9jXz6suY/+AUtr/y4er1hPDz4rpROrymdSmtK31CmX9dOe0obmYfUyVS0uxuvHOt2u7GCEZAF84Rx9vFCtt6UcnrlDCwssMZpeywVXgnBo4Es3iWPiIL74LoHxAoiCOTnSI/4hTADjfkeXgLIh8KyqsFJGTwq91ByRr2wr6hxa+dsYr/fDzs28N7kqgYnx+iHK5zDJI8FK5WKtra21Gw2VSqVYsiF7zgsRDgYV97HODhc9fbRNwQcIoe9dImfUD6E0o0O3pRneRyZyWT0yuzLek/UHVa/zCi0w5WDv9Eu/9wNfrfbDQum6YvHysgIsta966iqj96l8a++qOgZ6fjv/mOV3/KSlpvvUCM/p8yFnjLZJc0d+ppmvvttpZ/bVfqLTUXVnnpRWp33jWvzA49q6Z0fV63+YGD/Y9DZyKOD9qVNkkPJ66aEEBYTJen1emFf11arpfHx8SCEDI6Xh3l84wQI3pHBZKkZk8KziBOpxPAYyi2pC4yzpEyOQyxfkQA5kjy1mOfjvR1+Y2wYH0cDCA/3eAztkNMF2tGIQzPQA2Ras9kM+UOE1oUfDwXjitK6N0HIEXAfQ6/Q8gtjxRpabyNy4gYQdtLXzbILfbR8QUKIJV1/z9vDePMsj9EYI481GUfaTuURYwsp4x4zyGJ6Wud/9C/r3md/WZnlltJ/c1szqf+o6bH/JM1G0mtvtCzqK+op7OonSSl1lfvUumb+0x+q8qNP6vwn/wet9O5Wv38jsePOhLk/iPzZ6zrwbhaR0ulk7MF+PpTS+eA55GLiKHx29g2om81mQy7Vod3q6qokhdQISkLHseRYUk/uu6Ag0MBuBMoFGyHx3/kO8SVbTcJSoxB4aG8XZ5bsNTZOrjjccQPjsU0qNTjrsVarqVqthue6kXLv5ePgLKLfg3Iybv4s+u7K7X2lPxgsj7UdiTDOnU5Hm9ubOv31C8pa3r3dPRTmgXYnSUTnN3gvKRQ3SA4fnd110md3d1fr992nF//xJ7X99kn1M5K6UrTRV/RKT1Gvp6jXV9RVUMz+TKT2zJQ6PzkuHZWiupT739d0/KW/rUKhHRsT+u/j5UUnyWuvv3Ed6DmJY6jkwZIXCoUQuxEvSgrCy8Dg0mGunMHCW/hyLDbUAhLlcrnw7IWFhTDJeB6UjhSNs3VOKJAHY7I8JmFA/XcUxoVBUvCMDDaw25XK4SDeg34ntzxxA8Y4MRb+PrzV6uqqcrmcpqenY54Cw0H/MXYQJTzHEQ6GC0/F81xovNjB87nS0MM5eeVkUFLwZmZmlO5v68jrw3WYO1NTap04EeMImFsMApe3m83jMOhuhFwx3Um49+z2elqe+SE1fu0BTT79tA5lH1f2i2sauTbYCa/bH9H2x4+qPyWtrr9Nvfed1k7+vSpMNjX+ym9p8md/V9HVrrL/y5JG/8GLqqcfDXMI9wC/UalUYmPjl0P4va6bVgjt7u6q0WiEihUUELqegWDQOA7eLZ17XN8IyztE3hNyAqjZaDRUKpU0OzsbKzx3GMgzEDYUk/d4bhJP4MLj7SVp7h6LYogkycAznZ3luc7WUiLY7/fDIUxYeITJSTK8iHtWxnttbU2HDx8ObW+1WrHUh7OHfI/fUTgXCI9LKSyB2CB+dK/o6ZdkjMjz/P+eytjSso5cGBIgr98/rvahGUW9oaBioDBuPMPHE4NIPasjHTeySaPBszCgjeysOm//hNYzP6rsO1sq5F9Xr99Xuz2i7d3b1O121O32VRodkJLt7Ypqx/6Govcsa/K3v6j0SzvKXLqg6OhjMZQI/K/X66rVaoEnSF79/vBMnr2uA5WT8jzPnzExMLWec/NYweNVGu5EjBSnxPm/w7vz58+rVquFjYGplMEbIPgorMctbjScfHEox8XffGmSs4B4IzzTXpM/ZPnix/3RniiKQqHFxBulZg49fcJ6veFOBvQjm81qeno6pH04pAhvwhyxM4UXRQCRGVdXNuaQ0klnsplnN3i+QJw0jZNL9BtEhIfu9rraXn5G0c5wU6+de96tbi9eDukGMDkmtIcaaIeS/r39kADtTt7f60m7/RHttO8LTiaKdoIMeMFHJlPQ1p9+hyb+w5cUbfY1c+4LWj/8Y+r1hnXcFK20Wi2dO3dOU1NTmpycjK36YdxBhntdN2VrWejsBdf8360WS7eYPIeWjv090GfAuM/ZW9jPnZ0dTU1NxawPnsuJFY+dkkyk56uc2uf7SfYYxeM5eC8miOfRDsbE99/BqyYrV3Z2drS+vq5KpRKr+8Rb8RyehefOZDKanZ1VtVrV9vZ28BoQVL53MLvxObvpioNA44WS1UZJ9tYVNckpJOEj3/e+p1IpKdVX+uWvqPSGLLaLRTVO3n+D4UAe3FB7egqnwPjwPU8R0bdkzM5Ye4zqMuLz5POdzElvnT4jPRJJX+kr/aWG+m/vhfCv0+loamoqyFij0dDq6mosz4mB8L5/z8rprB2Cwx43NJgUC5CRl+9F0DjVjCdFcEhPoNwM6Orqqk6ePKl8Ph/by8e9mjPA7mmI/bjPobnHdl4O5zW0STjnDKHfx3OIf7zPsNuupGwOhYIBb4DtCLhXFKVSg31q2Cv10KFDsY2/CDF4PptO8Rl1xbTNtzx1TgChRbkcFTgKIjzwvG0QqjfGhn/T6bSqjVf0oX873N384qm8lucmVDAl9FPOPE3lhoP73Nj6O92wOFPqsb3HtA6J3dASYjncZ26KxW31f1KKviKNFC+rUHhVu+17tLW1pWq1qlQqFfLznLq+sbERwhmfpzed56SjeE/+lkqlAqxyIaQY2mMVhAGBJlcKtc4zsSIoFxbv0KFDQbD4HGjAgHkC3pXDhY4BcU/NhVAC05LsoBSPpXiXowP3crTDhcS9MzFzvV4PAsXxfLwLQ+iMZb8/KOejCoqd8PCaKCgEGH3f60xJ1tO6kcTjMR7ZbDbGsKMwhUIhtpjd54L28nzg9uraf9Chq2+ksCRV73i7dqKc2m8c2Iuh8jlkrH0Okkws99BX5h20g7dHzphT5wpoe5IDAOYzxyhSNtuR5t5o04e21e1vhmWH/X5fly9fVjqd1rFjx8JqFJZOevVct9sNp5ftdd2Scjpp4LW0DoNIvCf35kHo3dKhpFje5D4uDJJjdReEJBRwiIlVT3pSVxA8qzOn0rAwnGe6QnvezMeG3z1/iHI7eYIwYC1TqVTwlp7e4f/UarKrHbnlSqWier2uer0e8nsw4CznQyHZiMpRQ2Asu8MaUbwsfdzZ2QmxK1twMp7MG/PLnOGF6INXT3Wiqj70e68P52syravveKc6u8ODfKUh8nCDlyw6cdKIcUY+fH6YB7gClCw558wd/MrOzo6azWZQbO5xBU+lUtIDku6S9KjUabRDFVW5XFaz2QwOx8cG2OsVYG+arfXEuscsyYJqgnRYVjydFD+botfrhXNAfDDpBIMBZFpYWIgpGZ85K8kFFHMo7pCM92MYWq2WRkZGYgwkSogQ8h7GwuF5Mj51S4/Vpl3J/CN9pp8+zqyeZ52rv5MqnVKpdEPcSgE4XhFBTHoU94osSmdcmVuYcPYA8hCg2+1qc3MzKC59QXBZ2xt2SUx1dfzzn9ID31ofbB+SkrY+8f9r71x+I83OMv585fKtXC67bJeru+0Zd3cyk5nMkIzCKAqKQEIRsABGoMCSSLBgxSWb/AMs2LHJDjYgRexAQUKJACECGiUZZohmOiAGxj3uybTtLndV2XVxVflSFxY1v1PPd9rtTppFZuEjtdqu8vd95/Jenvd533O+l6TnntNU8yzMVa/XU6/XSymoy5CzuMRvoAUUz0MM+uMNeXGEhAKfnJyEPrh8Y1i5vt/vq9ud1WlxWfMbh9KmVDr8jnZ3P5FClX4Cgju1TqejQqGQyu0+rl2qnP4eCWANAoWwxHEWg+c6t8h4SYe4XnDsCuDxThwPMUlc62eNYnXpY0xMYY3xYMQVHlO60XDlYHwOWz32pd8x1HWFzmazIRcM8XN8fKylpaVAw5NukZR6L83x8bGmp6dVLBZVr9fV7Xa1vLyc8ra5XC6UHHqBA2MD2Th0dPIlZrTd+zscPjg40NLSUshLu2LzDElqdf9XL/3z95RgSF+RKr//m1rorCmbHZNj7XZbmUwmEF0eqqCg9Bn58RdMOcPvu534LuYXgJMuC35sjsfmvnZkB5rNgtq9G5r/k0MlI2l25oFOT481M7MQlBKjhgzguDgBBOQRx+s/tnI64+WL5YtG4tdpeyaEwQErUG6UzGMgrJlbY6Ao8ZXDKFcU75MrL9+TP8RouGIjkFwbw2Efv6cZnGjwiien/d3yxjCKtEe/3w/KSJzppAH3z2azwdNTrNFut1UoFMKz5+fnw86ei+I35p2KFrwGaxX3HSED4pHzJkau1+sB9oKePOzJTme19b1/0bX9j2LNaan1uy+oOfyZgGKAgf1+P3guh31OLkmTo099fCgd8nRR+OFxZWz4IbsuincdtTlKenj4eZW+8F9K/kwqf+Vd3bixp273laCUrPFFKCuuzX4q5ZSUWkxiJFcCZxdhuByOMRnARTpNB3H1CLHDTKwV1o6feaYXxDP5Hld5ygdvitfGCl7kGekXz0ERqfGNFdRTMu6dyce693EmmoUmNvRyNBcOSannDgaDoJwc/oXAMf9+UgX3oYDE96JKSpUB+jhAH8w7BSacPIGAcWRNqVQK485ms1pst/WLf3cvnIZ3/uWC3v7UV9U5nJN0GhSUw8KBl+59nIMA/p6dTeAwyId59i10yB6ekbXyezJ2lwOHxs6ruEd98OBLunnzH5TP7Gvqb8/0qd/4ln74wxd0dpaEM3Yd4XF9zBBfFnM+8WjMJEkCOwcEoLNuhZhoJomHEscxmZyrwiRTOcM1rrR4O6oofHJ9cDFjyPVO6LD30a0qwgXkcCXyyXWow0K70nOdQyknNlxpHXVwrddfohQx7c/3jJcYs1qtBgXibymUz+VyyufzKhQKKhQKwUPHDLqkQACxjp6L45kYM3+Ge3lOlcjlclpYWNCt119XrlYfr1U+0f/8yu+o2V0PBQ/MvxN2Xo0Wx4LMPUfdAP2JE10m3PsRsvjxM6wNRt8b6+SGFpRBOzlZ1J07v6fT3yoo+Xvp2r99Xy8+/xeamjoNMsfcuIK6AYwNQdyeGHOy8KPR5N0jwBiHqL4TBUVhoCwsLt2rRyAfGHgcIMdxJsLBMxhgDKt90VkkYisWLz7aETaVxeHZbhxIl2CRM5lJLa2nkbyiCOWK2UTQAn1cWFgIAhenhWK2URorQ6vVUrFYTIUgfg1z4NAVT4jRdDTiYQjz6iQZ/fQ8dy6XCyRHyC2enqr09tvBa7a+dFv75V/Q6DRd9BHXR1OrzRr5906alUolTU9Ph/eBXpSXpLgjDlncwbCuLk8xV8C6+VqPRiNVKp/Vm8M/1Oe/+nXN/nVbz3z6n9S/NdTbP/yKkmQc++dyOW1sbDySBvJN80+lnLVaLVhDFosY0MkbhCu27jQne/AKfObBspeCsSgImyuqpyr4exdg75t7CSbYSaeFhYWUdyS28ljFiwEQUkgYBAqWDiEIm4un0iVjcfzK+FFqj51jr+tGJkkSra2tqdPp6PDwUIVCIRUzEvd4HTGe39fI42+fOwSJ9XKL74wz6QNPO8zMzGi+XtfK3Y/SJxnpR1/4VQ1HeWUyk2ql2Bu7F/aUl7+unveMAvM59aLX66ndbgfOAsO/v7+v4XColZWVFHJxOXKOIvZsLgf87MiwVvtZvVX6Y73651/X3L2WRt/4riq3f17njXEsjf4QusSh4GWw9lLlfPjwoc7Pz0OuEW9DkhvByWYfPRHbYaLHZySmURqv7HAPi1LicTwWimGIVyR5qsUT4sBMjweZGK/TZZFQWCdofC+jxyLcz2MW+sG9XLDdWsZePo7lndKP2Wmg5f7+vorFYjjwKu5bLFBOoHksxS4g3xvLd448+Nwhd+xtCoX/Vib5iJT7TFbdV9eV6aePlpwk9adTBt7nlONo3JsDUb3CBtabUzEkqdFoqFarBSKQap2YmfbmxRSMifDNYSmyORwOVal8Rm+99Qf67Et/peNf+4SSt1c1GPTCfLRarZTsxuHQ49qlylkqlbS4uBgG7+9NIW7gIQg/ncazsMh+kgLC6LEeP8fEjgsZRAALyT1QVhTJCSQXQCldJE0c4d4XYfF4wK27NHn1upSuEeYenoPleS64jNcFhf648DJGh0OuLFNT43OAj46OHmG0UXQn7zigDEjpNcP+LOpzfV1iw+lIiOex3sPhUPnGh8qMxuvS+/U1HZ2/qExmgm58uyDhBf2BgHSGHwQDAcVnDnul8ZuniVmPj49DasrP5fXnSunTKFg/Z635zmFtunxzpN3dl9Rs/qnOz/tays8pN5sLZZTIkCOj/7dylstlra6uBqViUiWFd1lATjB4AngYN68fdYJmNBqlPIFPCsqDQHqNZgw3XPCcCfMktnt4F3KHlV4nGsMPh3/uff2+eG765J7EFdYhjf+NQ1fvNwYv9nKQI/l8XteuXVOj0QhnGDlcgtDx/rvldgZ7Zmby0uLYELiwuuJ7n1Nz+o8j6VTSopS8ltHUYVbSo+f2otQIMEYAOXG2nuod0k9eYDE3N6fV1dWQVYAjKRQK4QRD5pL702fmiDVxj4kseZjiawmBJkm93vRHz82oUCikHBhjc/l6UntinjObzQYcz+RAnWcy405QkM1A/QVDzWZTU1OTV9pReyqlq32wmnhct2YeZ/GzWzAnFFgsj10hPsg3etUL13v6xU9WwLs5FPS8rVtCj8c8tuYeToBwHeVwMQJxZWVLF58NBoNgjHK5nNbW1vTee+/p6OhIa2trqfGDRlAC9xwIJ8IjKQUj3bC4J3flce+P0Ul5hYzUzxeUaWQ1HKahths/rsHoOWT1GJgXBEG2eL42m81qZWUlXOtGnLlFoZlDN/zOZcShR0xwes4eXfExxPG5G+WLHMhPrJweS2KtgBwoKQu4uLgYDmfmYCyC9U6nk1JSacIEM4EObXxBPL5zwiIO6Ck18+8RLIeuGAOfaJSNCfYUgd+POeHvEAq3sr5QPMs96HA4DPCbz2JEQT/dsnuKw+N7Th2s1Wqq1WpaW1tLnQJIP9iokMlMkuD026u2gIHAzphQ8rAkjvEdxQx+aSB9ezwfo/NxbHxyMlYQXk1PuSFeUZq82S0uWPEwYzAYHwRHOSOe3D2xr4GHC4ybMblSAUHdEcQQ1/kTX0MPcySlDiJg7nEsjhov1b/LvvQDgxkwOUNpsqmW81/X1tYC6zocDkPh8+rqavi7brcbrsWLInTulRgcp+WRqiHuSJIklLZ5oM3kO8tIfEHaIkkmKRf3hg5nUWonArrdbgo24vGcTXYlc8uJMDijjTBwjVt5xs39PdeJonjlEdC2Xq+rUCikFMlJG4raQTc8jzVYXl7WzMyMer1eKOD2mBwIiQAC8/BGzEvm5bxGr0jJW1LuOx9o7gt7Gg63godztNBoNHR4eKjZ2dlwqj7zF7O4zmS7ovA7/UCR3dhxvWcJvO/IBNfFoZPfxxXTU0j8jJcfjUbhZVzOozgqfCrldNjosQt0vDO4g8FAe3t7mp6eVqFQCEXZflR/Pp8PJWDAWKyoJ+yZPHJ+F8U9NIexg8HkBa2ueFzrk+/sKDDJK2Ic6pG8lhQYXFdSFtnjWe8nRIoTKB57eTzMfR1SoxCee/R0xnA4Pn9pdXVVlUpF6+vrWlpaSgmYlzOi9O5dPB6maidGFXznCo1CUkxSKBQ0MzOjD4+/qNt/+S3Nf7Wpmb2WrpW+offe/yNNTS0G2Mmp/dS1QvZ5PhfFi2PdmF1n7lBYPg95V0NfrCUEWfy9IyuP9X0+PDZFRn09nSNhzXmu8yqXtUuV06Eb/6jscUHxYJltS9R/elxC/SWbtYld+/3xa+N4GzULwyK5wPtRJUxCzNghSEBwt25xcYKTI8C9OObAOLnndtgsKbVwfh3PcuYztr4XxSMeo3IP967009NKm5ubOjo60t7enorFYqov9IGGoQRWYrw8SR4fQcM6eIWUNFF8EM4YppZ1f+qX9dy3/0aJRtrQv2p75zPae/A5jWZHymfzGpxP0hYcFg6k9LxiDP8dljL3ToD5NbGSopDMhSMLv4+vkadefG1caVkbZNY5BJSTvsTXPK49UTmxbF4N40c9MgF4ExSvWq1qfn5exWIxeECEDSaNQu1ms6l2ux0gj2+ExZp67MUkeiwJa+yF+P4st1ou+E4UuCdxJhK04ELC4sR7TZ2R8xPxuY/HcR4TOenB4jpR4fFjLKQ8m212Dx48ULVaValUSgkvUDo+/BvPCPyiDy4HhAQQLR6DE1fxescxVB3pvfd+W1NTLc3P/7veeOMVNZvX9R+l7+ubX/ymvvbdr2lrsBXWAAWJyTWHs24k3ANmMplQscZceVpOmnhMYlTmA7lG6Zgv5M8JS9bfWVqPWTHWjkRcsZ3EdPLyqZTTLQ83B5KwKLzFy1+bsLi4qE6no4cPH+r4+Fg3btwIhJHHhjBuSZKEs1hZoHw+n/KEbp0dnnouLM5HxpPikNChOs9gYZxppb4zTqXwtx6vetyMIeNZHot433iO52X5DAPmtD3CGRMV/H/9+nUdHh7q/v37Wl5eDmSes7sc+YJQY9hiogrUAdRzUgYh5HeH6wh1pyO9/vqXVa//nLLZ5XG402molq/pR4MfqXxSfsT7g0DcmNKfGPLHMZt7Iw9n+OfjRb6ZS1cmj/097OAa/sVlgMwL9/X0nBtwaWKo0YWL2hPflQLTdH4+PmCaeJH0xNnZWdiPVywWw+QWi0VNT0+rVqtpd3dX6+vrKpVKqbpUPB/XZrNZHR0dBULJj+AE5jrR42QMk8WiOmwlPgE++a6P2FK7B4NFhAV06+hCw/Md2iKgTt374qKAzKVbZeJ5EAdjwco7ocS8uJJeu3ZNOzs7qtfrunnzZkA/fi6wv2wXg4RRIYmPQIFmCCc8sQ4U7fcnL5HyeDBJZrS19VKQl9nsrIaZod659Y4+987nUrwGz/V4DrjN3wXBjUIUr9d2hfK5d0TiiuRr6aGFX+NenM89buRnN8SeInQIzBr6G8V/YuVEOPmfShQG7Z6Bv6ednJwEkqJarWp7e1vdblebm5vBY/nkZ7PZUDxdq9XUaDQ0HA7DnkfujzBKk9cHOKR0Gt530kDqYBwc6nFvT2lQZsh9+My9rcd0cX9c8N2bAqe5Pia6PKd4UTqEe3riHE9Iv1ZWVnR2dqZGY3y2DXOTyUzergaJ5B7BiTOH53h8N5KkM3gmpXNeJUR8J00KTe48f0eDzEDz/fkUMeLeyskwh4XOsmIk3AsilxgMRxTsGOKzk5OTwINwffw8PncZc8V1cow19qIKZ2bpPwrrMvy49sQT38kfOgHi1ggLQUedHMpkMlpZWVEul9Pe3p52d3fV6/W0sbERBsu93AMtLS2Fk8w42Q/B9b18KA+FEhgRIBkEVrvdDhDaF8QZV4/rQAQoJsKCwKAgQD5Xdj+b1BfE86hAaoTNF/Yi7+vkg8dUeKnRaBSKGQgtNjY2tL29rbt37+rWrVsp4guj5cUGzsYzLi+QcJbe2WzQDOtBP52hDqcAzpzpnfI7ygwzevXeq4GgcXjpRRrOKbjBcqSAzMAxkHZyI8N6eAjBuD2G514+37ExdOX0dJX31+fW88TMP3D4cYdN/1jKiWARbBMz0kmPIekIlD+njCEwt27dUqVSCYXI5XI5FZQD0xgEsZbDaPrkgTdGhHeI1Gq1sL+v3W6H82yk8Rar1157TaVSKSiS9CjckR49Q5XPkiQJHoT+es7UFRAhQFHdGjssQkBGo5EWFhbCgsY7QjzWIn6iLxwEhvLOzc1pfX1dOzs7arfbKpfLKUHEACBgzANzwCsRMSLE7S7EPk7WiNP6nDxDjnqZnh4uPNRSd0nFfjEVKjCPzpI6bOSe/r2zn8wLc+198HvBQoPKnFyKyS7mxEMR5t7JHD7z73mWG0/Wq9lshl0qjOGi9sQKIa++iA8V5jtpXGubz+dDhYk0OYMWhpIDu9jtUiqVUrQ4FR8MDLIIQoZN3xBTR0dHqlar4a3TWPLRaJR6a9j6+rpWVla0tbWlzc3N4NFQSFdAh1SeAHcrzaLFu1S4HzlTBMVjN092+/wlyXgHPQIDqwqz6s9FoZytdK/HM5999tlw4NbW1laqoMSLLqhF9fnwnTwOrXke6+oQnzlwb4ECZjIZ/ef1/1RntqNP1z+t0llJ2dlsKkyJ59LH7WN2Jt5jQY8bXSljDxcbA4fS7mz8PvweK3/MfdBXTqcghPC/dWdzqf5d9qUrYDabDZUhw+Ew9VKeXC6n5eXlFK53mOSQ8ZOf/KROT0/VarU0NzenZ599VsPhMLycBijL6+HPz8/VarUC6YQVAsL6hGcyGS0tLSmfzyuXy6lYLKpUKqlUKqWOjgBmOex04ebvXDCd0XMBpaHIrqTuIVEshMWPFgF1cD0sLXlDhJT+uqB4POO7R1CMF198Udvb26pWq7p9+/aFhR0olDOg3Acv40jFjYR7Rie3MEDOrHeGHY0yI2WSTAqm8jzuj4Jwf59zJ11cQTzGdBYXWErj3g6LXd4dfvIcEEa8ls5cu1F35zUYDEL4wbY+zl1id81TKSeKQKc5J5VqDiYDKIuAu2UBhiK0FCjs7+9rf39fH3zwQVBqdrTg+RDyMS3fSUGybHa81y+Xy4U0zfz8vBYXF8OO+CRJQikbQtLpdFJn6Hh8BHx3ocUYAWNZxDj28IV1aOP3Y07cI3ihBAIJS8280gfiYObaBc/vCQoZDAbK5/Pa2NjQvXv3NDc3p62trSDMKLWnBJgDTvVz5pV/rDPr4QLve3o9ZsMAS9Kr77+aYtl9Li9itV2JYu/osBDlwfB5jtTHhyxelGt0RWc9YlLIvSDIymWcNZ+dndXKyorq9bqazWZIK+bz+aAnl3nQJ+Y5ER46PTs7G2I54lC+Z9HIg+JpK5WK7ty5E/KYSTI+1Kter+vs7EzPP/+8FhYWQl5ud3dX29vbKevOtp/Z2Vmtra1pdXU1bBty6x3n5rrd7iMkCweKeT4NQ+RC44vmRIJDUZ7jf+uezT2cLzjsK3FlNpsN7/70jeyuaAiXexQXVJojjCRJtLq6qlqtpp2dnVAUwqbsWLAYD/dxoUZR3ZtCGrknQ+CdCBuNRkoyifLdvG5XbqfiQ0cY3JfPnATzfroHcyVGWbmfK7srnqSU3HphgssAChwz8Xwe94++kw5ja+XCwoIajUbYd4vexIz/j62cU1NT4VxaSAZgIfQ9yovnSZIkCAXedWpqSjdu3AgnmPOinTfeeENJkujll18OAfLp6aneffdd1et1FYtFPffcc+H4R9IDS0tLgXl1ISQWca8F5mfvKRPpCu35MVfAONbyGMuV3a2fF06wYB4a+O+kHihrdMPiMSGK4h6ZRb2oXM1jKI7KuHHjhg4ODrS7u6tnnnkmCJcLHvdESDHOMWzEQLCJgWf5DibmgDSZJK32V7VZ2dTS4ZLOl85TpzQyRubVFcENncNO1oEWowfGcZERc4jK37Iu/O5wFeTlxoFTDvkdI+tzKY1ZWdYZp8Rb2t2o/kTK6bAE4ce7MBCv5GGADIjPVlZWVCqVUknzbHa89+7BgwdqNBrq9/uqVqu6f/++fvCDH4QtaS+88EIYaC6X03A4TqAvLi4GhSXG8ngp7jNeH8jrE+15KrfIvlAe5+BhgGkIJsrisbeTKDyXGBPD4HG6NIlh3GPGCXXvO9DcYaaTVRihlZUVVSqVsP823lvrhAhCw9xjzBwiSgoQEkWOvRwnYEjSy/deVvndsrpnXR0ODsOrHb0PHgLwuXslDFScfnMFcS/s3/la8l2s0BeRQawH37lC+wkfKCByDmrz0OTw8FD7+/sql8va3Nx8+nelkI6gWiXuzMzMjAqFQvCuMKmVSkVHR0ehIshL1xCUVqulg4MDVatVvfnmm4GprdVqYVCnp6fa29sLpymwOOfn52o2m6FY2sv3vPCACeWfE0HOBjoR4RbWBdKvg3L3PFYmk3mEDUURY1LNUyDAH/6XJm8Jc0Y0TiW5B/E+ghyAn9yD705PT3X37l3dvHkzeDm33p7PcwIGD+GwFoVxJXDEwd/zjPPzc2U6GTVOxlvEZmZmArRHCf1a7odwY8w8jAAtxTAfBXKm1pEBRsjjXe7nikkoxX1iyOtMP4aBTIMrOfr0/vvv6/79++r1esrn82o2m0+nnL1eT81mM1SmsN2LeGV6elrtdjt8/vDhw1A2trGxoeFwGA5Vohql0+no4OBAOzs7qlQqYefK9evXw7ESx8fHarVawRu12+1gYan0qVarqRPNPMaMmVgvovCXzLDQTDYNy40yuyd2+EL1El7NTytwo+DEDgvnggQJhHIjJG7QXHnor8dVbjzimAgOgLwbxSCQPh6zIbwXNRQcY+he1NneODbz09RhLyH+UC6EmznC+4DUfDcP8xZ7QYeptNi4XETmxX8TbzCPSSGfL59zrsEw8zvfc85wu93W3t6eZmZm1Gq1Hqt/lyrnhx9+qEajESyH70xxC9NoNHRwcBAO+r1586bK5XKoxSUOrdfrqlQqarfbAbPfvn1b5XJZ/X5fR0dHymbHrx3gXRbD4VD37t0LND1sIqQUFs139yMI7p1YFBTA48eLYhrGCHxkVwY/I2gsMF6ARUDISIfgPd0jOGSS0kdpeLyDsqOE/mzGSYOixxMT22LoTk5OtLe3p+FwqOvXr6eMgYcj3jAU/i4UGF5nZxHkmK3HszFujA75Ve7L8zFg0gSi+7w76cL9ndl3JOQKwnesfWxMGCvywlzHkNlTPTzHja3zEYyda7h+enpa29vblxYhJI+zklftql21n2679HUMV+2qXbWfXrtSzqt21T6m7Uo5r9pV+5i2K+W8alftY9qulPOqXbWPabtSzqt21T6m7f8AgANXLpQXF9wAAAAASUVORK5CYII=", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "00409.png\n", + "IoU: 17.98\n", + "F1 Score: 30.47\n", + "Precision: 24.99\n", + "Recall: 39.04\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "00473.png\n", + "IoU: 41.94\n", + "F1 Score: 59.09\n", + "Precision: 43.62\n", + "Recall: 91.59\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "00492.png\n", + "IoU: 31.08\n", + "F1 Score: 47.42\n", + "Precision: 38.47\n", + "Recall: 61.80\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "00954.png\n", + "IoU: 34.70\n", + "F1 Score: 51.52\n", + "Precision: 52.61\n", + "Recall: 50.47\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Average IoU: 83.11\n", + "Average F1 Score: 90.48\n", + "Average Precision: 90.45\n", + "Average Recall: 91.03\n" + ] + } + ], + "source": [ + "import cv2\n", + "import matplotlib.pyplot as plt\n", + "import numpy as np\n", + "import os\n", + "\n", + "iou_list, f1_score_list, precision_list, recall_list = [],[],[],[]\n", + "\n", + "file_name_list = os.listdir('/workspaces/mmsegmentation/OUTPUT/MAEOUT_240422/')\n", + "file_name_list = sorted(file_name_list)\n", + "\n", + "def evaluate_segmentation(label, output):\n", + " # Flatten label and output arrays\n", + " label_flat = label.flatten()\n", + " output_flat = output.flatten()\n", + "\n", + " # True positives, false positives, false negatives\n", + " TP = np.sum(np.logical_and(label_flat == 1, output_flat == 1))\n", + " FP = np.sum(np.logical_and(label_flat == 0, output_flat == 1))\n", + " FN = np.sum(np.logical_and(label_flat == 1, output_flat == 0))\n", + "\n", + " # Intersection over Union (IoU)\n", + " intersection = TP\n", + " union = TP + FP + FN\n", + " iou = intersection / union if union > 0 else 0\n", + "\n", + " # Precision\n", + " precision = TP / (TP + FP) if (TP + FP) > 0 else 0\n", + "\n", + " # Recall\n", + " recall = TP / (TP + FN) if (TP + FN) > 0 else 0\n", + "\n", + " # F1 score\n", + " f1_score = 2 * (precision * recall) / (precision + recall) if (precision + recall) > 0 else 0\n", + "\n", + " return iou, f1_score, precision, recall\n", + "\n", + "\n", + "for file_name in file_name_list:\n", + " image_path_output = f'/workspaces/mmsegmentation/OUTPUT/MAEOUT_240422/{file_name}'\n", + " image_path_img = f'/workspaces/mmsegmentation/data/cag/images/test/{file_name}'\n", + " image_path_label = f'/workspaces/mmsegmentation/data/cag/annotations/test/{file_name}'\n", + "\n", + " label_img = cv2.imread(image_path_label, 0)\n", + " img = cv2.imread(image_path_img)\n", + " output_img = cv2.imread(image_path_output, 0)\n", + "\n", + " iou, f1_score, precision, recall = evaluate_segmentation(label_img, output_img)\n", + " \n", + " iou_list.append(iou)\n", + " f1_score_list.append(f1_score)\n", + " precision_list.append(precision)\n", + " recall_list.append(recall)\n", + " \n", + " if f1_score < 0.6:\n", + " # Create a mask for pixels where label and output are both 1\n", + " overlap_mask = (label_img == 1) & (output_img == 1)\n", + "\n", + " # Create a mask for pixels where prediction is not possible (label is 1 but output is 0)\n", + " not_possible_mask = (label_img == 1) & (output_img == 0)\n", + "\n", + " # Create a mask for pixels where prediction is better (output is 1 but label is 0)\n", + " better_prediction_mask = (label_img == 0) & (output_img == 1)\n", + "\n", + " # Create a copy of the original image\n", + " result_img = img.copy()\n", + "\n", + " # Define colors\n", + " green = (0, 255, 0) # Green for overlapping parts\n", + " red = (0, 0, 255) # Red for parts where prediction is not possible\n", + " yellow = (0, 255, 255) # Yellow for parts where prediction is better\n", + "\n", + " # Draw masks on the result image\n", + " result_img[overlap_mask] = red\n", + " result_img[not_possible_mask] = yellow\n", + " result_img[better_prediction_mask] = green\n", + "\n", + " # Display the result\n", + " print(file_name)\n", + " print(\"IoU: %.2f\" % (iou*100))\n", + " print(\"F1 Score: %.2f\" % (f1_score*100))\n", + " print(\"Precision: %.2f\" % (precision*100))\n", + " print(\"Recall: %.2f\" % (recall*100))\n", + " \n", + " plt.imshow(cv2.cvtColor(result_img, cv2.COLOR_BGR2RGB))\n", + " plt.axis('off')\n", + " plt.show()\n", + " \n", + "iou_average = sum(iou_list) / len(iou_list)\n", + "f1_score_average = sum(f1_score_list) / len(f1_score_list)\n", + "precision_average = sum(precision_list) / len(precision_list)\n", + "recall_average = sum(recall_list) / len(recall_list)\n", + "\n", + "print(\"Average IoU: %.2f\" % (iou_average * 100))\n", + "print(\"Average F1 Score: %.2f\" % (f1_score_average * 100))\n", + "print(\"Average Precision: %.2f\" % (precision_average * 100))\n", + "print(\"Average Recall: %.2f\" % (recall_average * 100))\n", + "\n" + ] + }, + { + "cell_type": "code", + "execution_count": 27, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "1421\n" + ] + } + ], + "source": [ + "print(len(file_name_list))" + ] + }, + { + "cell_type": "code", + "execution_count": 28, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Number of sheets for portion 1: 2400.0\n", + "Number of sheets for portion 2: 800.0\n", + "Number of sheets for portion 3: 800.0\n" + ] + } + ], + "source": [ + "total_sheets = 4000\n", + "\n", + "# Calculate the number of sheets for each portion\n", + "portion_1 = total_sheets * 6 / (6 + 2 + 2)\n", + "portion_2 = total_sheets * 2 / (6 + 2 + 2)\n", + "portion_3 = total_sheets * 2 / (6 + 2 + 2)\n", + "\n", + "print(\"Number of sheets for portion 1:\", portion_1)\n", + "print(\"Number of sheets for portion 2:\", portion_2)\n", + "print(\"Number of sheets for portion 3:\", portion_3)" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "4262\n" + ] + } + ], + "source": [ + "import os\n", + "#annotations\n", + "# images\n", + "# Directory containing the images\n", + "directory = '/workspaces/mmsegmentation/data/cag/annotations/training'\n", + "print(len(os.listdir('/workspaces/mmsegmentation/data/cag/annotations/training')))\n", + "# Iterate over the files in the directory\n", + "# for filename in os.listdir(directory):\n", + "# # Check if the filename is in the range to be deleted\n", + "# if filename.endswith('.png'):\n", + "# file_number = int(filename.split('.')[0])\n", + "# if file_number >= 2400:\n", + "# # Construct the full file path\n", + "# file_path = os.path.join(directory, filename)\n", + "# # Delete the file\n", + "# os.remove(file_path)\n", + " \n", + "# print(len(os.listdir('/workspaces/mmsegmentation-1/data/cag/annotations/training')))" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "base", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.8.12" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/configs/_base_/datasets/cag.py b/configs/_base_/datasets/cag.py new file mode 100644 index 0000000000..08e6785fdf --- /dev/null +++ b/configs/_base_/datasets/cag.py @@ -0,0 +1,95 @@ +# dataset settings +dataset_type = 'CoronaryAngiographyDataset' +data_root = 'data/cag_2100' +# augmentation setting from YoungIn's jupyter notebook +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(type='LoadAnnotations', reduce_zero_label=False), + dict( + type='AlbuShiftScaleRotateTransform', + scale_limit=(-0.2,0), + rotate_limit=20, + shift_limit=0.1, + border_mode=0, value=[0.3,0.4,0.5], + p=1 + ), + dict( + type='AlbuRandomContrastTransform', + brightness_limit=(-0.2, 0.2), + contrast_limit=(-0.2, 0.2), + p=0.5 + ), + dict( + type='AlbuGaussNoiseTransform', + var_limit=(0, 0.01), + p=0.5 + ), + dict(type='RandomFlip', prob=0.5), + # dict(type='PhotoMetricDistortion'), + dict(type='PackSegInputs') +] +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(type='Resize', scale=(512, 512), keep_ratio=True), + # add loading annotation after ``Resize`` because ground truth + # does not need to do resize data transform + dict(type='LoadAnnotations', reduce_zero_label=False), + dict(type='PackSegInputs') +] +img_ratios = [1.0] +tta_pipeline = [ + dict(type='LoadImageFromFile', backend_args=None), + dict( + type='TestTimeAug', + transforms=[ + [ + dict(type='Resize', scale_factor=r, keep_ratio=True) + for r in img_ratios + ], + [ + dict(type='RandomFlip', prob=0., direction='horizontal'), + dict(type='RandomFlip', prob=1., direction='horizontal') + ], [dict(type='LoadAnnotations')], [dict(type='PackSegInputs')] + ]) +] +train_dataloader = dict( + batch_size=4, + num_workers=4, + persistent_workers=True, + sampler=dict(type='InfiniteSampler', shuffle=True), + dataset=dict( + type=dataset_type, + data_root=data_root, + reduce_zero_label=False, + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + pipeline=train_pipeline)) +val_dataloader = dict( + batch_size=1, + num_workers=4, + persistent_workers=True, + sampler=dict(type='DefaultSampler', shuffle=False), + dataset=dict( + type=dataset_type, + data_root=data_root, + reduce_zero_label=False, + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + pipeline=test_pipeline)) +test_dataloader = dict( + batch_size=1, + num_workers=4, + persistent_workers=True, + sampler=dict(type='DefaultSampler', shuffle=False), + dataset=dict( + type=dataset_type, + data_root=data_root, + reduce_zero_label=False, + data_prefix=dict( + img_path='images/test', + seg_map_path='annotations/test'), + pipeline=test_pipeline)) + +val_evaluator = dict(type='IoUMetric', iou_metrics=['mIoU', 'mFscore']) +test_evaluator = dict(type='IoUMetric', iou_metrics=['mIoU', 'mFscore']) diff --git a/configs/_base_/schedules/schedule_160k.py b/configs/_base_/schedules/schedule_160k.py index 60d7bec762..1221375d8c 100644 --- a/configs/_base_/schedules/schedule_160k.py +++ b/configs/_base_/schedules/schedule_160k.py @@ -13,13 +13,13 @@ ] # training schedule for 160k train_cfg = dict( - type='IterBasedTrainLoop', max_iters=160000, val_interval=16000) + type='IterBasedTrainLoop', max_iters=160000, val_interval=10000) val_cfg = dict(type='ValLoop') test_cfg = dict(type='TestLoop') default_hooks = dict( timer=dict(type='IterTimerHook'), logger=dict(type='LoggerHook', interval=50, log_metric_by_epoch=False), param_scheduler=dict(type='ParamSchedulerHook'), - checkpoint=dict(type='CheckpointHook', by_epoch=False, interval=16000), + checkpoint=dict(type='CheckpointHook', by_epoch=False, interval=10000), sampler_seed=dict(type='DistSamplerSeedHook'), visualization=dict(type='SegVisualizationHook')) diff --git a/configs/mae/mae-base_upernet_8xb2-amp-160k_ade20k-512x512.py b/configs/mae/mae-base_upernet_8xb2-amp-160k_ade20k-512x512.py index b8eae174e9..7045dfc87e 100644 --- a/configs/mae/mae-base_upernet_8xb2-amp-160k_ade20k-512x512.py +++ b/configs/mae/mae-base_upernet_8xb2-amp-160k_ade20k-512x512.py @@ -1,12 +1,12 @@ _base_ = [ - '../_base_/models/upernet_mae.py', '../_base_/datasets/ade20k.py', + '../_base_/models/upernet_mae.py', '../_base_/datasets/cag.py', '../_base_/default_runtime.py', '../_base_/schedules/schedule_160k.py' ] crop_size = (512, 512) data_preprocessor = dict(size=crop_size) model = dict( data_preprocessor=data_preprocessor, - pretrained='./pretrain/mae_pretrain_vit_base_mmcls.pth', + pretrained='/workspaces/mmsegmentation-1/configs/mae/mae_pretrain_vit_base.pth', backbone=dict( type='MAE', img_size=(512, 512), @@ -51,4 +51,4 @@ # By default, models are trained on 8 GPUs with 2 images per GPU train_dataloader = dict(batch_size=2) val_dataloader = dict(batch_size=1) -test_dataloader = val_dataloader +test_dataloader = dict(batch_size=1) diff --git a/configs/mae/mae-base_upernet_8xb2-amp-160k_cag-512x512.py b/configs/mae/mae-base_upernet_8xb2-amp-160k_cag-512x512.py new file mode 100644 index 0000000000..0225c1e9f7 --- /dev/null +++ b/configs/mae/mae-base_upernet_8xb2-amp-160k_cag-512x512.py @@ -0,0 +1,54 @@ +_base_ = [ + '../_base_/models/upernet_mae.py', '../_base_/datasets/cag.py', + '../_base_/default_runtime.py', '../_base_/schedules/schedule_160k.py' +] +crop_size = (512, 512) +data_preprocessor = dict(size=crop_size) +model = dict( + data_preprocessor=data_preprocessor, + pretrained='/workspaces/mmsegmentation-1/converted_model.pth', + backbone=dict( + type='MAE', + img_size=(512, 512), + patch_size=16, + embed_dims=768, + num_layers=12, + num_heads=12, + mlp_ratio=4, + init_values=1.0, + drop_path_rate=0.1, + out_indices=[3, 5, 7, 11]), + neck=dict(embed_dim=768, rescales=[4, 2, 1, 0.5]), + decode_head=dict( + in_channels=[768, 768, 768, 768], num_classes=150, channels=768), + auxiliary_head=dict(in_channels=768, num_classes=150), + test_cfg=dict(mode='slide', crop_size=(512, 512), stride=(341, 341))) + +optim_wrapper = dict( + _delete_=True, + type='OptimWrapper', + optimizer=dict( + type='AdamW', lr=1e-4, betas=(0.9, 0.999), weight_decay=0.05), + paramwise_cfg=dict(num_layers=12, layer_decay_rate=0.65), + constructor='LayerDecayOptimizerConstructor') + +param_scheduler = [ + dict( + type='LinearLR', start_factor=1e-6, by_epoch=False, begin=0, end=1500), + dict( + type='PolyLR', + eta_min=0.0, + power=1.0, + begin=1500, + end=160000, + by_epoch=False, + ) +] + +# mixed precision +fp16 = dict(loss_scale='dynamic') + +# By default, models are trained on 8 GPUs with 2 images per GPU +train_dataloader = dict(batch_size=2) +val_dataloader = dict(batch_size=1) +test_dataloader = dict(batch_size=1) diff --git a/configs/upernet/upernet_r101_4xb4-160k_cag-512x512.py b/configs/upernet/upernet_r101_4xb4-160k_cag-512x512.py new file mode 100644 index 0000000000..1e80ddc436 --- /dev/null +++ b/configs/upernet/upernet_r101_4xb4-160k_cag-512x512.py @@ -0,0 +1,11 @@ +_base_ = [ + '../_base_/models/upernet_r50.py', '../_base_/datasets/cag.py', + '../_base_/default_runtime.py', '../_base_/schedules/schedule_160k.py' +] +crop_size = (512, 512) +data_preprocessor = dict(size=crop_size) +model = dict( + data_preprocessor=data_preprocessor, + decode_head=dict(num_classes=2), + auxiliary_head=dict(num_classes=2), + pretrained='open-mmlab://resnet101_v1c', backbone=dict(depth=101)) diff --git a/docker/Dockerfile b/docker/Dockerfile index 26420ddbbb..e0efc81172 100644 --- a/docker/Dockerfile +++ b/docker/Dockerfile @@ -33,3 +33,4 @@ WORKDIR /mmsegmentation ENV FORCE_CUDA="1" RUN pip install -r requirements.txt RUN pip install --no-cache-dir -e . +RUN pip install wandb diff --git a/feature_map_visual.py b/feature_map_visual.py new file mode 100644 index 0000000000..266ceee01c --- /dev/null +++ b/feature_map_visual.py @@ -0,0 +1,137 @@ +# Copyright (c) OpenMMLab. All rights reserved. +from argparse import ArgumentParser +from typing import Type + +import mmcv +import torch +import torch.nn as nn + +from mmengine.model import revert_sync_batchnorm +from mmengine.structures import PixelData +from mmseg.apis import inference_model, init_model +from mmseg.structures import SegDataSample +from mmseg.utils import register_all_modules +from mmseg.visualization import SegLocalVisualizer + + +class Recorder: + """record the forward output feature map and save to data_buffer.""" + + def __init__(self) -> None: + self.data_buffer = list() + + def __enter__(self, ): + self._data_buffer = list() + + def record_data_hook(self, model: nn.Module, input: Type, output: Type): + self.data_buffer.append(output) + + def __exit__(self, *args, **kwargs): + pass + + +def visualize(args, model, recorder, result): + seg_visualizer = SegLocalVisualizer( + vis_backends=[dict(type='WandbVisBackend')], + save_dir='temp_dir', + alpha=0.5) + seg_visualizer.dataset_meta = dict( + classes=model.dataset_meta['classes'], + palette=model.dataset_meta['palette']) + + image = mmcv.imread(args.img, 'gray') + + seg_visualizer.add_datasample( + name='predict', + image=image, + data_sample=result, + draw_gt=False, + draw_pred=True, + wait_time=0, + out_file=None, + show=False) + + # add feature map to wandb visualizer + for i in range(len(recorder.data_buffer)): + feature = recorder.data_buffer[i][0] # remove the batch + drawn_img = seg_visualizer.draw_featmap( + feature, image, channel_reduction='select_max') + seg_visualizer.add_image(f'feature_map{i}', drawn_img) + + if args.gt_mask: + sem_seg = mmcv.imread(args.gt_mask, 'unchanged') + sem_seg = torch.from_numpy(sem_seg) + gt_mask = dict(data=sem_seg) + gt_mask = PixelData(**gt_mask) + data_sample = SegDataSample() + data_sample.gt_sem_seg = gt_mask + + seg_visualizer.add_datasample( + name='gt_mask', + image=image, + data_sample=data_sample, + draw_gt=True, + draw_pred=False, + wait_time=0, + out_file=None, + show=False) + + seg_visualizer.add_image('image', image) + + +def main(): + parser = ArgumentParser( + description='Draw the Feature Map During Inference') + parser.add_argument('img', help='Image file') + parser.add_argument('config', help='Config file') + parser.add_argument('checkpoint', help='Checkpoint file') + parser.add_argument('--gt_mask', default=None, help='Path of gt mask file') + parser.add_argument('--out-file', default=None, help='Path to output file') + parser.add_argument( + '--device', default='cuda:0', help='Device used for inference') + parser.add_argument( + '--opacity', + type=float, + default=0.5, + help='Opacity of painted segmentation map. In (0, 1] range.') + parser.add_argument( + '--title', default='result', help='The image identifier.') + args = parser.parse_args() + + register_all_modules() + + # build the model from a config file and a checkpoint file + model = init_model(args.config, args.checkpoint, device=args.device) + if args.device == 'cpu': + model = revert_sync_batchnorm(model) + + # show all named module in the model and use it in source list below + for name, module in model.named_modules(): + print(name) + + source = [ + 'decode_head.fusion.stages.0.query_project.activate', + 'decode_head.context.stages.0.key_project.activate', + 'decode_head.context.bottleneck.activate' + ] + source = dict.fromkeys(source) + + count = 0 + recorder = Recorder() + # registry the forward hook + for name, module in model.named_modules(): + if name in source: + count += 1 + module.register_forward_hook(recorder.record_data_hook) + if count == len(source): + break + + with recorder: + # test a single image, and record feature map to data_buffer + result = inference_model(model, args.img) + + visualize(args, model, recorder, result) + + +if __name__ == '__main__': + main() \ No newline at end of file diff --git a/mmseg/datasets/__init__.py b/mmseg/datasets/__init__.py index a2bdb63d01..2cb0ec0694 100644 --- a/mmseg/datasets/__init__.py +++ b/mmseg/datasets/__init__.py @@ -1,6 +1,7 @@ # Copyright (c) OpenMMLab. All rights reserved. # yapf: disable from .ade import ADE20KDataset +from .cag import CoronaryAngiographyDataset from .basesegdataset import BaseCDDataset, BaseSegDataset from .bdd100k import BDD100KDataset from .chase_db1 import ChaseDB1Dataset @@ -37,7 +38,7 @@ PhotoMetricDistortion, RandomCrop, RandomCutOut, RandomMosaic, RandomRotate, RandomRotFlip, Rerange, ResizeShortestEdge, ResizeToMultiple, RGB2Gray, - SegRescale) + SegRescale, AlbuShiftScaleRotateTransform, AlbuRandomContrastTransform, AlbuGaussNoiseTransform) from .voc import PascalVOCDataset # yapf: enable @@ -58,7 +59,7 @@ 'BioMedicalRandomGamma', 'BioMedical3DPad', 'RandomRotFlip', 'SynapseDataset', 'REFUGEDataset', 'MapillaryDataset_v1', 'MapillaryDataset_v2', 'Albu', 'LEVIRCDDataset', - 'LoadMultipleRSImageFromFile', 'LoadSingleRSImageFromFile', + 'LoadMultipleRSImageFromFile', 'CoronaryAngiographyDataset', 'LoadSingleRSImageFromFile', 'ConcatCDInput', 'BaseCDDataset', 'DSDLSegDataset', 'BDD100KDataset', 'NYUDataset' ] diff --git a/mmseg/datasets/cag.py b/mmseg/datasets/cag.py new file mode 100644 index 0000000000..7eb7ba43bb --- /dev/null +++ b/mmseg/datasets/cag.py @@ -0,0 +1,24 @@ +# Copyright (c) OpenMMLab. All rights reserved. +from mmseg.registry import DATASETS +from .basesegdataset import BaseSegDataset + + +@DATASETS.register_module() +class CoronaryAngiographyDataset(BaseSegDataset): + """Angiography dataset. + """ + METAINFO = dict( + classes=('background', 'contrast'), + palette=[[0, 0, 0], [255, 0, 0]]) + + def __init__(self, + img_suffix='.png', + seg_map_suffix='.png', + reduce_zero_label=False, + + **kwargs) -> None: + super().__init__( + img_suffix=img_suffix, + seg_map_suffix=seg_map_suffix, + reduce_zero_label=reduce_zero_label, + **kwargs) diff --git a/mmseg/datasets/transforms/__init__.py b/mmseg/datasets/transforms/__init__.py index 125f070818..9fb64c78ac 100644 --- a/mmseg/datasets/transforms/__init__.py +++ b/mmseg/datasets/transforms/__init__.py @@ -13,7 +13,7 @@ RandomDepthMix, RandomFlip, RandomMosaic, RandomRotate, RandomRotFlip, Rerange, Resize, ResizeShortestEdge, ResizeToMultiple, RGB2Gray, - SegRescale) + SegRescale, AlbuGaussNoiseTransform, AlbuRandomContrastTransform, AlbuShiftScaleRotateTransform) # yapf: enable __all__ = [ @@ -26,5 +26,5 @@ 'BioMedical3DRandomFlip', 'BioMedicalRandomGamma', 'BioMedical3DPad', 'RandomRotFlip', 'Albu', 'LoadSingleRSImageFromFile', 'ConcatCDInput', 'LoadMultipleRSImageFromFile', 'LoadDepthAnnotation', 'RandomDepthMix', - 'RandomFlip', 'Resize' + 'RandomFlip', 'Resize', 'AlbuGaussNoiseTransform', 'AlbuRandomContrastTransform', 'AlbuShiftScaleRotateTransform' ] diff --git a/mmseg/datasets/transforms/formatting.py b/mmseg/datasets/transforms/formatting.py index bd250551e9..83f566d03c 100644 --- a/mmseg/datasets/transforms/formatting.py +++ b/mmseg/datasets/transforms/formatting.py @@ -64,14 +64,13 @@ def transform(self, results: dict) -> dict: if 'img' in results: img = results['img'] if len(img.shape) < 3: - img = np.expand_dims(img, -1) + img = np.stack([img] * 3, axis=-1) if not img.flags.c_contiguous: img = to_tensor(np.ascontiguousarray(img.transpose(2, 0, 1))) else: img = img.transpose(2, 0, 1) img = to_tensor(img).contiguous() packed_results['inputs'] = img - data_sample = SegDataSample() if 'gt_seg_map' in results: if len(results['gt_seg_map'].shape) == 2: diff --git a/mmseg/datasets/transforms/transforms.py b/mmseg/datasets/transforms/transforms.py index 082ae5b440..916186c3af 100644 --- a/mmseg/datasets/transforms/transforms.py +++ b/mmseg/datasets/transforms/transforms.py @@ -15,6 +15,7 @@ from mmengine.utils import is_tuple_of from numpy import random from scipy.ndimage import gaussian_filter +import albumentations as album from mmseg.datasets.dataset_wrappers import MultiImageMixDataset from mmseg.registry import TRANSFORMS @@ -670,40 +671,40 @@ def contrast(self, img: np.ndarray) -> np.ndarray: alpha=random.uniform(self.contrast_lower, self.contrast_upper)) return img - def saturation(self, img: np.ndarray) -> np.ndarray: - """Saturation distortion. - - Args: - img (np.ndarray): The input image. - Returns: - np.ndarray: Image after saturation change. - """ - - if random.randint(2): - img = mmcv.bgr2hsv(img) - img[:, :, 1] = self.convert( - img[:, :, 1], - alpha=random.uniform(self.saturation_lower, - self.saturation_upper)) - img = mmcv.hsv2bgr(img) - return img - - def hue(self, img: np.ndarray) -> np.ndarray: - """Hue distortion. - - Args: - img (np.ndarray): The input image. - Returns: - np.ndarray: Image after hue change. - """ - - if random.randint(2): - img = mmcv.bgr2hsv(img) - img[:, :, - 0] = (img[:, :, 0].astype(int) + - random.randint(-self.hue_delta, self.hue_delta)) % 180 - img = mmcv.hsv2bgr(img) - return img + # def saturation(self, img: np.ndarray) -> np.ndarray: + # """Saturation distortion. + + # Args: + # img (np.ndarray): The input image. + # Returns: + # np.ndarray: Image after saturation change. + # """ + + # if random.randint(2): + # img = mmcv.bgr2hsv(img) + # img[:, :, 1] = self.convert( + # img[:, :, 1], + # alpha=random.uniform(self.saturation_lower, + # self.saturation_upper)) + # img = mmcv.hsv2bgr(img) + # return img + + # def hue(self, img: np.ndarray) -> np.ndarray: + # """Hue distortion. + + # Args: + # img (np.ndarray): The input image. + # Returns: + # np.ndarray: Image after hue change. + # """ + + # if random.randint(2): + # img = mmcv.bgr2hsv(img) + # img[:, :, + # 0] = (img[:, :, 0].astype(int) + + # random.randint(-self.hue_delta, self.hue_delta)) % 180 + # img = mmcv.hsv2bgr(img) + # return img def transform(self, results: dict) -> dict: """Transform function to perform photometric distortion on images. @@ -2512,3 +2513,159 @@ def transform(self, results: dict) -> dict: results['img'] = img return results + + +class BaseAlbuTransform(BaseTransform): + + def __repr__(self): + repr_str = self.__class__.__name__ + num_params = len(self.params) + for i, (k, v) in enumerate(self.params.items()): + if i == 0: + repr_str += f'({k}={v}, ' + elif 0 < i < num_params - 1: + repr_str += f'{k}={v}, ' + elif i == num_params - 1: + repr_str += f'{k}={v})' + return repr_str + + +@TRANSFORMS.register_module() +class AlbuShiftScaleRotateTransform(BaseAlbuTransform): + """ + Wrapper for album.ShiftScaleRotate + + Required Keys: + + - img + - gt_seg_map + + Modified Keys: + + - img + - gt_seg_map + + Args: + shift_limit ((float, float) or float): shift factor range for both height and width. If shift_limit + is a single float value, the range will be (-shift_limit, shift_limit). Absolute values for lower and + upper bounds should lie in range [0, 1]. Default: (-0.0625, 0.0625). + scale_limit ((float, float) or float): scaling factor range. If scale_limit is a single float value, the + range will be (-scale_limit, scale_limit). Note that the scale_limit will be biased by 1. + If scale_limit is a tuple, like (low, high), sampling will be done from the range (1 + low, 1 + high). + Default: (-0.1, 0.1). + rotate_limit ((int, int) or int): rotation range. If rotate_limit is a single int value, the + range will be (-rotate_limit, rotate_limit). Default: (-45, 45). + interpolation (OpenCV flag): flag that is used to specify the interpolation algorithm. Should be one of: + cv2.INTER_NEAREST, cv2.INTER_LINEAR, cv2.INTER_CUBIC, cv2.INTER_AREA, cv2.INTER_LANCZOS4. + Default: cv2.INTER_LINEAR. + border_mode (OpenCV flag): flag that is used to specify the pixel extrapolation method. Should be one of: + cv2.BORDER_CONSTANT, cv2.BORDER_REPLICATE, cv2.BORDER_REFLECT, cv2.BORDER_WRAP, cv2.BORDER_REFLECT_101. + Default: cv2.BORDER_REFLECT_101 + value (int, float, list of int, list of float): padding value if border_mode is cv2.BORDER_CONSTANT. + mask_value (int, float, + list of int, + list of float): padding value if border_mode is cv2.BORDER_CONSTANT applied for masks. + shift_limit_x ((float, float) or float): shift factor range for width. If it is set then this value + instead of shift_limit will be used for shifting width. If shift_limit_x is a single float value, + the range will be (-shift_limit_x, shift_limit_x). Absolute values for lower and upper bounds should lie in + the range [0, 1]. Default: None. + shift_limit_y ((float, float) or float): shift factor range for height. If it is set then this value + instead of shift_limit will be used for shifting height. If shift_limit_y is a single float value, + the range will be (-shift_limit_y, shift_limit_y). Absolute values for lower and upper bounds should lie + in the range [0, 1]. Default: None. + rotate_method (str): rotation method used for the bounding boxes. Should be one of "largest_box" or "ellipse". + Default: "largest_box" + p (float): probability of applying the transform. Default: 0.5. + """ + def __init__(self, **kwargs): + self.params = kwargs + self.albu_transform = album.ShiftScaleRotate(**kwargs) + + def transform(self, results: dict) -> dict: + """Transform function to perform `album.ShiftScaleRotate`. + + Args: + results (dict): Result dict from loading pipeline. + + Returns: + dict: Result dict with images distorted. + """ + aug = self.albu_transform(image=results['img'], mask=results['gt_seg_map']) + results['img'] = aug['image'] + results['gt_seg_map'] = aug['mask'] + return results + + +@TRANSFORMS.register_module() +class AlbuRandomContrastTransform(BaseAlbuTransform): + """ + Wrapper for album.RandomContrast + + Required Keys: + + - img + + Modified Keys: + + - img + + Args: + limit ((float, float) or float): factor range for changing contrast. + If limit is a single float, the range will be (-limit, limit). Default: (-0.2, 0.2). + p (float): probability of applying the transform. Default: 0.5. + """ + def __init__(self, **kwargs): + self.params = kwargs + self.albu_transform = album.RandomBrightnessContrast(**kwargs) + + def transform(self, results: dict) -> dict: + """Transform function to perform `album.ShiftScaleRotate`. + + Args: + results (dict): Result dict from loading pipeline. + + Returns: + dict: Result dict with images distorted. + """ + aug = self.albu_transform(image=results['img']) + results['img'] = aug['image'] + return results + + +@TRANSFORMS.register_module() +class AlbuGaussNoiseTransform(BaseAlbuTransform): + """ + Wrapper for album.GaussNoise + + Required Keys: + + - img + + Modified Keys: + + - img + + Args: + var_limit ((float, float) or float): variance range for noise. If var_limit is a single float, the range + will be (0, var_limit). Default: (10.0, 50.0). + mean (float): mean of the noise. Default: 0 + per_channel (bool): if set to True, noise will be sampled for each channel independently. + Otherwise, the noise will be sampled once for all channels. Default: True + p (float): probability of applying the transform. Default: 0.5. + """ + def __init__(self, **kwargs): + self.params = kwargs + self.albu_transform = album.GaussNoise(**kwargs) + + def transform(self, results: dict) -> dict: + """Transform function to perform `album.ShiftScaleRotate`. + + Args: + results (dict): Result dict from loading pipeline. + + Returns: + dict: Result dict with images distorted. + """ + aug = self.albu_transform(image=results['img']) + results['img'] = aug['image'] + return results diff --git a/mmseg/utils/class_names.py b/mmseg/utils/class_names.py index 122e63fcc4..e93555e12d 100644 --- a/mmseg/utils/class_names.py +++ b/mmseg/utils/class_names.py @@ -1,6 +1,14 @@ # Copyright (c) OpenMMLab. All rights reserved. from mmengine.utils import is_str +def cag_classes(): + return [ + 'background', 'contrast' + ] + +def cag_palette(): + return [[0, 0, 0], [255, 0, 0]] + def cityscapes_classes(): """Cityscapes class names for external use.""" @@ -440,6 +448,7 @@ def bdd100k_palette(): dataset_aliases = { + 'cag' : ['cag'], 'cityscapes': ['cityscapes'], 'ade': ['ade', 'ade20k'], 'voc': ['voc', 'pascal_voc', 'voc12', 'voc12aug'], diff --git a/nohup.out b/nohup.out new file mode 100644 index 0000000000..c2a6817ea9 --- /dev/null +++ b/nohup.out @@ -0,0 +1,43300 @@ +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/16 08:10:01 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 1275060518 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 1275060518 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/16 08:10:02 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +fp16 = dict(loss_scale='dynamic') +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = None +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=768, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + num_convs=1, + type='FCNHead'), + backbone=dict( + act_cfg=dict(type='GELU'), + attn_drop_rate=0.0, + drop_path_rate=0.1, + embed_dims=768, + img_size=( + 512, + 512, + ), + in_channels=3, + init_values=1.0, + mlp_ratio=4, + norm_cfg=dict(eps=1e-06, type='LN'), + norm_eval=False, + num_heads=12, + num_layers=12, + out_indices=[ + 3, + 5, + 7, + 11, + ], + patch_size=16, + type='MAE'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=768, + dropout_ratio=0.1, + in_channels=[ + 768, + 768, + 768, + 768, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + neck=dict( + embed_dim=768, rescales=[ + 4, + 2, + 1, + 0.5, + ], type='Feature2Pyramid'), + pretrained='/workspaces/mmsegmentation-1/converted_model.pth', + test_cfg=dict(crop_size=( + 512, + 512, + ), mode='slide', stride=( + 341, + 341, + )), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + constructor='LayerDecayOptimizerConstructor', + optimizer=dict( + betas=( + 0.9, + 0.999, + ), lr=0.0001, type='AdamW', weight_decay=0.05), + paramwise_cfg=dict(layer_decay_rate=0.65, num_layers=12), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, by_epoch=False, end=1500, start_factor=1e-06, + type='LinearLR'), + dict( + begin=1500, + by_epoch=False, + end=160000, + eta_min=0.0, + power=1.0, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=2, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = './work_dirs/mae-base_upernet_8xb2-amp-160k_cag-512x512' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/16 08:10:04 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' + +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65}set param backbone.layers.3.ffn.layers.1.bias as id 4 + +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14set param backbone.layers.4.gamma_1 as id 5 + +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.cls_token as id 0 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.pos_embed as id 0set param backbone.layers.4.ffn.layers.0.0.bias as id 5 + +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.0.gamma_1 as id 1set param backbone.layers.5.attn.qkv.weight as id 6 + +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6set param backbone.layers.0.ln1.bias as id 1 + +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6set param backbone.layers.0.attn.relative_position_bias_table as id 1 + +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.6.gamma_1 as id 7set param backbone.layers.0.attn.qkv.bias as id 1 + +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7set param backbone.layers.0.ln2.weight as id 1 + +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7set param backbone.layers.0.ffn.layers.0.0.weight as id 1 + +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.1.attn.relative_position_bias_table as id 2set param backbone.layers.6.ffn.layers.0.0.bias as id 7 + +set param backbone.layers.1.attn.qkv.weight as id 2set param backbone.layers.6.ffn.layers.1.weight as id 7 + +set param backbone.layers.6.ffn.layers.1.bias as id 7set param backbone.layers.1.attn.qkv.bias as id 2 + +set param backbone.layers.7.gamma_1 as id 8set param backbone.layers.1.attn.proj.weight as id 2 + +set param backbone.layers.7.gamma_2 as id 8set param backbone.layers.1.attn.proj.bias as id 2 + +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.1.ln2.weight as id 2set param backbone.layers.7.ln1.bias as id 8 + +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2set param backbone.layers.7.attn.qkv.bias as id 8 + +set param backbone.layers.1.ffn.layers.0.0.bias as id 2set param backbone.layers.7.attn.proj.weight as id 8 + +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8set param backbone.layers.2.gamma_1 as id 3 + +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.2.ln1.weight as id 3set param backbone.layers.7.ffn.layers.1.bias as id 8 + +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.2.attn.relative_position_bias_table as id 3set param backbone.layers.8.gamma_2 as id 9 + +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9set param backbone.layers.2.attn.proj.weight as id 3 + +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.2.ln2.weight as id 3set param backbone.layers.8.attn.proj.bias as id 9 + +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.2.ffn.layers.1.weight as id 3set param backbone.layers.8.ffn.layers.1.weight as id 9 + +set param backbone.layers.8.ffn.layers.1.bias as id 9set param backbone.layers.2.ffn.layers.1.bias as id 3 + +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.3.gamma_2 as id 4set param backbone.layers.9.ln1.weight as id 10 + +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.3.attn.qkv.weight as id 4set param backbone.layers.9.attn.proj.weight as id 10 + +set param backbone.layers.9.attn.proj.bias as id 10set param backbone.layers.3.attn.qkv.bias as id 4 + +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11set param backbone.layers.4.gamma_1 as id 5 + +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11set param backbone.layers.4.ln1.weight as id 5 + +set param backbone.layers.4.ln1.bias as id 5set param backbone.layers.10.ln2.weight as id 11 + +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11set param backbone.layers.4.attn.qkv.weight as id 5 + +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11set param backbone.layers.4.attn.proj.weight as id 5 + +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12set param backbone.layers.4.ln2.weight as id 5 + +set param backbone.layers.11.ln1.weight as id 12set param backbone.layers.4.ln2.bias as id 5 + +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12set param backbone.layers.4.ffn.layers.0.0.bias as id 5 + +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.4.ffn.layers.1.weight as id 5set param backbone.layers.11.attn.proj.bias as id 12 + +set param backbone.layers.4.ffn.layers.1.bias as id 5set param backbone.layers.11.ln2.weight as id 12 + +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.5.gamma_2 as id 6set param backbone.layers.11.ffn.layers.0.0.bias as id 12 + +set param backbone.layers.11.ffn.layers.1.weight as id 12set param backbone.layers.5.ln1.weight as id 6 + +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6set param neck.upsample_4x.0.weight as id 13 + +set param backbone.layers.5.attn.qkv.bias as id 6set param neck.upsample_4x.0.bias as id 13 + +set param backbone.layers.5.attn.proj.weight as id 6set param neck.upsample_4x.1.weight as id 13 + +set param neck.upsample_4x.1.bias as id 13set param backbone.layers.5.attn.proj.bias as id 6 + +set param neck.upsample_4x.3.weight as id 13 +set param backbone.layers.5.ln2.weight as id 6 +set param neck.upsample_4x.3.bias as id 13 +set param backbone.layers.5.ln2.bias as id 6 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6set param decode_head.conv_seg.weight as id 13 + +set param decode_head.conv_seg.bias as id 13set param backbone.layers.5.ffn.layers.1.bias as id 6 + +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7set param decode_head.psp_modules.0.1.conv.weight as id 13 + +set param backbone.layers.6.ln1.weight as id 7set param decode_head.psp_modules.0.1.bn.weight as id 13 + +set param decode_head.psp_modules.0.1.bn.bias as id 13set param backbone.layers.6.ln1.bias as id 7 + +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param backbone.layers.6.attn.qkv.weight as id 7set param decode_head.lateral_convs.0.bn.bias as id 13 + +set param backbone.layers.6.attn.qkv.bias as id 7 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param backbone.layers.6.attn.proj.weight as id 7set param decode_head.lateral_convs.1.bn.bias as id 13 + +set param backbone.layers.6.attn.proj.bias as id 7 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13set param backbone.layers.6.ln2.weight as id 7 + +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param backbone.layers.6.ln2.bias as id 7 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param backbone.layers.7.gamma_1 as id 8set param decode_head.fpn_convs.2.bn.bias as id 13 + +set param backbone.layers.7.gamma_2 as id 8set param decode_head.fpn_bottleneck.conv.weight as id 13 + +set param decode_head.fpn_bottleneck.bn.weight as id 13set param backbone.layers.7.ln1.weight as id 8 + +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8set param auxiliary_head.conv_seg.weight as id 13 + +set param auxiliary_head.conv_seg.bias as id 13set param backbone.layers.7.attn.qkv.bias as id 8 + +set param auxiliary_head.convs.0.conv.weight as id 13set param backbone.layers.7.attn.proj.weight as id 8 + +set param auxiliary_head.convs.0.bn.weight as id 13 +set param backbone.layers.7.attn.proj.bias as id 8 +set param auxiliary_head.convs.0.bn.bias as id 13 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +Param groups = { + "layer_0_decay": { + "param_names": [ + "backbone.cls_token", + "backbone.pos_embed", + "backbone.patch_embed.projection.weight" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.05 + }, + "layer_0_no_decay": { + "param_names": [ + "backbone.patch_embed.projection.bias" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.0 + }, + "layer_1_no_decay": { + "param_names": [ + "backbone.layers.0.gamma_1", + "backbone.layers.0.gamma_2", + "backbone.layers.0.ln1.weight", + "backbone.layers.0.ln1.bias", + "backbone.layers.0.attn.qkv.bias", + "backbone.layers.0.attn.proj.bias", + "backbone.layers.0.ln2.weight", + "backbone.layers.0.ln2.bias", + "backbone.layers.0.ffn.layers.0.0.bias", + "backbone.layers.0.ffn.layers.1.bias" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.0 + }, + "layer_1_decay": { + "param_names": [ + "backbone.layers.0.attn.relative_position_bias_table", + "backbone.layers.0.attn.qkv.weight", + "backbone.layers.0.attn.proj.weight", + "backbone.layers.0.ffn.layers.0.0.weight", + "backbone.layers.0.ffn.layers.1.weight" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.05 + }, + "layer_2_no_decay": { + "param_names": [ + "backbone.layers.1.gamma_1", + "backbone.layers.1.gamma_2", + "backbone.layers.1.ln1.weight", + "backbone.layers.1.ln1.bias", + "backbone.layers.1.attn.qkv.bias", + "backbone.layers.1.attn.proj.bias", + "backbone.layers.1.ln2.weight", + "backbone.layers.1.ln2.bias", + "backbone.layers.1.ffn.layers.0.0.bias", + "backbone.layers.1.ffn.layers.1.bias" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.0 + }, + "layer_2_decay": { + "param_names": [ + "backbone.layers.1.attn.relative_position_bias_table", + "backbone.layers.1.attn.qkv.weight", + "backbone.layers.1.attn.proj.weight", + "backbone.layers.1.ffn.layers.0.0.weight", + "backbone.layers.1.ffn.layers.1.weight" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.05 + }, + "layer_3_no_decay": { + "param_names": [ + "backbone.layers.2.gamma_1", + "backbone.layers.2.gamma_2", + "backbone.layers.2.ln1.weight", + "backbone.layers.2.ln1.bias", + "backbone.layers.2.attn.qkv.bias", + "backbone.layers.2.attn.proj.bias", + "backbone.layers.2.ln2.weight", + "backbone.layers.2.ln2.bias", + "backbone.layers.2.ffn.layers.0.0.bias", + "backbone.layers.2.ffn.layers.1.bias" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.0 + }, + "layer_3_decay": { + "param_names": [ + "backbone.layers.2.attn.relative_position_bias_table", + "backbone.layers.2.attn.qkv.weight", + "backbone.layers.2.attn.proj.weight", + "backbone.layers.2.ffn.layers.0.0.weight", + "backbone.layers.2.ffn.layers.1.weight" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.05 + }, + "layer_4_no_decay": { + "param_names": [ + "backbone.layers.3.gamma_1", + "backbone.layers.3.gamma_2", + "backbone.layers.3.ln1.weight", + "backbone.layers.3.ln1.bias", + "backbone.layers.3.attn.qkv.bias", + "backbone.layers.3.attn.proj.bias", + "backbone.layers.3.ln2.weight", + "backbone.layers.3.ln2.bias", + "backbone.layers.3.ffn.layers.0.0.bias", + "backbone.layers.3.ffn.layers.1.bias" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.0 + }, + "layer_4_decay": { + "param_names": [ + "backbone.layers.3.attn.relative_position_bias_table", + "backbone.layers.3.attn.qkv.weight", + "backbone.layers.3.attn.proj.weight", + "backbone.layers.3.ffn.layers.0.0.weight", + "backbone.layers.3.ffn.layers.1.weight" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.05 + }, + "layer_5_no_decay": { + "param_names": [ + "backbone.layers.4.gamma_1", + "backbone.layers.4.gamma_2", + "backbone.layers.4.ln1.weight", + "backbone.layers.4.ln1.bias", + "backbone.layers.4.attn.qkv.bias", + "backbone.layers.4.attn.proj.bias", + "backbone.layers.4.ln2.weight", + "backbone.layers.4.ln2.bias", + "backbone.layers.4.ffn.layers.0.0.bias", + "backbone.layers.4.ffn.layers.1.bias" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.0 + }, + "layer_5_decay": { + "param_names": [ + "backbone.layers.4.attn.relative_position_bias_table", + "backbone.layers.4.attn.qkv.weight", + "backbone.layers.4.attn.proj.weight", + "backbone.layers.4.ffn.layers.0.0.weight", + "backbone.layers.4.ffn.layers.1.weight" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.05 + }, + "layer_6_no_decay": { + "param_names": [ + "backbone.layers.5.gamma_1", + "backbone.layers.5.gamma_2", + "backbone.layers.5.ln1.weight", + "backbone.layers.5.ln1.bias", + "backbone.layers.5.attn.qkv.bias", + "backbone.layers.5.attn.proj.bias", + "backbone.layers.5.ln2.weight", + "backbone.layers.5.ln2.bias", + "backbone.layers.5.ffn.layers.0.0.bias", + "backbone.layers.5.ffn.layers.1.bias" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.0 + }, + "layer_6_decay": { + "param_names": [ + "backbone.layers.5.attn.relative_position_bias_table", + "backbone.layers.5.attn.qkv.weight", + "backbone.layers.5.attn.proj.weight", + "backbone.layers.5.ffn.layers.0.0.weight", + "backbone.layers.5.ffn.layers.1.weight" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.05 + }, + "layer_7_no_decay": { + "param_names": [ + "backbone.layers.6.gamma_1", + "backbone.layers.6.gamma_2", + "backbone.layers.6.ln1.weight", + "backbone.layers.6.ln1.bias", + "backbone.layers.6.attn.qkv.bias", + "backbone.layers.6.attn.proj.bias", + "backbone.layers.6.ln2.weight", + "backbone.layers.6.ln2.bias", + "backbone.layers.6.ffn.layers.0.0.bias", + "backbone.layers.6.ffn.layers.1.bias" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.0 + }, + "layer_7_decay": { + "param_names": [ + "backbone.layers.6.attn.relative_position_bias_table", + "backbone.layers.6.attn.qkv.weight", + "backbone.layers.6.attn.proj.weight", + "backbone.layers.6.ffn.layers.0.0.weight", + "backbone.layers.6.ffn.layers.1.weight" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.05 + }, + "layer_8_no_decay": { + "param_names": [ + "backbone.layers.7.gamma_1", + "backbone.layers.7.gamma_2", + "backbone.layers.7.ln1.weight", + "backbone.layers.7.ln1.bias", + "backbone.layers.7.attn.qkv.bias", + "backbone.layers.7.attn.proj.bias", + "backbone.layers.7.ln2.weight", + "backbone.layers.7.ln2.bias", + "backbone.layers.7.ffn.layers.0.0.bias", + "backbone.layers.7.ffn.layers.1.bias" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.0 + }, + "layer_8_decay": { + "param_names": [ + "backbone.layers.7.attn.relative_position_bias_table", + "backbone.layers.7.attn.qkv.weight", + "backbone.layers.7.attn.proj.weight", + "backbone.layers.7.ffn.layers.0.0.weight", + "backbone.layers.7.ffn.layers.1.weight" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.05 + }, + "layer_9_no_decay": { + "param_names": [ + "backbone.layers.8.gamma_1", + "backbone.layers.8.gamma_2", + "backbone.layers.8.ln1.weight", + "backbone.layers.8.ln1.bias", + "backbone.layers.8.attn.qkv.bias", + "backbone.layers.8.attn.proj.bias", + "backbone.layers.8.ln2.weight", + "backbone.layers.8.ln2.bias", + "backbone.layers.8.ffn.layers.0.0.bias", + "backbone.layers.8.ffn.layers.1.bias" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.0 + }, + "layer_9_decay": { + "param_names": [ + "backbone.layers.8.attn.relative_position_bias_table", + "backbone.layers.8.attn.qkv.weight", + "backbone.layers.8.attn.proj.weight", + "backbone.layers.8.ffn.layers.0.0.weight", + "backbone.layers.8.ffn.layers.1.weight" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.05 + }, + "layer_10_no_decay": { + "param_names": [ + "backbone.layers.9.gamma_1", + "backbone.layers.9.gamma_2", + "backbone.layers.9.ln1.weight", + "backbone.layers.9.ln1.bias", + "backbone.layers.9.attn.qkv.bias", + "backbone.layers.9.attn.proj.bias", + "backbone.layers.9.ln2.weight", + "backbone.layers.9.ln2.bias", + "backbone.layers.9.ffn.layers.0.0.bias", + "backbone.layers.9.ffn.layers.1.bias" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.0 + }, + "layer_10_decay": { + "param_names": [ + "backbone.layers.9.attn.relative_position_bias_table", + "backbone.layers.9.attn.qkv.weight", + "backbone.layers.9.attn.proj.weight", + "backbone.layers.9.ffn.layers.0.0.weight", + "backbone.layers.9.ffn.layers.1.weight" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.05 + }, + "layer_11_no_decay": { + "param_names": [ + "backbone.layers.10.gamma_1", + "backbone.layers.10.gamma_2", + "backbone.layers.10.ln1.weight", + "backbone.layers.10.ln1.bias", + "backbone.layers.10.attn.qkv.bias", + "backbone.layers.10.attn.proj.bias", + "backbone.layers.10.ln2.weight", + "backbone.layers.10.ln2.bias", + "backbone.layers.10.ffn.layers.0.0.bias", + "backbone.layers.10.ffn.layers.1.bias" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.0 + }, + "layer_11_decay": { + "param_names": [ + "backbone.layers.10.attn.relative_position_bias_table", + "backbone.layers.10.attn.qkv.weight", + "backbone.layers.10.attn.proj.weight", + "backbone.layers.10.ffn.layers.0.0.weight", + "backbone.layers.10.ffn.layers.1.weight" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.05 + }, + "layer_12_no_decay": { + "param_names": [ + "backbone.layers.11.gamma_1", + "backbone.layers.11.gamma_2", + "backbone.layers.11.ln1.weight", + "backbone.layers.11.ln1.bias", + "backbone.layers.11.attn.qkv.bias", + "backbone.layers.11.attn.proj.bias", + "backbone.layers.11.ln2.weight", + "backbone.layers.11.ln2.bias", + "backbone.layers.11.ffn.layers.0.0.bias", + "backbone.layers.11.ffn.layers.1.bias" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.0 + }, + "layer_12_decay": { + "param_names": [ + "backbone.layers.11.attn.relative_position_bias_table", + "backbone.layers.11.attn.qkv.weight", + "backbone.layers.11.attn.proj.weight", + "backbone.layers.11.ffn.layers.0.0.weight", + "backbone.layers.11.ffn.layers.1.weight" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.05 + }, + "layer_13_decay": { + "param_names": [ + "neck.upsample_4x.0.weight", + "neck.upsample_4x.3.weight", + "neck.upsample_2x.0.weight", + "decode_head.conv_seg.weight", + "decode_head.psp_modules.0.1.conv.weight", + "decode_head.psp_modules.1.1.conv.weight", + "decode_head.psp_modules.2.1.conv.weight", + "decode_head.psp_modules.3.1.conv.weight", + "decode_head.bottleneck.conv.weight", + "decode_head.lateral_convs.0.conv.weight", + "decode_head.lateral_convs.1.conv.weight", + "decode_head.lateral_convs.2.conv.weight", + "decode_head.fpn_convs.0.conv.weight", + "decode_head.fpn_convs.1.conv.weight", + "decode_head.fpn_convs.2.conv.weight", + "decode_head.fpn_bottleneck.conv.weight", + "auxiliary_head.conv_seg.weight", + "auxiliary_head.convs.0.conv.weight" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.05 + }, + "layer_13_no_decay": { + "param_names": [ + "neck.upsample_4x.0.bias", + "neck.upsample_4x.1.weight", + "neck.upsample_4x.1.bias", + "neck.upsample_4x.3.bias", + "neck.upsample_2x.0.bias", + "decode_head.conv_seg.bias", + "decode_head.psp_modules.0.1.bn.weight", + "decode_head.psp_modules.0.1.bn.bias", + "decode_head.psp_modules.1.1.bn.weight", + "decode_head.psp_modules.1.1.bn.bias", + "decode_head.psp_modules.2.1.bn.weight", + "decode_head.psp_modules.2.1.bn.bias", + "decode_head.psp_modules.3.1.bn.weight", + "decode_head.psp_modules.3.1.bn.bias", + "decode_head.bottleneck.bn.weight", + "decode_head.bottleneck.bn.bias", + "decode_head.lateral_convs.0.bn.weight", + "decode_head.lateral_convs.0.bn.bias", + "decode_head.lateral_convs.1.bn.weight", + "decode_head.lateral_convs.1.bn.bias", + "decode_head.lateral_convs.2.bn.weight", + "decode_head.lateral_convs.2.bn.bias", + "decode_head.fpn_convs.0.bn.weight", + "decode_head.fpn_convs.0.bn.bias", + "decode_head.fpn_convs.1.bn.weight", + "decode_head.fpn_convs.1.bn.bias", + "decode_head.fpn_convs.2.bn.weight", + "decode_head.fpn_convs.2.bn.bias", + "decode_head.fpn_bottleneck.bn.weight", + "decode_head.fpn_bottleneck.bn.bias", + "auxiliary_head.conv_seg.bias", + "auxiliary_head.convs.0.bn.weight", + "auxiliary_head.convs.0.bn.bias" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.0 + } +} +04/16 08:10:05 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +/opt/conda/lib/python3.8/site-packages/mmengine/model/weight_init.py:649: UserWarning: Specified kernel cache directory could not be created! This disables kernel caching. Specified directory is /root/.cache/torch/kernels. This warning will appear only once per process. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/cuda/jit_utils.cpp:860.) + tensor.erfinv_() +/opt/conda/lib/python3.8/site-packages/mmengine/model/weight_init.py:649: UserWarning: Specified kernel cache directory could not be created! This disables kernel caching. Specified directory is /root/.cache/torch/kernels. This warning will appear only once per process. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/cuda/jit_utils.cpp:860.) + tensor.erfinv_() +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +/opt/conda/lib/python3.8/site-packages/mmengine/model/weight_init.py:649: UserWarning: Specified kernel cache directory could not be created! This disables kernel caching. Specified directory is /root/.cache/torch/kernels. This warning will appear only once per process. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/cuda/jit_utils.cpp:860.) + tensor.erfinv_() +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +/opt/conda/lib/python3.8/site-packages/mmengine/model/weight_init.py:649: UserWarning: Specified kernel cache directory could not be created! This disables kernel caching. Specified directory is /root/.cache/torch/kernels. This warning will appear only once per process. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/cuda/jit_utils.cpp:860.) + tensor.erfinv_() +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +04/16 08:10:06 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +04/16 08:10:06 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +04/16 08:10:06 - mmengine - INFO - Checkpoints will be saved to /workspaces/mmsegmentation-1/work_dirs/mae-base_upernet_8xb2-amp-160k_cag-512x512. +04/16 08:11:03 - mmengine - INFO - Iter(train) [ 50/160000] base_lr: 3.2689e-06 lr: 1.2086e-08 eta: 2 days, 2:27:28 time: 1.0323 data_time: 0.0047 memory: 8935 loss: 7.1374 decode.loss_ce: 5.1061 decode.acc_seg: 2.7090 aux.loss_ce: 2.0313 aux.acc_seg: 0.0000 +04/16 08:11:54 - mmengine - INFO - Iter(train) [ 100/160000] base_lr: 6.6045e-06 lr: 2.4418e-08 eta: 2 days, 0:06:13 time: 1.0280 data_time: 0.0044 memory: 8462 loss: 6.8298 decode.loss_ce: 4.8244 decode.acc_seg: 26.8021 aux.loss_ce: 2.0054 aux.acc_seg: 0.1680 +04/16 08:12:46 - mmengine - INFO - Iter(train) [ 150/160000] base_lr: 9.9401e-06 lr: 3.6750e-08 eta: 1 day, 23:16:43 time: 1.0264 data_time: 0.0042 memory: 8462 loss: 6.3951 decode.loss_ce: 4.4258 decode.acc_seg: 57.5527 aux.loss_ce: 1.9693 aux.acc_seg: 27.9995 +04/16 08:13:37 - mmengine - INFO - Iter(train) [ 200/160000] base_lr: 1.3276e-05 lr: 4.9083e-08 eta: 1 day, 22:48:42 time: 1.0202 data_time: 0.0040 memory: 8462 loss: 5.8755 decode.loss_ce: 3.9571 decode.acc_seg: 70.3098 aux.loss_ce: 1.9185 aux.acc_seg: 25.9087 +04/16 08:14:28 - mmengine - INFO - Iter(train) [ 250/160000] base_lr: 1.6611e-05 lr: 6.1415e-08 eta: 1 day, 22:27:43 time: 1.0164 data_time: 0.0046 memory: 8462 loss: 5.1345 decode.loss_ce: 3.2971 decode.acc_seg: 80.0583 aux.loss_ce: 1.8375 aux.acc_seg: 45.8790 +04/16 08:15:19 - mmengine - INFO - Iter(train) [ 300/160000] base_lr: 1.9947e-05 lr: 7.3747e-08 eta: 1 day, 22:12:04 time: 1.0117 data_time: 0.0040 memory: 8462 loss: 4.6058 decode.loss_ce: 2.8301 decode.acc_seg: 84.3102 aux.loss_ce: 1.7756 aux.acc_seg: 37.2120 +04/16 08:16:09 - mmengine - INFO - Iter(train) [ 350/160000] base_lr: 2.3282e-05 lr: 8.6079e-08 eta: 1 day, 21:58:41 time: 1.0075 data_time: 0.0046 memory: 8462 loss: 3.9343 decode.loss_ce: 2.2677 decode.acc_seg: 90.3549 aux.loss_ce: 1.6666 aux.acc_seg: 51.3355 +04/16 08:16:59 - mmengine - INFO - Iter(train) [ 400/160000] base_lr: 2.6618e-05 lr: 9.8412e-08 eta: 1 day, 21:47:59 time: 1.0072 data_time: 0.0041 memory: 8462 loss: 3.2720 decode.loss_ce: 1.7116 decode.acc_seg: 92.8013 aux.loss_ce: 1.5604 aux.acc_seg: 62.0686 +04/16 08:17:50 - mmengine - INFO - Iter(train) [ 450/160000] base_lr: 2.9953e-05 lr: 1.1074e-07 eta: 1 day, 21:39:08 time: 1.0060 data_time: 0.0041 memory: 8462 loss: 2.5921 decode.loss_ce: 1.1644 decode.acc_seg: 97.5016 aux.loss_ce: 1.4277 aux.acc_seg: 68.5354 +04/16 08:18:40 - mmengine - INFO - Iter(train) [ 500/160000] base_lr: 3.3289e-05 lr: 1.2308e-07 eta: 1 day, 21:31:50 time: 1.0055 data_time: 0.0042 memory: 8462 loss: 2.0673 decode.loss_ce: 0.7804 decode.acc_seg: 97.7400 aux.loss_ce: 1.2869 aux.acc_seg: 67.5140 +04/16 08:19:30 - mmengine - INFO - Iter(train) [ 550/160000] base_lr: 3.6624e-05 lr: 1.3541e-07 eta: 1 day, 21:25:21 time: 1.0046 data_time: 0.0044 memory: 8462 loss: 1.6140 decode.loss_ce: 0.5170 decode.acc_seg: 96.9492 aux.loss_ce: 1.0970 aux.acc_seg: 83.8314 +04/16 08:20:20 - mmengine - INFO - Iter(train) [ 600/160000] base_lr: 3.9960e-05 lr: 1.4774e-07 eta: 1 day, 21:19:55 time: 1.0043 data_time: 0.0044 memory: 8462 loss: 1.3474 decode.loss_ce: 0.3544 decode.acc_seg: 97.2013 aux.loss_ce: 0.9931 aux.acc_seg: 82.7944 +04/16 08:21:11 - mmengine - INFO - Iter(train) [ 650/160000] base_lr: 4.3296e-05 lr: 1.6007e-07 eta: 1 day, 21:14:59 time: 1.0030 data_time: 0.0046 memory: 8462 loss: 1.0697 decode.loss_ce: 0.2786 decode.acc_seg: 98.2925 aux.loss_ce: 0.7911 aux.acc_seg: 94.7325 +04/16 08:22:01 - mmengine - INFO - Iter(train) [ 700/160000] base_lr: 4.6631e-05 lr: 1.7240e-07 eta: 1 day, 21:10:12 time: 0.9999 data_time: 0.0045 memory: 8462 loss: 0.8254 decode.loss_ce: 0.2025 decode.acc_seg: 98.8443 aux.loss_ce: 0.6229 aux.acc_seg: 97.8575 +04/16 08:22:51 - mmengine - INFO - Iter(train) [ 750/160000] base_lr: 4.9967e-05 lr: 1.8474e-07 eta: 1 day, 21:05:44 time: 0.9987 data_time: 0.0043 memory: 8462 loss: 0.6389 decode.loss_ce: 0.1660 decode.acc_seg: 96.1958 aux.loss_ce: 0.4728 aux.acc_seg: 94.6260 +04/16 08:23:41 - mmengine - INFO - Iter(train) [ 800/160000] base_lr: 5.3302e-05 lr: 1.9707e-07 eta: 1 day, 21:01:16 time: 0.9981 data_time: 0.0045 memory: 8462 loss: 0.5138 decode.loss_ce: 0.1463 decode.acc_seg: 97.7625 aux.loss_ce: 0.3675 aux.acc_seg: 96.5370 +04/16 08:24:30 - mmengine - INFO - Iter(train) [ 850/160000] base_lr: 5.6638e-05 lr: 2.0940e-07 eta: 1 day, 20:57:08 time: 0.9965 data_time: 0.0042 memory: 8462 loss: 0.4012 decode.loss_ce: 0.1269 decode.acc_seg: 98.1451 aux.loss_ce: 0.2743 aux.acc_seg: 96.9545 +04/16 08:25:20 - mmengine - INFO - Iter(train) [ 900/160000] base_lr: 5.9973e-05 lr: 2.2173e-07 eta: 1 day, 20:53:32 time: 0.9979 data_time: 0.0043 memory: 8462 loss: 0.3109 decode.loss_ce: 0.0988 decode.acc_seg: 98.7818 aux.loss_ce: 0.2121 aux.acc_seg: 97.1060 +04/16 08:26:10 - mmengine - INFO - Iter(train) [ 950/160000] base_lr: 6.3309e-05 lr: 2.3407e-07 eta: 1 day, 20:50:07 time: 0.9976 data_time: 0.0040 memory: 8462 loss: 0.2525 decode.loss_ce: 0.0933 decode.acc_seg: 97.1182 aux.loss_ce: 0.1593 aux.acc_seg: 94.3285 +04/16 08:27:00 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 08:27:00 - mmengine - INFO - Iter(train) [ 1000/160000] base_lr: 6.6644e-05 lr: 2.4640e-07 eta: 1 day, 20:46:57 time: 0.9972 data_time: 0.0042 memory: 8462 loss: 0.2075 decode.loss_ce: 0.0836 decode.acc_seg: 98.2325 aux.loss_ce: 0.1239 aux.acc_seg: 97.9246 +04/16 08:27:50 - mmengine - INFO - Iter(train) [ 1050/160000] base_lr: 6.9980e-05 lr: 2.5873e-07 eta: 1 day, 20:44:03 time: 0.9977 data_time: 0.0040 memory: 8462 loss: 0.1839 decode.loss_ce: 0.0781 decode.acc_seg: 98.1035 aux.loss_ce: 0.1059 aux.acc_seg: 97.7102 +04/16 08:28:40 - mmengine - INFO - Iter(train) [ 1100/160000] base_lr: 7.3316e-05 lr: 2.7106e-07 eta: 1 day, 20:41:20 time: 0.9965 data_time: 0.0041 memory: 8462 loss: 0.1804 decode.loss_ce: 0.0764 decode.acc_seg: 96.8763 aux.loss_ce: 0.1039 aux.acc_seg: 96.4491 +04/16 08:29:30 - mmengine - INFO - Iter(train) [ 1150/160000] base_lr: 7.6651e-05 lr: 2.8339e-07 eta: 1 day, 20:38:45 time: 0.9978 data_time: 0.0045 memory: 8462 loss: 0.1393 decode.loss_ce: 0.0634 decode.acc_seg: 97.7964 aux.loss_ce: 0.0759 aux.acc_seg: 96.7056 +04/16 08:30:20 - mmengine - INFO - Iter(train) [ 1200/160000] base_lr: 7.9987e-05 lr: 2.9573e-07 eta: 1 day, 20:36:19 time: 0.9976 data_time: 0.0041 memory: 8462 loss: 0.1442 decode.loss_ce: 0.0723 decode.acc_seg: 97.6814 aux.loss_ce: 0.0719 aux.acc_seg: 96.8489 +04/16 08:31:09 - mmengine - INFO - Iter(train) [ 1250/160000] base_lr: 8.3322e-05 lr: 3.0806e-07 eta: 1 day, 20:34:06 time: 0.9977 data_time: 0.0044 memory: 8462 loss: 0.1477 decode.loss_ce: 0.0750 decode.acc_seg: 97.8846 aux.loss_ce: 0.0727 aux.acc_seg: 97.8550 +04/16 08:31:59 - mmengine - INFO - Iter(train) [ 1300/160000] base_lr: 8.6658e-05 lr: 3.2039e-07 eta: 1 day, 20:31:57 time: 0.9982 data_time: 0.0041 memory: 8462 loss: 0.1178 decode.loss_ce: 0.0598 decode.acc_seg: 98.3858 aux.loss_ce: 0.0580 aux.acc_seg: 97.7406 +04/16 08:32:49 - mmengine - INFO - Iter(train) [ 1350/160000] base_lr: 8.9993e-05 lr: 3.3272e-07 eta: 1 day, 20:29:53 time: 0.9979 data_time: 0.0041 memory: 8462 loss: 0.1109 decode.loss_ce: 0.0613 decode.acc_seg: 97.9813 aux.loss_ce: 0.0496 aux.acc_seg: 97.3036 +04/16 08:33:39 - mmengine - INFO - Iter(train) [ 1400/160000] base_lr: 9.3329e-05 lr: 3.4506e-07 eta: 1 day, 20:27:50 time: 0.9982 data_time: 0.0044 memory: 8462 loss: 0.1039 decode.loss_ce: 0.0561 decode.acc_seg: 97.9427 aux.loss_ce: 0.0478 aux.acc_seg: 97.4939 +04/16 08:34:29 - mmengine - INFO - Iter(train) [ 1450/160000] base_lr: 9.6664e-05 lr: 3.5739e-07 eta: 1 day, 20:25:53 time: 0.9973 data_time: 0.0041 memory: 8462 loss: 0.1053 decode.loss_ce: 0.0591 decode.acc_seg: 98.8289 aux.loss_ce: 0.0462 aux.acc_seg: 98.2611 +04/16 08:35:19 - mmengine - INFO - Iter(train) [ 1500/160000] base_lr: 1.0000e-04 lr: 3.6972e-07 eta: 1 day, 20:24:03 time: 0.9975 data_time: 0.0044 memory: 8462 loss: 0.0957 decode.loss_ce: 0.0518 decode.acc_seg: 98.0558 aux.loss_ce: 0.0439 aux.acc_seg: 97.3547 +04/16 08:36:09 - mmengine - INFO - Iter(train) [ 1550/160000] base_lr: 9.9969e-05 lr: 3.6961e-07 eta: 1 day, 20:22:18 time: 0.9972 data_time: 0.0040 memory: 8462 loss: 0.1076 decode.loss_ce: 0.0628 decode.acc_seg: 96.7735 aux.loss_ce: 0.0448 aux.acc_seg: 96.2715 +04/16 08:36:59 - mmengine - INFO - Iter(train) [ 1600/160000] base_lr: 9.9938e-05 lr: 3.6949e-07 eta: 1 day, 20:20:37 time: 0.9957 data_time: 0.0043 memory: 8462 loss: 0.0848 decode.loss_ce: 0.0441 decode.acc_seg: 98.0389 aux.loss_ce: 0.0407 aux.acc_seg: 97.3362 +04/16 08:37:49 - mmengine - INFO - Iter(train) [ 1650/160000] base_lr: 9.9906e-05 lr: 3.6937e-07 eta: 1 day, 20:19:02 time: 0.9975 data_time: 0.0042 memory: 8462 loss: 0.0880 decode.loss_ce: 0.0508 decode.acc_seg: 98.3910 aux.loss_ce: 0.0371 aux.acc_seg: 97.4863 +04/16 08:38:38 - mmengine - INFO - Iter(train) [ 1700/160000] base_lr: 9.9874e-05 lr: 3.6926e-07 eta: 1 day, 20:17:28 time: 0.9974 data_time: 0.0040 memory: 8462 loss: 0.0877 decode.loss_ce: 0.0514 decode.acc_seg: 98.4911 aux.loss_ce: 0.0362 aux.acc_seg: 97.8970 +04/16 08:39:28 - mmengine - INFO - Iter(train) [ 1750/160000] base_lr: 9.9843e-05 lr: 3.6914e-07 eta: 1 day, 20:15:54 time: 0.9977 data_time: 0.0044 memory: 8462 loss: 0.0777 decode.loss_ce: 0.0452 decode.acc_seg: 97.4014 aux.loss_ce: 0.0325 aux.acc_seg: 97.5168 +04/16 08:40:18 - mmengine - INFO - Iter(train) [ 1800/160000] base_lr: 9.9811e-05 lr: 3.6902e-07 eta: 1 day, 20:14:21 time: 0.9983 data_time: 0.0044 memory: 8462 loss: 0.0848 decode.loss_ce: 0.0489 decode.acc_seg: 98.3080 aux.loss_ce: 0.0359 aux.acc_seg: 97.6761 +04/16 08:41:08 - mmengine - INFO - Iter(train) [ 1850/160000] base_lr: 9.9780e-05 lr: 3.6891e-07 eta: 1 day, 20:12:52 time: 0.9967 data_time: 0.0042 memory: 8462 loss: 0.0878 decode.loss_ce: 0.0555 decode.acc_seg: 98.5981 aux.loss_ce: 0.0323 aux.acc_seg: 98.0816 +04/16 08:41:58 - mmengine - INFO - Iter(train) [ 1900/160000] base_lr: 9.9748e-05 lr: 3.6879e-07 eta: 1 day, 20:11:23 time: 0.9979 data_time: 0.0047 memory: 8462 loss: 0.0696 decode.loss_ce: 0.0407 decode.acc_seg: 98.9454 aux.loss_ce: 0.0289 aux.acc_seg: 98.5573 +04/16 08:42:48 - mmengine - INFO - Iter(train) [ 1950/160000] base_lr: 9.9717e-05 lr: 3.6867e-07 eta: 1 day, 20:09:59 time: 0.9975 data_time: 0.0041 memory: 8462 loss: 0.0650 decode.loss_ce: 0.0386 decode.acc_seg: 98.4501 aux.loss_ce: 0.0264 aux.acc_seg: 97.8333 +04/16 08:43:38 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 08:43:38 - mmengine - INFO - Iter(train) [ 2000/160000] base_lr: 9.9685e-05 lr: 3.6856e-07 eta: 1 day, 20:08:37 time: 0.9983 data_time: 0.0048 memory: 8462 loss: 0.0783 decode.loss_ce: 0.0462 decode.acc_seg: 98.0640 aux.loss_ce: 0.0322 aux.acc_seg: 97.0270 +04/16 08:44:28 - mmengine - INFO - Iter(train) [ 2050/160000] base_lr: 9.9654e-05 lr: 3.6844e-07 eta: 1 day, 20:07:15 time: 0.9977 data_time: 0.0046 memory: 8462 loss: 0.0654 decode.loss_ce: 0.0391 decode.acc_seg: 98.8264 aux.loss_ce: 0.0263 aux.acc_seg: 98.0137 +04/16 08:45:17 - mmengine - INFO - Iter(train) [ 2100/160000] base_lr: 9.9622e-05 lr: 3.6832e-07 eta: 1 day, 20:05:56 time: 0.9981 data_time: 0.0042 memory: 8462 loss: 0.0681 decode.loss_ce: 0.0394 decode.acc_seg: 99.0908 aux.loss_ce: 0.0287 aux.acc_seg: 98.6416 +04/16 08:45:48 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 08:46:07 - mmengine - INFO - Iter(train) [ 2150/160000] base_lr: 9.9591e-05 lr: 3.6821e-07 eta: 1 day, 20:04:41 time: 0.9983 data_time: 0.0042 memory: 8462 loss: 0.0750 decode.loss_ce: 0.0456 decode.acc_seg: 98.0452 aux.loss_ce: 0.0294 aux.acc_seg: 96.4743 +04/16 08:46:57 - mmengine - INFO - Iter(train) [ 2200/160000] base_lr: 9.9559e-05 lr: 3.6809e-07 eta: 1 day, 20:03:24 time: 0.9969 data_time: 0.0043 memory: 8462 loss: 0.0626 decode.loss_ce: 0.0372 decode.acc_seg: 98.7965 aux.loss_ce: 0.0254 aux.acc_seg: 98.4007 +04/16 08:47:47 - mmengine - INFO - Iter(train) [ 2250/160000] base_lr: 9.9527e-05 lr: 3.6797e-07 eta: 1 day, 20:02:07 time: 0.9962 data_time: 0.0041 memory: 8462 loss: 0.0595 decode.loss_ce: 0.0356 decode.acc_seg: 99.3166 aux.loss_ce: 0.0238 aux.acc_seg: 98.3961 +04/16 08:48:37 - mmengine - INFO - Iter(train) [ 2300/160000] base_lr: 9.9496e-05 lr: 3.6786e-07 eta: 1 day, 20:00:51 time: 0.9996 data_time: 0.0047 memory: 8462 loss: 0.0570 decode.loss_ce: 0.0344 decode.acc_seg: 99.5049 aux.loss_ce: 0.0226 aux.acc_seg: 98.8173 +04/16 08:49:27 - mmengine - INFO - Iter(train) [ 2350/160000] base_lr: 9.9464e-05 lr: 3.6774e-07 eta: 1 day, 19:59:38 time: 0.9978 data_time: 0.0044 memory: 8462 loss: 0.0608 decode.loss_ce: 0.0368 decode.acc_seg: 97.5445 aux.loss_ce: 0.0240 aux.acc_seg: 97.2980 +04/16 08:50:17 - mmengine - INFO - Iter(train) [ 2400/160000] base_lr: 9.9433e-05 lr: 3.6762e-07 eta: 1 day, 19:58:24 time: 0.9973 data_time: 0.0044 memory: 8462 loss: 0.0567 decode.loss_ce: 0.0334 decode.acc_seg: 98.2815 aux.loss_ce: 0.0232 aux.acc_seg: 97.8489 +04/16 08:51:07 - mmengine - INFO - Iter(train) [ 2450/160000] base_lr: 9.9401e-05 lr: 3.6751e-07 eta: 1 day, 19:57:11 time: 0.9974 data_time: 0.0041 memory: 8462 loss: 0.0542 decode.loss_ce: 0.0325 decode.acc_seg: 98.7648 aux.loss_ce: 0.0217 aux.acc_seg: 98.1602 +04/16 08:51:57 - mmengine - INFO - Iter(train) [ 2500/160000] base_lr: 9.9370e-05 lr: 3.6739e-07 eta: 1 day, 19:56:01 time: 0.9988 data_time: 0.0043 memory: 8462 loss: 0.0509 decode.loss_ce: 0.0299 decode.acc_seg: 99.0875 aux.loss_ce: 0.0210 aux.acc_seg: 98.3164 +04/16 08:52:46 - mmengine - INFO - Iter(train) [ 2550/160000] base_lr: 9.9338e-05 lr: 3.6727e-07 eta: 1 day, 19:54:50 time: 0.9980 data_time: 0.0044 memory: 8462 loss: 0.0517 decode.loss_ce: 0.0306 decode.acc_seg: 98.6189 aux.loss_ce: 0.0211 aux.acc_seg: 98.1190 +04/16 08:53:36 - mmengine - INFO - Iter(train) [ 2600/160000] base_lr: 9.9307e-05 lr: 3.6716e-07 eta: 1 day, 19:53:41 time: 0.9969 data_time: 0.0042 memory: 8462 loss: 0.0544 decode.loss_ce: 0.0320 decode.acc_seg: 98.2189 aux.loss_ce: 0.0223 aux.acc_seg: 97.2631 +04/16 08:54:26 - mmengine - INFO - Iter(train) [ 2650/160000] base_lr: 9.9275e-05 lr: 3.6704e-07 eta: 1 day, 19:52:30 time: 0.9963 data_time: 0.0040 memory: 8462 loss: 0.0589 decode.loss_ce: 0.0355 decode.acc_seg: 98.5888 aux.loss_ce: 0.0233 aux.acc_seg: 98.1344 +04/16 08:55:16 - mmengine - INFO - Iter(train) [ 2700/160000] base_lr: 9.9244e-05 lr: 3.6692e-07 eta: 1 day, 19:51:22 time: 0.9980 data_time: 0.0045 memory: 8462 loss: 0.0546 decode.loss_ce: 0.0329 decode.acc_seg: 98.9714 aux.loss_ce: 0.0218 aux.acc_seg: 98.4739 +04/16 08:56:06 - mmengine - INFO - Iter(train) [ 2750/160000] base_lr: 9.9212e-05 lr: 3.6681e-07 eta: 1 day, 19:50:14 time: 0.9976 data_time: 0.0042 memory: 8462 loss: 0.0515 decode.loss_ce: 0.0310 decode.acc_seg: 99.3862 aux.loss_ce: 0.0204 aux.acc_seg: 98.8497 +04/16 08:56:56 - mmengine - INFO - Iter(train) [ 2800/160000] base_lr: 9.9180e-05 lr: 3.6669e-07 eta: 1 day, 19:49:06 time: 0.9974 data_time: 0.0041 memory: 8462 loss: 0.0478 decode.loss_ce: 0.0284 decode.acc_seg: 99.3832 aux.loss_ce: 0.0194 aux.acc_seg: 98.8251 +04/16 08:57:46 - mmengine - INFO - Iter(train) [ 2850/160000] base_lr: 9.9149e-05 lr: 3.6657e-07 eta: 1 day, 19:47:59 time: 0.9973 data_time: 0.0040 memory: 8462 loss: 0.0514 decode.loss_ce: 0.0303 decode.acc_seg: 99.1659 aux.loss_ce: 0.0211 aux.acc_seg: 98.4051 +04/16 08:58:36 - mmengine - INFO - Iter(train) [ 2900/160000] base_lr: 9.9117e-05 lr: 3.6646e-07 eta: 1 day, 19:46:52 time: 0.9975 data_time: 0.0040 memory: 8462 loss: 0.0500 decode.loss_ce: 0.0297 decode.acc_seg: 99.0492 aux.loss_ce: 0.0203 aux.acc_seg: 98.0768 +04/16 08:59:25 - mmengine - INFO - Iter(train) [ 2950/160000] base_lr: 9.9086e-05 lr: 3.6634e-07 eta: 1 day, 19:45:48 time: 0.9980 data_time: 0.0047 memory: 8462 loss: 0.0498 decode.loss_ce: 0.0296 decode.acc_seg: 98.2616 aux.loss_ce: 0.0202 aux.acc_seg: 97.5533 +04/16 09:00:15 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 09:00:15 - mmengine - INFO - Iter(train) [ 3000/160000] base_lr: 9.9054e-05 lr: 3.6622e-07 eta: 1 day, 19:44:44 time: 0.9967 data_time: 0.0041 memory: 8462 loss: 0.0492 decode.loss_ce: 0.0290 decode.acc_seg: 99.5146 aux.loss_ce: 0.0203 aux.acc_seg: 98.7753 +04/16 09:01:05 - mmengine - INFO - Iter(train) [ 3050/160000] base_lr: 9.9023e-05 lr: 3.6611e-07 eta: 1 day, 19:43:40 time: 0.9972 data_time: 0.0041 memory: 8462 loss: 0.0420 decode.loss_ce: 0.0242 decode.acc_seg: 99.1709 aux.loss_ce: 0.0178 aux.acc_seg: 98.9540 +04/16 09:01:55 - mmengine - INFO - Iter(train) [ 3100/160000] base_lr: 9.8991e-05 lr: 3.6599e-07 eta: 1 day, 19:42:37 time: 0.9975 data_time: 0.0040 memory: 8462 loss: 0.0550 decode.loss_ce: 0.0337 decode.acc_seg: 97.5500 aux.loss_ce: 0.0213 aux.acc_seg: 97.6732 +04/16 09:02:45 - mmengine - INFO - Iter(train) [ 3150/160000] base_lr: 9.8960e-05 lr: 3.6587e-07 eta: 1 day, 19:41:34 time: 0.9973 data_time: 0.0041 memory: 8462 loss: 0.0424 decode.loss_ce: 0.0242 decode.acc_seg: 99.4364 aux.loss_ce: 0.0183 aux.acc_seg: 98.9170 +04/16 09:03:35 - mmengine - INFO - Iter(train) [ 3200/160000] base_lr: 9.8928e-05 lr: 3.6576e-07 eta: 1 day, 19:40:29 time: 0.9974 data_time: 0.0043 memory: 8462 loss: 0.0388 decode.loss_ce: 0.0224 decode.acc_seg: 98.3208 aux.loss_ce: 0.0164 aux.acc_seg: 97.7089 +04/16 09:04:25 - mmengine - INFO - Iter(train) [ 3250/160000] base_lr: 9.8897e-05 lr: 3.6564e-07 eta: 1 day, 19:39:27 time: 0.9980 data_time: 0.0043 memory: 8462 loss: 0.0416 decode.loss_ce: 0.0248 decode.acc_seg: 99.2579 aux.loss_ce: 0.0168 aux.acc_seg: 98.5394 +04/16 09:05:15 - mmengine - INFO - Iter(train) [ 3300/160000] base_lr: 9.8865e-05 lr: 3.6552e-07 eta: 1 day, 19:38:26 time: 0.9982 data_time: 0.0045 memory: 8462 loss: 0.0442 decode.loss_ce: 0.0261 decode.acc_seg: 99.2485 aux.loss_ce: 0.0182 aux.acc_seg: 98.9531 +04/16 09:06:05 - mmengine - INFO - Iter(train) [ 3350/160000] base_lr: 9.8833e-05 lr: 3.6541e-07 eta: 1 day, 19:37:25 time: 0.9977 data_time: 0.0042 memory: 8462 loss: 0.0524 decode.loss_ce: 0.0322 decode.acc_seg: 97.8529 aux.loss_ce: 0.0202 aux.acc_seg: 97.3797 +04/16 09:06:54 - mmengine - INFO - Iter(train) [ 3400/160000] base_lr: 9.8802e-05 lr: 3.6529e-07 eta: 1 day, 19:36:25 time: 0.9973 data_time: 0.0043 memory: 8462 loss: 0.0447 decode.loss_ce: 0.0265 decode.acc_seg: 98.5550 aux.loss_ce: 0.0182 aux.acc_seg: 98.0677 +04/16 09:07:44 - mmengine - INFO - Iter(train) [ 3450/160000] base_lr: 9.8770e-05 lr: 3.6517e-07 eta: 1 day, 19:35:26 time: 0.9996 data_time: 0.0043 memory: 8462 loss: 0.0530 decode.loss_ce: 0.0320 decode.acc_seg: 99.4623 aux.loss_ce: 0.0210 aux.acc_seg: 98.9674 +04/16 09:08:34 - mmengine - INFO - Iter(train) [ 3500/160000] base_lr: 9.8739e-05 lr: 3.6506e-07 eta: 1 day, 19:34:25 time: 0.9973 data_time: 0.0041 memory: 8462 loss: 0.0466 decode.loss_ce: 0.0287 decode.acc_seg: 99.1919 aux.loss_ce: 0.0179 aux.acc_seg: 98.7110 +04/16 09:09:24 - mmengine - INFO - Iter(train) [ 3550/160000] base_lr: 9.8707e-05 lr: 3.6494e-07 eta: 1 day, 19:33:25 time: 0.9972 data_time: 0.0039 memory: 8462 loss: 0.0358 decode.loss_ce: 0.0201 decode.acc_seg: 99.3044 aux.loss_ce: 0.0157 aux.acc_seg: 98.6568 +04/16 09:10:14 - mmengine - INFO - Iter(train) [ 3600/160000] base_lr: 9.8676e-05 lr: 3.6482e-07 eta: 1 day, 19:32:24 time: 0.9962 data_time: 0.0041 memory: 8462 loss: 0.0403 decode.loss_ce: 0.0239 decode.acc_seg: 99.0747 aux.loss_ce: 0.0163 aux.acc_seg: 98.2910 +04/16 09:11:04 - mmengine - INFO - Iter(train) [ 3650/160000] base_lr: 9.8644e-05 lr: 3.6471e-07 eta: 1 day, 19:31:23 time: 0.9971 data_time: 0.0042 memory: 8462 loss: 0.0385 decode.loss_ce: 0.0218 decode.acc_seg: 99.1697 aux.loss_ce: 0.0166 aux.acc_seg: 98.4314 +04/16 09:11:54 - mmengine - INFO - Iter(train) [ 3700/160000] base_lr: 9.8613e-05 lr: 3.6459e-07 eta: 1 day, 19:30:24 time: 0.9977 data_time: 0.0045 memory: 8462 loss: 0.0433 decode.loss_ce: 0.0254 decode.acc_seg: 98.5950 aux.loss_ce: 0.0179 aux.acc_seg: 97.6034 +04/16 09:12:44 - mmengine - INFO - Iter(train) [ 3750/160000] base_lr: 9.8581e-05 lr: 3.6447e-07 eta: 1 day, 19:29:26 time: 0.9987 data_time: 0.0046 memory: 8462 loss: 0.0491 decode.loss_ce: 0.0297 decode.acc_seg: 99.2804 aux.loss_ce: 0.0193 aux.acc_seg: 98.4362 +04/16 09:13:34 - mmengine - INFO - Iter(train) [ 3800/160000] base_lr: 9.8550e-05 lr: 3.6436e-07 eta: 1 day, 19:28:27 time: 0.9966 data_time: 0.0042 memory: 8462 loss: 0.0460 decode.loss_ce: 0.0283 decode.acc_seg: 99.6212 aux.loss_ce: 0.0178 aux.acc_seg: 99.1259 +04/16 09:14:23 - mmengine - INFO - Iter(train) [ 3850/160000] base_lr: 9.8518e-05 lr: 3.6424e-07 eta: 1 day, 19:27:29 time: 0.9976 data_time: 0.0043 memory: 8462 loss: 0.0404 decode.loss_ce: 0.0241 decode.acc_seg: 98.6940 aux.loss_ce: 0.0163 aux.acc_seg: 97.9586 +04/16 09:15:13 - mmengine - INFO - Iter(train) [ 3900/160000] base_lr: 9.8486e-05 lr: 3.6412e-07 eta: 1 day, 19:26:31 time: 0.9969 data_time: 0.0045 memory: 8462 loss: 0.0406 decode.loss_ce: 0.0241 decode.acc_seg: 98.7534 aux.loss_ce: 0.0165 aux.acc_seg: 98.0196 +04/16 09:16:03 - mmengine - INFO - Iter(train) [ 3950/160000] base_lr: 9.8455e-05 lr: 3.6401e-07 eta: 1 day, 19:25:32 time: 0.9976 data_time: 0.0045 memory: 8462 loss: 0.0419 decode.loss_ce: 0.0252 decode.acc_seg: 99.4352 aux.loss_ce: 0.0167 aux.acc_seg: 98.8262 +04/16 09:16:53 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 09:16:53 - mmengine - INFO - Iter(train) [ 4000/160000] base_lr: 9.8423e-05 lr: 3.6389e-07 eta: 1 day, 19:24:34 time: 0.9976 data_time: 0.0042 memory: 8462 loss: 0.0354 decode.loss_ce: 0.0202 decode.acc_seg: 98.6773 aux.loss_ce: 0.0152 aux.acc_seg: 97.7581 +04/16 09:17:43 - mmengine - INFO - Iter(train) [ 4050/160000] base_lr: 9.8392e-05 lr: 3.6377e-07 eta: 1 day, 19:23:36 time: 0.9966 data_time: 0.0042 memory: 8462 loss: 0.0412 decode.loss_ce: 0.0250 decode.acc_seg: 97.9300 aux.loss_ce: 0.0162 aux.acc_seg: 97.4089 +04/16 09:18:33 - mmengine - INFO - Iter(train) [ 4100/160000] base_lr: 9.8360e-05 lr: 3.6366e-07 eta: 1 day, 19:22:38 time: 0.9984 data_time: 0.0045 memory: 8462 loss: 0.0445 decode.loss_ce: 0.0277 decode.acc_seg: 98.4241 aux.loss_ce: 0.0168 aux.acc_seg: 97.7720 +04/16 09:19:23 - mmengine - INFO - Iter(train) [ 4150/160000] base_lr: 9.8329e-05 lr: 3.6354e-07 eta: 1 day, 19:21:39 time: 0.9965 data_time: 0.0042 memory: 8462 loss: 0.0387 decode.loss_ce: 0.0222 decode.acc_seg: 99.3420 aux.loss_ce: 0.0164 aux.acc_seg: 98.7667 +04/16 09:20:13 - mmengine - INFO - Iter(train) [ 4200/160000] base_lr: 9.8297e-05 lr: 3.6342e-07 eta: 1 day, 19:20:42 time: 0.9976 data_time: 0.0041 memory: 8462 loss: 0.0386 decode.loss_ce: 0.0228 decode.acc_seg: 98.6504 aux.loss_ce: 0.0158 aux.acc_seg: 98.1169 +04/16 09:21:03 - mmengine - INFO - Iter(train) [ 4250/160000] base_lr: 9.8266e-05 lr: 3.6331e-07 eta: 1 day, 19:19:45 time: 0.9970 data_time: 0.0043 memory: 8462 loss: 0.0370 decode.loss_ce: 0.0221 decode.acc_seg: 98.6811 aux.loss_ce: 0.0149 aux.acc_seg: 98.2132 +04/16 09:21:52 - mmengine - INFO - Iter(train) [ 4300/160000] base_lr: 9.8234e-05 lr: 3.6319e-07 eta: 1 day, 19:18:48 time: 0.9976 data_time: 0.0044 memory: 8462 loss: 0.0386 decode.loss_ce: 0.0231 decode.acc_seg: 99.5756 aux.loss_ce: 0.0155 aux.acc_seg: 99.0969 +04/16 09:22:42 - mmengine - INFO - Iter(train) [ 4350/160000] base_lr: 9.8203e-05 lr: 3.6307e-07 eta: 1 day, 19:17:51 time: 0.9989 data_time: 0.0047 memory: 8462 loss: 0.0338 decode.loss_ce: 0.0193 decode.acc_seg: 99.4419 aux.loss_ce: 0.0145 aux.acc_seg: 98.6378 +04/16 09:23:32 - mmengine - INFO - Iter(train) [ 4400/160000] base_lr: 9.8171e-05 lr: 3.6296e-07 eta: 1 day, 19:16:55 time: 0.9984 data_time: 0.0045 memory: 8462 loss: 0.0375 decode.loss_ce: 0.0225 decode.acc_seg: 98.9044 aux.loss_ce: 0.0150 aux.acc_seg: 98.0289 +04/16 09:24:22 - mmengine - INFO - Iter(train) [ 4450/160000] base_lr: 9.8139e-05 lr: 3.6284e-07 eta: 1 day, 19:15:58 time: 0.9972 data_time: 0.0047 memory: 8462 loss: 0.0413 decode.loss_ce: 0.0249 decode.acc_seg: 99.3196 aux.loss_ce: 0.0163 aux.acc_seg: 98.7162 +04/16 09:25:12 - mmengine - INFO - Iter(train) [ 4500/160000] base_lr: 9.8108e-05 lr: 3.6273e-07 eta: 1 day, 19:15:00 time: 0.9966 data_time: 0.0045 memory: 8462 loss: 0.0434 decode.loss_ce: 0.0268 decode.acc_seg: 98.5575 aux.loss_ce: 0.0166 aux.acc_seg: 97.8743 +04/16 09:26:02 - mmengine - INFO - Iter(train) [ 4550/160000] base_lr: 9.8076e-05 lr: 3.6261e-07 eta: 1 day, 19:14:02 time: 0.9963 data_time: 0.0043 memory: 8462 loss: 0.0366 decode.loss_ce: 0.0225 decode.acc_seg: 99.2420 aux.loss_ce: 0.0141 aux.acc_seg: 99.0316 +04/16 09:26:52 - mmengine - INFO - Iter(train) [ 4600/160000] base_lr: 9.8045e-05 lr: 3.6249e-07 eta: 1 day, 19:13:05 time: 0.9984 data_time: 0.0043 memory: 8462 loss: 0.0461 decode.loss_ce: 0.0290 decode.acc_seg: 99.2903 aux.loss_ce: 0.0171 aux.acc_seg: 98.5401 +04/16 09:27:41 - mmengine - INFO - Iter(train) [ 4650/160000] base_lr: 9.8013e-05 lr: 3.6238e-07 eta: 1 day, 19:12:08 time: 0.9968 data_time: 0.0048 memory: 8462 loss: 0.0399 decode.loss_ce: 0.0238 decode.acc_seg: 98.9355 aux.loss_ce: 0.0161 aux.acc_seg: 97.9330 +04/16 09:28:31 - mmengine - INFO - Iter(train) [ 4700/160000] base_lr: 9.7982e-05 lr: 3.6226e-07 eta: 1 day, 19:11:10 time: 0.9964 data_time: 0.0043 memory: 8462 loss: 0.0359 decode.loss_ce: 0.0218 decode.acc_seg: 98.7331 aux.loss_ce: 0.0141 aux.acc_seg: 98.1449 +04/16 09:29:21 - mmengine - INFO - Iter(train) [ 4750/160000] base_lr: 9.7950e-05 lr: 3.6214e-07 eta: 1 day, 19:10:13 time: 0.9964 data_time: 0.0043 memory: 8462 loss: 0.0330 decode.loss_ce: 0.0199 decode.acc_seg: 99.3498 aux.loss_ce: 0.0131 aux.acc_seg: 98.6938 +04/16 09:30:11 - mmengine - INFO - Iter(train) [ 4800/160000] base_lr: 9.7919e-05 lr: 3.6203e-07 eta: 1 day, 19:09:17 time: 0.9975 data_time: 0.0042 memory: 8462 loss: 0.0399 decode.loss_ce: 0.0239 decode.acc_seg: 99.3769 aux.loss_ce: 0.0160 aux.acc_seg: 98.5725 +04/16 09:31:01 - mmengine - INFO - Iter(train) [ 4850/160000] base_lr: 9.7887e-05 lr: 3.6191e-07 eta: 1 day, 19:08:21 time: 0.9966 data_time: 0.0041 memory: 8462 loss: 0.0399 decode.loss_ce: 0.0244 decode.acc_seg: 99.0669 aux.loss_ce: 0.0155 aux.acc_seg: 98.1323 +04/16 09:31:51 - mmengine - INFO - Iter(train) [ 4900/160000] base_lr: 9.7856e-05 lr: 3.6179e-07 eta: 1 day, 19:07:25 time: 0.9974 data_time: 0.0046 memory: 8462 loss: 0.0367 decode.loss_ce: 0.0224 decode.acc_seg: 98.8964 aux.loss_ce: 0.0142 aux.acc_seg: 98.2525 +04/16 09:32:41 - mmengine - INFO - Iter(train) [ 4950/160000] base_lr: 9.7824e-05 lr: 3.6168e-07 eta: 1 day, 19:06:28 time: 0.9960 data_time: 0.0041 memory: 8462 loss: 0.0437 decode.loss_ce: 0.0267 decode.acc_seg: 98.8968 aux.loss_ce: 0.0170 aux.acc_seg: 98.6990 +04/16 09:33:30 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 09:33:30 - mmengine - INFO - Iter(train) [ 5000/160000] base_lr: 9.7792e-05 lr: 3.6156e-07 eta: 1 day, 19:05:33 time: 0.9968 data_time: 0.0041 memory: 8462 loss: 0.0353 decode.loss_ce: 0.0210 decode.acc_seg: 99.4287 aux.loss_ce: 0.0143 aux.acc_seg: 99.0656 +04/16 09:34:20 - mmengine - INFO - Iter(train) [ 5050/160000] base_lr: 9.7761e-05 lr: 3.6144e-07 eta: 1 day, 19:04:37 time: 0.9966 data_time: 0.0041 memory: 8462 loss: 0.0453 decode.loss_ce: 0.0280 decode.acc_seg: 99.3763 aux.loss_ce: 0.0173 aux.acc_seg: 98.9359 +04/16 09:35:10 - mmengine - INFO - Iter(train) [ 5100/160000] base_lr: 9.7729e-05 lr: 3.6133e-07 eta: 1 day, 19:03:41 time: 0.9977 data_time: 0.0045 memory: 8462 loss: 0.0360 decode.loss_ce: 0.0216 decode.acc_seg: 99.3940 aux.loss_ce: 0.0144 aux.acc_seg: 98.8123 +04/16 09:36:00 - mmengine - INFO - Iter(train) [ 5150/160000] base_lr: 9.7698e-05 lr: 3.6121e-07 eta: 1 day, 19:02:45 time: 0.9974 data_time: 0.0046 memory: 8462 loss: 0.0324 decode.loss_ce: 0.0197 decode.acc_seg: 99.4511 aux.loss_ce: 0.0126 aux.acc_seg: 98.7846 +04/16 09:36:50 - mmengine - INFO - Iter(train) [ 5200/160000] base_lr: 9.7666e-05 lr: 3.6109e-07 eta: 1 day, 19:01:49 time: 0.9962 data_time: 0.0046 memory: 8462 loss: 0.0390 decode.loss_ce: 0.0239 decode.acc_seg: 99.4148 aux.loss_ce: 0.0151 aux.acc_seg: 99.0433 +04/16 09:37:40 - mmengine - INFO - Iter(train) [ 5250/160000] base_lr: 9.7635e-05 lr: 3.6098e-07 eta: 1 day, 19:00:54 time: 0.9971 data_time: 0.0048 memory: 8462 loss: 0.0351 decode.loss_ce: 0.0207 decode.acc_seg: 99.2474 aux.loss_ce: 0.0144 aux.acc_seg: 98.5289 +04/16 09:38:30 - mmengine - INFO - Iter(train) [ 5300/160000] base_lr: 9.7603e-05 lr: 3.6086e-07 eta: 1 day, 18:59:58 time: 0.9967 data_time: 0.0042 memory: 8462 loss: 0.0368 decode.loss_ce: 0.0223 decode.acc_seg: 99.2064 aux.loss_ce: 0.0145 aux.acc_seg: 98.6559 +04/16 09:39:19 - mmengine - INFO - Iter(train) [ 5350/160000] base_lr: 9.7572e-05 lr: 3.6074e-07 eta: 1 day, 18:59:03 time: 0.9974 data_time: 0.0045 memory: 8462 loss: 0.0340 decode.loss_ce: 0.0205 decode.acc_seg: 99.4291 aux.loss_ce: 0.0135 aux.acc_seg: 98.8754 +04/16 09:40:09 - mmengine - INFO - Iter(train) [ 5400/160000] base_lr: 9.7540e-05 lr: 3.6063e-07 eta: 1 day, 18:58:08 time: 0.9968 data_time: 0.0043 memory: 8462 loss: 0.0366 decode.loss_ce: 0.0218 decode.acc_seg: 99.3673 aux.loss_ce: 0.0149 aux.acc_seg: 98.6958 +04/16 09:40:59 - mmengine - INFO - Iter(train) [ 5450/160000] base_lr: 9.7509e-05 lr: 3.6051e-07 eta: 1 day, 18:57:12 time: 0.9963 data_time: 0.0045 memory: 8462 loss: 0.0440 decode.loss_ce: 0.0265 decode.acc_seg: 98.9367 aux.loss_ce: 0.0175 aux.acc_seg: 98.4566 +04/16 09:41:49 - mmengine - INFO - Iter(train) [ 5500/160000] base_lr: 9.7477e-05 lr: 3.6039e-07 eta: 1 day, 18:56:16 time: 0.9957 data_time: 0.0042 memory: 8462 loss: 0.0370 decode.loss_ce: 0.0222 decode.acc_seg: 99.1501 aux.loss_ce: 0.0148 aux.acc_seg: 98.9159 +04/16 09:42:39 - mmengine - INFO - Iter(train) [ 5550/160000] base_lr: 9.7445e-05 lr: 3.6028e-07 eta: 1 day, 18:55:22 time: 0.9966 data_time: 0.0041 memory: 8462 loss: 0.0297 decode.loss_ce: 0.0172 decode.acc_seg: 99.3139 aux.loss_ce: 0.0125 aux.acc_seg: 98.8188 +04/16 09:43:29 - mmengine - INFO - Iter(train) [ 5600/160000] base_lr: 9.7414e-05 lr: 3.6016e-07 eta: 1 day, 18:54:27 time: 0.9961 data_time: 0.0042 memory: 8462 loss: 0.0362 decode.loss_ce: 0.0216 decode.acc_seg: 99.2807 aux.loss_ce: 0.0145 aux.acc_seg: 98.4791 +04/16 09:44:18 - mmengine - INFO - Iter(train) [ 5650/160000] base_lr: 9.7382e-05 lr: 3.6004e-07 eta: 1 day, 18:53:32 time: 0.9976 data_time: 0.0045 memory: 8462 loss: 0.0384 decode.loss_ce: 0.0235 decode.acc_seg: 99.5243 aux.loss_ce: 0.0149 aux.acc_seg: 98.6925 +04/16 09:45:08 - mmengine - INFO - Iter(train) [ 5700/160000] base_lr: 9.7351e-05 lr: 3.5993e-07 eta: 1 day, 18:52:37 time: 0.9969 data_time: 0.0050 memory: 8462 loss: 0.0367 decode.loss_ce: 0.0220 decode.acc_seg: 98.9822 aux.loss_ce: 0.0146 aux.acc_seg: 98.4602 +04/16 09:45:58 - mmengine - INFO - Iter(train) [ 5750/160000] base_lr: 9.7319e-05 lr: 3.5981e-07 eta: 1 day, 18:51:43 time: 0.9968 data_time: 0.0044 memory: 8462 loss: 0.0276 decode.loss_ce: 0.0160 decode.acc_seg: 99.3057 aux.loss_ce: 0.0116 aux.acc_seg: 98.7503 +04/16 09:46:48 - mmengine - INFO - Iter(train) [ 5800/160000] base_lr: 9.7288e-05 lr: 3.5969e-07 eta: 1 day, 18:50:48 time: 0.9964 data_time: 0.0045 memory: 8462 loss: 0.0401 decode.loss_ce: 0.0247 decode.acc_seg: 98.8018 aux.loss_ce: 0.0154 aux.acc_seg: 98.1627 +04/16 09:47:38 - mmengine - INFO - Iter(train) [ 5850/160000] base_lr: 9.7256e-05 lr: 3.5958e-07 eta: 1 day, 18:49:53 time: 0.9957 data_time: 0.0044 memory: 8462 loss: 0.0321 decode.loss_ce: 0.0194 decode.acc_seg: 99.3196 aux.loss_ce: 0.0127 aux.acc_seg: 98.8062 +04/16 09:48:28 - mmengine - INFO - Iter(train) [ 5900/160000] base_lr: 9.7225e-05 lr: 3.5946e-07 eta: 1 day, 18:48:59 time: 0.9982 data_time: 0.0055 memory: 8462 loss: 0.0308 decode.loss_ce: 0.0182 decode.acc_seg: 99.1032 aux.loss_ce: 0.0125 aux.acc_seg: 98.3934 +04/16 09:49:17 - mmengine - INFO - Iter(train) [ 5950/160000] base_lr: 9.7193e-05 lr: 3.5934e-07 eta: 1 day, 18:48:04 time: 0.9968 data_time: 0.0044 memory: 8462 loss: 0.0358 decode.loss_ce: 0.0219 decode.acc_seg: 99.2605 aux.loss_ce: 0.0139 aux.acc_seg: 98.6856 +04/16 09:50:07 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 09:50:07 - mmengine - INFO - Iter(train) [ 6000/160000] base_lr: 9.7161e-05 lr: 3.5923e-07 eta: 1 day, 18:47:10 time: 0.9970 data_time: 0.0041 memory: 8462 loss: 0.0337 decode.loss_ce: 0.0205 decode.acc_seg: 99.1886 aux.loss_ce: 0.0132 aux.acc_seg: 98.7307 +04/16 09:50:57 - mmengine - INFO - Iter(train) [ 6050/160000] base_lr: 9.7130e-05 lr: 3.5911e-07 eta: 1 day, 18:46:15 time: 0.9956 data_time: 0.0043 memory: 8462 loss: 0.0309 decode.loss_ce: 0.0182 decode.acc_seg: 99.3471 aux.loss_ce: 0.0127 aux.acc_seg: 98.9792 +04/16 09:51:47 - mmengine - INFO - Iter(train) [ 6100/160000] base_lr: 9.7098e-05 lr: 3.5899e-07 eta: 1 day, 18:45:20 time: 0.9949 data_time: 0.0041 memory: 8462 loss: 0.0321 decode.loss_ce: 0.0186 decode.acc_seg: 99.1684 aux.loss_ce: 0.0135 aux.acc_seg: 98.6595 +04/16 09:52:37 - mmengine - INFO - Iter(train) [ 6150/160000] base_lr: 9.7067e-05 lr: 3.5888e-07 eta: 1 day, 18:44:25 time: 0.9968 data_time: 0.0046 memory: 8462 loss: 0.0433 decode.loss_ce: 0.0277 decode.acc_seg: 99.3948 aux.loss_ce: 0.0156 aux.acc_seg: 98.8504 +04/16 09:53:27 - mmengine - INFO - Iter(train) [ 6200/160000] base_lr: 9.7035e-05 lr: 3.5876e-07 eta: 1 day, 18:43:30 time: 0.9963 data_time: 0.0042 memory: 8462 loss: 0.0343 decode.loss_ce: 0.0209 decode.acc_seg: 99.0700 aux.loss_ce: 0.0133 aux.acc_seg: 98.5773 +04/16 09:54:16 - mmengine - INFO - Iter(train) [ 6250/160000] base_lr: 9.7004e-05 lr: 3.5864e-07 eta: 1 day, 18:42:36 time: 0.9964 data_time: 0.0050 memory: 8462 loss: 0.0350 decode.loss_ce: 0.0214 decode.acc_seg: 98.7816 aux.loss_ce: 0.0136 aux.acc_seg: 98.6498 +04/16 09:55:06 - mmengine - INFO - Iter(train) [ 6300/160000] base_lr: 9.6972e-05 lr: 3.5853e-07 eta: 1 day, 18:41:41 time: 0.9958 data_time: 0.0046 memory: 8462 loss: 0.0377 decode.loss_ce: 0.0219 decode.acc_seg: 99.1528 aux.loss_ce: 0.0157 aux.acc_seg: 98.4425 +04/16 09:55:56 - mmengine - INFO - Iter(train) [ 6350/160000] base_lr: 9.6941e-05 lr: 3.5841e-07 eta: 1 day, 18:40:46 time: 0.9952 data_time: 0.0042 memory: 8462 loss: 0.0356 decode.loss_ce: 0.0218 decode.acc_seg: 99.3008 aux.loss_ce: 0.0138 aux.acc_seg: 98.6506 +04/16 09:56:46 - mmengine - INFO - Iter(train) [ 6400/160000] base_lr: 9.6909e-05 lr: 3.5829e-07 eta: 1 day, 18:39:52 time: 0.9957 data_time: 0.0043 memory: 8462 loss: 0.0380 decode.loss_ce: 0.0230 decode.acc_seg: 99.4038 aux.loss_ce: 0.0149 aux.acc_seg: 99.0522 +04/16 09:57:36 - mmengine - INFO - Iter(train) [ 6450/160000] base_lr: 9.6878e-05 lr: 3.5818e-07 eta: 1 day, 18:38:57 time: 0.9969 data_time: 0.0045 memory: 8462 loss: 0.0304 decode.loss_ce: 0.0181 decode.acc_seg: 99.2731 aux.loss_ce: 0.0124 aux.acc_seg: 98.5628 +04/16 09:58:25 - mmengine - INFO - Iter(train) [ 6500/160000] base_lr: 9.6846e-05 lr: 3.5806e-07 eta: 1 day, 18:38:03 time: 0.9959 data_time: 0.0044 memory: 8462 loss: 0.0323 decode.loss_ce: 0.0195 decode.acc_seg: 98.5191 aux.loss_ce: 0.0128 aux.acc_seg: 97.7932 +04/16 09:59:15 - mmengine - INFO - Iter(train) [ 6550/160000] base_lr: 9.6814e-05 lr: 3.5794e-07 eta: 1 day, 18:37:09 time: 0.9971 data_time: 0.0048 memory: 8462 loss: 0.0379 decode.loss_ce: 0.0234 decode.acc_seg: 99.3090 aux.loss_ce: 0.0145 aux.acc_seg: 99.0202 +04/16 10:00:05 - mmengine - INFO - Iter(train) [ 6600/160000] base_lr: 9.6783e-05 lr: 3.5783e-07 eta: 1 day, 18:36:14 time: 0.9966 data_time: 0.0043 memory: 8462 loss: 0.0382 decode.loss_ce: 0.0229 decode.acc_seg: 99.1682 aux.loss_ce: 0.0153 aux.acc_seg: 98.7230 +04/16 10:00:55 - mmengine - INFO - Iter(train) [ 6650/160000] base_lr: 9.6751e-05 lr: 3.5771e-07 eta: 1 day, 18:35:20 time: 0.9966 data_time: 0.0043 memory: 8462 loss: 0.0407 decode.loss_ce: 0.0246 decode.acc_seg: 98.9925 aux.loss_ce: 0.0161 aux.acc_seg: 98.5542 +04/16 10:01:45 - mmengine - INFO - Iter(train) [ 6700/160000] base_lr: 9.6720e-05 lr: 3.5759e-07 eta: 1 day, 18:34:27 time: 0.9964 data_time: 0.0045 memory: 8462 loss: 0.0347 decode.loss_ce: 0.0211 decode.acc_seg: 98.9086 aux.loss_ce: 0.0136 aux.acc_seg: 98.3412 +04/16 10:02:35 - mmengine - INFO - Iter(train) [ 6750/160000] base_lr: 9.6688e-05 lr: 3.5748e-07 eta: 1 day, 18:33:33 time: 0.9980 data_time: 0.0042 memory: 8462 loss: 0.0358 decode.loss_ce: 0.0218 decode.acc_seg: 99.2535 aux.loss_ce: 0.0140 aux.acc_seg: 98.5071 +04/16 10:03:24 - mmengine - INFO - Iter(train) [ 6800/160000] base_lr: 9.6657e-05 lr: 3.5736e-07 eta: 1 day, 18:32:40 time: 0.9962 data_time: 0.0044 memory: 8462 loss: 0.0338 decode.loss_ce: 0.0205 decode.acc_seg: 99.4959 aux.loss_ce: 0.0133 aux.acc_seg: 98.9510 +04/16 10:04:14 - mmengine - INFO - Iter(train) [ 6850/160000] base_lr: 9.6625e-05 lr: 3.5724e-07 eta: 1 day, 18:31:46 time: 0.9964 data_time: 0.0042 memory: 8462 loss: 0.0336 decode.loss_ce: 0.0201 decode.acc_seg: 99.3446 aux.loss_ce: 0.0135 aux.acc_seg: 98.9294 +04/16 10:05:04 - mmengine - INFO - Iter(train) [ 6900/160000] base_lr: 9.6594e-05 lr: 3.5713e-07 eta: 1 day, 18:30:52 time: 0.9963 data_time: 0.0046 memory: 8462 loss: 0.0331 decode.loss_ce: 0.0198 decode.acc_seg: 99.5260 aux.loss_ce: 0.0133 aux.acc_seg: 98.6692 +04/16 10:05:54 - mmengine - INFO - Iter(train) [ 6950/160000] base_lr: 9.6562e-05 lr: 3.5701e-07 eta: 1 day, 18:29:59 time: 0.9967 data_time: 0.0046 memory: 8462 loss: 0.0321 decode.loss_ce: 0.0187 decode.acc_seg: 99.4892 aux.loss_ce: 0.0134 aux.acc_seg: 99.0576 +04/16 10:06:44 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 10:06:44 - mmengine - INFO - Iter(train) [ 7000/160000] base_lr: 9.6531e-05 lr: 3.5689e-07 eta: 1 day, 18:29:05 time: 0.9960 data_time: 0.0046 memory: 8462 loss: 0.0254 decode.loss_ce: 0.0143 decode.acc_seg: 99.4932 aux.loss_ce: 0.0110 aux.acc_seg: 98.8232 +04/16 10:07:34 - mmengine - INFO - Iter(train) [ 7050/160000] base_lr: 9.6499e-05 lr: 3.5678e-07 eta: 1 day, 18:28:12 time: 0.9958 data_time: 0.0041 memory: 8462 loss: 0.0332 decode.loss_ce: 0.0205 decode.acc_seg: 99.0192 aux.loss_ce: 0.0127 aux.acc_seg: 98.2471 +04/16 10:08:23 - mmengine - INFO - Iter(train) [ 7100/160000] base_lr: 9.6467e-05 lr: 3.5666e-07 eta: 1 day, 18:27:19 time: 0.9957 data_time: 0.0041 memory: 8462 loss: 0.0317 decode.loss_ce: 0.0194 decode.acc_seg: 99.5409 aux.loss_ce: 0.0123 aux.acc_seg: 98.7761 +04/16 10:09:13 - mmengine - INFO - Iter(train) [ 7150/160000] base_lr: 9.6436e-05 lr: 3.5654e-07 eta: 1 day, 18:26:26 time: 0.9973 data_time: 0.0046 memory: 8462 loss: 0.0292 decode.loss_ce: 0.0174 decode.acc_seg: 99.0015 aux.loss_ce: 0.0118 aux.acc_seg: 98.7383 +04/16 10:10:03 - mmengine - INFO - Iter(train) [ 7200/160000] base_lr: 9.6404e-05 lr: 3.5643e-07 eta: 1 day, 18:25:33 time: 0.9981 data_time: 0.0041 memory: 8462 loss: 0.0318 decode.loss_ce: 0.0184 decode.acc_seg: 99.5110 aux.loss_ce: 0.0134 aux.acc_seg: 98.5100 +04/16 10:10:53 - mmengine - INFO - Iter(train) [ 7250/160000] base_lr: 9.6373e-05 lr: 3.5631e-07 eta: 1 day, 18:24:40 time: 0.9971 data_time: 0.0042 memory: 8462 loss: 0.0427 decode.loss_ce: 0.0266 decode.acc_seg: 98.6675 aux.loss_ce: 0.0161 aux.acc_seg: 98.4175 +04/16 10:11:43 - mmengine - INFO - Iter(train) [ 7300/160000] base_lr: 9.6341e-05 lr: 3.5619e-07 eta: 1 day, 18:23:47 time: 0.9957 data_time: 0.0042 memory: 8462 loss: 0.0292 decode.loss_ce: 0.0170 decode.acc_seg: 99.3309 aux.loss_ce: 0.0122 aux.acc_seg: 98.8367 +04/16 10:12:33 - mmengine - INFO - Iter(train) [ 7350/160000] base_lr: 9.6310e-05 lr: 3.5608e-07 eta: 1 day, 18:22:54 time: 0.9966 data_time: 0.0043 memory: 8462 loss: 0.0334 decode.loss_ce: 0.0192 decode.acc_seg: 99.2157 aux.loss_ce: 0.0141 aux.acc_seg: 98.7862 +04/16 10:13:22 - mmengine - INFO - Iter(train) [ 7400/160000] base_lr: 9.6278e-05 lr: 3.5596e-07 eta: 1 day, 18:22:01 time: 0.9970 data_time: 0.0046 memory: 8462 loss: 0.0284 decode.loss_ce: 0.0166 decode.acc_seg: 99.2878 aux.loss_ce: 0.0118 aux.acc_seg: 98.8157 +04/16 10:14:12 - mmengine - INFO - Iter(train) [ 7450/160000] base_lr: 9.6247e-05 lr: 3.5584e-07 eta: 1 day, 18:21:08 time: 0.9957 data_time: 0.0043 memory: 8462 loss: 0.0394 decode.loss_ce: 0.0249 decode.acc_seg: 99.0482 aux.loss_ce: 0.0145 aux.acc_seg: 98.3685 +04/16 10:15:02 - mmengine - INFO - Iter(train) [ 7500/160000] base_lr: 9.6215e-05 lr: 3.5573e-07 eta: 1 day, 18:20:15 time: 0.9963 data_time: 0.0044 memory: 8462 loss: 0.0347 decode.loss_ce: 0.0213 decode.acc_seg: 99.2590 aux.loss_ce: 0.0134 aux.acc_seg: 99.0177 +04/16 10:15:52 - mmengine - INFO - Iter(train) [ 7550/160000] base_lr: 9.6184e-05 lr: 3.5561e-07 eta: 1 day, 18:19:21 time: 0.9945 data_time: 0.0044 memory: 8462 loss: 0.0362 decode.loss_ce: 0.0219 decode.acc_seg: 99.4312 aux.loss_ce: 0.0143 aux.acc_seg: 98.5786 +04/16 10:16:42 - mmengine - INFO - Iter(train) [ 7600/160000] base_lr: 9.6152e-05 lr: 3.5549e-07 eta: 1 day, 18:18:27 time: 0.9960 data_time: 0.0045 memory: 8462 loss: 0.0369 decode.loss_ce: 0.0225 decode.acc_seg: 99.1697 aux.loss_ce: 0.0144 aux.acc_seg: 98.7450 +04/16 10:17:31 - mmengine - INFO - Iter(train) [ 7650/160000] base_lr: 9.6120e-05 lr: 3.5538e-07 eta: 1 day, 18:17:35 time: 0.9966 data_time: 0.0046 memory: 8462 loss: 0.0293 decode.loss_ce: 0.0172 decode.acc_seg: 99.2083 aux.loss_ce: 0.0121 aux.acc_seg: 98.6141 +04/16 10:18:21 - mmengine - INFO - Iter(train) [ 7700/160000] base_lr: 9.6089e-05 lr: 3.5526e-07 eta: 1 day, 18:16:42 time: 0.9966 data_time: 0.0043 memory: 8462 loss: 0.0255 decode.loss_ce: 0.0147 decode.acc_seg: 98.9990 aux.loss_ce: 0.0108 aux.acc_seg: 98.5468 +04/16 10:19:11 - mmengine - INFO - Iter(train) [ 7750/160000] base_lr: 9.6057e-05 lr: 3.5514e-07 eta: 1 day, 18:15:49 time: 0.9962 data_time: 0.0042 memory: 8462 loss: 0.0340 decode.loss_ce: 0.0202 decode.acc_seg: 99.3233 aux.loss_ce: 0.0138 aux.acc_seg: 98.8796 +04/16 10:20:01 - mmengine - INFO - Iter(train) [ 7800/160000] base_lr: 9.6026e-05 lr: 3.5503e-07 eta: 1 day, 18:14:56 time: 0.9964 data_time: 0.0042 memory: 8462 loss: 0.0342 decode.loss_ce: 0.0212 decode.acc_seg: 99.3309 aux.loss_ce: 0.0129 aux.acc_seg: 98.4568 +04/16 10:20:51 - mmengine - INFO - Iter(train) [ 7850/160000] base_lr: 9.5994e-05 lr: 3.5491e-07 eta: 1 day, 18:14:03 time: 0.9965 data_time: 0.0046 memory: 8462 loss: 0.0345 decode.loss_ce: 0.0211 decode.acc_seg: 99.7005 aux.loss_ce: 0.0134 aux.acc_seg: 99.2964 +04/16 10:21:41 - mmengine - INFO - Iter(train) [ 7900/160000] base_lr: 9.5963e-05 lr: 3.5479e-07 eta: 1 day, 18:13:10 time: 0.9969 data_time: 0.0046 memory: 8462 loss: 0.0323 decode.loss_ce: 0.0190 decode.acc_seg: 99.2842 aux.loss_ce: 0.0132 aux.acc_seg: 98.2010 +04/16 10:22:30 - mmengine - INFO - Iter(train) [ 7950/160000] base_lr: 9.5931e-05 lr: 3.5468e-07 eta: 1 day, 18:12:18 time: 0.9965 data_time: 0.0043 memory: 8462 loss: 0.0318 decode.loss_ce: 0.0195 decode.acc_seg: 98.9491 aux.loss_ce: 0.0123 aux.acc_seg: 98.3175 +04/16 10:23:20 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 10:23:20 - mmengine - INFO - Iter(train) [ 8000/160000] base_lr: 9.5900e-05 lr: 3.5456e-07 eta: 1 day, 18:11:25 time: 0.9964 data_time: 0.0046 memory: 8462 loss: 0.0294 decode.loss_ce: 0.0170 decode.acc_seg: 99.1680 aux.loss_ce: 0.0124 aux.acc_seg: 98.7823 +04/16 10:24:10 - mmengine - INFO - Iter(train) [ 8050/160000] base_lr: 9.5868e-05 lr: 3.5444e-07 eta: 1 day, 18:10:31 time: 0.9959 data_time: 0.0045 memory: 8462 loss: 0.0313 decode.loss_ce: 0.0187 decode.acc_seg: 99.3095 aux.loss_ce: 0.0126 aux.acc_seg: 99.0952 +04/16 10:25:00 - mmengine - INFO - Iter(train) [ 8100/160000] base_lr: 9.5837e-05 lr: 3.5433e-07 eta: 1 day, 18:09:39 time: 0.9958 data_time: 0.0047 memory: 8462 loss: 0.0326 decode.loss_ce: 0.0194 decode.acc_seg: 99.3422 aux.loss_ce: 0.0132 aux.acc_seg: 99.0778 +04/16 10:25:50 - mmengine - INFO - Iter(train) [ 8150/160000] base_lr: 9.5805e-05 lr: 3.5421e-07 eta: 1 day, 18:08:46 time: 0.9956 data_time: 0.0046 memory: 8462 loss: 0.0367 decode.loss_ce: 0.0225 decode.acc_seg: 99.4356 aux.loss_ce: 0.0142 aux.acc_seg: 98.9025 +04/16 10:26:39 - mmengine - INFO - Iter(train) [ 8200/160000] base_lr: 9.5773e-05 lr: 3.5409e-07 eta: 1 day, 18:07:53 time: 0.9953 data_time: 0.0045 memory: 8462 loss: 0.0349 decode.loss_ce: 0.0208 decode.acc_seg: 99.1325 aux.loss_ce: 0.0141 aux.acc_seg: 98.4537 +04/16 10:27:29 - mmengine - INFO - Iter(train) [ 8250/160000] base_lr: 9.5742e-05 lr: 3.5398e-07 eta: 1 day, 18:07:00 time: 0.9963 data_time: 0.0043 memory: 8462 loss: 0.0361 decode.loss_ce: 0.0213 decode.acc_seg: 99.3942 aux.loss_ce: 0.0148 aux.acc_seg: 98.8081 +04/16 10:28:19 - mmengine - INFO - Iter(train) [ 8300/160000] base_lr: 9.5710e-05 lr: 3.5386e-07 eta: 1 day, 18:06:08 time: 0.9962 data_time: 0.0043 memory: 8462 loss: 0.0302 decode.loss_ce: 0.0178 decode.acc_seg: 99.3151 aux.loss_ce: 0.0124 aux.acc_seg: 98.9853 +04/16 10:29:09 - mmengine - INFO - Iter(train) [ 8350/160000] base_lr: 9.5679e-05 lr: 3.5374e-07 eta: 1 day, 18:05:15 time: 0.9968 data_time: 0.0052 memory: 8462 loss: 0.0287 decode.loss_ce: 0.0167 decode.acc_seg: 99.5903 aux.loss_ce: 0.0120 aux.acc_seg: 99.0244 +04/16 10:29:59 - mmengine - INFO - Iter(train) [ 8400/160000] base_lr: 9.5647e-05 lr: 3.5363e-07 eta: 1 day, 18:04:23 time: 0.9959 data_time: 0.0043 memory: 8462 loss: 0.0308 decode.loss_ce: 0.0186 decode.acc_seg: 99.3853 aux.loss_ce: 0.0122 aux.acc_seg: 98.7701 +04/16 10:30:48 - mmengine - INFO - Iter(train) [ 8450/160000] base_lr: 9.5616e-05 lr: 3.5351e-07 eta: 1 day, 18:03:30 time: 0.9964 data_time: 0.0045 memory: 8462 loss: 0.0365 decode.loss_ce: 0.0223 decode.acc_seg: 99.5937 aux.loss_ce: 0.0143 aux.acc_seg: 99.1161 +04/16 10:31:38 - mmengine - INFO - Iter(train) [ 8500/160000] base_lr: 9.5584e-05 lr: 3.5339e-07 eta: 1 day, 18:02:38 time: 0.9969 data_time: 0.0050 memory: 8462 loss: 0.0301 decode.loss_ce: 0.0183 decode.acc_seg: 99.2689 aux.loss_ce: 0.0118 aux.acc_seg: 98.8541 +04/16 10:32:28 - mmengine - INFO - Iter(train) [ 8550/160000] base_lr: 9.5553e-05 lr: 3.5328e-07 eta: 1 day, 18:01:45 time: 0.9951 data_time: 0.0043 memory: 8462 loss: 0.0368 decode.loss_ce: 0.0231 decode.acc_seg: 98.9004 aux.loss_ce: 0.0137 aux.acc_seg: 98.5189 +04/16 10:33:18 - mmengine - INFO - Iter(train) [ 8600/160000] base_lr: 9.5521e-05 lr: 3.5316e-07 eta: 1 day, 18:00:53 time: 0.9961 data_time: 0.0044 memory: 8462 loss: 0.0374 decode.loss_ce: 0.0226 decode.acc_seg: 99.1444 aux.loss_ce: 0.0149 aux.acc_seg: 98.4131 +04/16 10:34:08 - mmengine - INFO - Iter(train) [ 8650/160000] base_lr: 9.5490e-05 lr: 3.5304e-07 eta: 1 day, 18:00:00 time: 0.9953 data_time: 0.0046 memory: 8462 loss: 0.0342 decode.loss_ce: 0.0209 decode.acc_seg: 99.5308 aux.loss_ce: 0.0133 aux.acc_seg: 99.1673 +04/16 10:34:57 - mmengine - INFO - Iter(train) [ 8700/160000] base_lr: 9.5458e-05 lr: 3.5293e-07 eta: 1 day, 17:59:08 time: 0.9964 data_time: 0.0042 memory: 8462 loss: 0.0286 decode.loss_ce: 0.0172 decode.acc_seg: 98.9365 aux.loss_ce: 0.0114 aux.acc_seg: 98.3425 +04/16 10:35:47 - mmengine - INFO - Iter(train) [ 8750/160000] base_lr: 9.5426e-05 lr: 3.5281e-07 eta: 1 day, 17:58:15 time: 0.9969 data_time: 0.0044 memory: 8462 loss: 0.0303 decode.loss_ce: 0.0179 decode.acc_seg: 98.5941 aux.loss_ce: 0.0124 aux.acc_seg: 97.9431 +04/16 10:36:37 - mmengine - INFO - Iter(train) [ 8800/160000] base_lr: 9.5395e-05 lr: 3.5269e-07 eta: 1 day, 17:57:23 time: 0.9961 data_time: 0.0045 memory: 8462 loss: 0.0308 decode.loss_ce: 0.0181 decode.acc_seg: 99.1993 aux.loss_ce: 0.0126 aux.acc_seg: 98.6078 +04/16 10:37:27 - mmengine - INFO - Iter(train) [ 8850/160000] base_lr: 9.5363e-05 lr: 3.5258e-07 eta: 1 day, 17:56:30 time: 0.9962 data_time: 0.0047 memory: 8462 loss: 0.0228 decode.loss_ce: 0.0129 decode.acc_seg: 99.1808 aux.loss_ce: 0.0099 aux.acc_seg: 98.4518 +04/16 10:38:17 - mmengine - INFO - Iter(train) [ 8900/160000] base_lr: 9.5332e-05 lr: 3.5246e-07 eta: 1 day, 17:55:38 time: 0.9968 data_time: 0.0045 memory: 8462 loss: 0.0284 decode.loss_ce: 0.0170 decode.acc_seg: 99.2653 aux.loss_ce: 0.0114 aux.acc_seg: 98.7051 +04/16 10:39:06 - mmengine - INFO - Iter(train) [ 8950/160000] base_lr: 9.5300e-05 lr: 3.5234e-07 eta: 1 day, 17:54:45 time: 0.9955 data_time: 0.0039 memory: 8462 loss: 0.0312 decode.loss_ce: 0.0188 decode.acc_seg: 99.2916 aux.loss_ce: 0.0124 aux.acc_seg: 98.5813 +04/16 10:39:56 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 10:39:56 - mmengine - INFO - Iter(train) [ 9000/160000] base_lr: 9.5269e-05 lr: 3.5223e-07 eta: 1 day, 17:53:53 time: 0.9953 data_time: 0.0041 memory: 8462 loss: 0.0279 decode.loss_ce: 0.0161 decode.acc_seg: 99.4371 aux.loss_ce: 0.0118 aux.acc_seg: 98.8926 +04/16 10:40:46 - mmengine - INFO - Iter(train) [ 9050/160000] base_lr: 9.5237e-05 lr: 3.5211e-07 eta: 1 day, 17:53:01 time: 0.9974 data_time: 0.0047 memory: 8462 loss: 0.0281 decode.loss_ce: 0.0167 decode.acc_seg: 99.0015 aux.loss_ce: 0.0114 aux.acc_seg: 98.5495 +04/16 10:41:36 - mmengine - INFO - Iter(train) [ 9100/160000] base_lr: 9.5206e-05 lr: 3.5199e-07 eta: 1 day, 17:52:08 time: 0.9961 data_time: 0.0042 memory: 8462 loss: 0.0305 decode.loss_ce: 0.0184 decode.acc_seg: 99.4125 aux.loss_ce: 0.0121 aux.acc_seg: 98.7947 +04/16 10:42:26 - mmengine - INFO - Iter(train) [ 9150/160000] base_lr: 9.5174e-05 lr: 3.5188e-07 eta: 1 day, 17:51:16 time: 0.9959 data_time: 0.0041 memory: 8462 loss: 0.0324 decode.loss_ce: 0.0195 decode.acc_seg: 99.2474 aux.loss_ce: 0.0129 aux.acc_seg: 98.8632 +04/16 10:43:15 - mmengine - INFO - Iter(train) [ 9200/160000] base_lr: 9.5143e-05 lr: 3.5176e-07 eta: 1 day, 17:50:24 time: 0.9963 data_time: 0.0044 memory: 8462 loss: 0.0319 decode.loss_ce: 0.0186 decode.acc_seg: 99.4095 aux.loss_ce: 0.0133 aux.acc_seg: 98.9197 +04/16 10:44:05 - mmengine - INFO - Iter(train) [ 9250/160000] base_lr: 9.5111e-05 lr: 3.5164e-07 eta: 1 day, 17:49:31 time: 0.9948 data_time: 0.0041 memory: 8462 loss: 0.0284 decode.loss_ce: 0.0168 decode.acc_seg: 99.4268 aux.loss_ce: 0.0116 aux.acc_seg: 98.7555 +04/16 10:44:55 - mmengine - INFO - Iter(train) [ 9300/160000] base_lr: 9.5079e-05 lr: 3.5153e-07 eta: 1 day, 17:48:39 time: 0.9956 data_time: 0.0046 memory: 8462 loss: 0.0292 decode.loss_ce: 0.0171 decode.acc_seg: 99.3982 aux.loss_ce: 0.0121 aux.acc_seg: 99.1098 +04/16 10:45:45 - mmengine - INFO - Iter(train) [ 9350/160000] base_lr: 9.5048e-05 lr: 3.5141e-07 eta: 1 day, 17:47:47 time: 0.9962 data_time: 0.0048 memory: 8462 loss: 0.0273 decode.loss_ce: 0.0162 decode.acc_seg: 99.3837 aux.loss_ce: 0.0112 aux.acc_seg: 98.9866 +04/16 10:46:35 - mmengine - INFO - Iter(train) [ 9400/160000] base_lr: 9.5016e-05 lr: 3.5130e-07 eta: 1 day, 17:46:55 time: 0.9950 data_time: 0.0044 memory: 8462 loss: 0.0246 decode.loss_ce: 0.0146 decode.acc_seg: 99.3452 aux.loss_ce: 0.0099 aux.acc_seg: 98.9985 +04/16 10:47:24 - mmengine - INFO - Iter(train) [ 9450/160000] base_lr: 9.4985e-05 lr: 3.5118e-07 eta: 1 day, 17:46:02 time: 0.9954 data_time: 0.0045 memory: 8462 loss: 0.0284 decode.loss_ce: 0.0171 decode.acc_seg: 99.1913 aux.loss_ce: 0.0113 aux.acc_seg: 98.4587 +04/16 10:48:14 - mmengine - INFO - Iter(train) [ 9500/160000] base_lr: 9.4953e-05 lr: 3.5106e-07 eta: 1 day, 17:45:10 time: 0.9956 data_time: 0.0043 memory: 8462 loss: 0.0302 decode.loss_ce: 0.0182 decode.acc_seg: 99.3170 aux.loss_ce: 0.0121 aux.acc_seg: 99.3048 +04/16 10:49:04 - mmengine - INFO - Iter(train) [ 9550/160000] base_lr: 9.4922e-05 lr: 3.5095e-07 eta: 1 day, 17:44:18 time: 0.9964 data_time: 0.0043 memory: 8462 loss: 0.0267 decode.loss_ce: 0.0154 decode.acc_seg: 99.4465 aux.loss_ce: 0.0113 aux.acc_seg: 99.1415 +04/16 10:49:54 - mmengine - INFO - Iter(train) [ 9600/160000] base_lr: 9.4890e-05 lr: 3.5083e-07 eta: 1 day, 17:43:26 time: 0.9974 data_time: 0.0041 memory: 8462 loss: 0.0400 decode.loss_ce: 0.0244 decode.acc_seg: 99.1270 aux.loss_ce: 0.0156 aux.acc_seg: 98.4724 +04/16 10:50:44 - mmengine - INFO - Iter(train) [ 9650/160000] base_lr: 9.4859e-05 lr: 3.5071e-07 eta: 1 day, 17:42:35 time: 0.9968 data_time: 0.0047 memory: 8462 loss: 0.0286 decode.loss_ce: 0.0171 decode.acc_seg: 98.8754 aux.loss_ce: 0.0115 aux.acc_seg: 98.0198 +04/16 10:51:34 - mmengine - INFO - Iter(train) [ 9700/160000] base_lr: 9.4827e-05 lr: 3.5060e-07 eta: 1 day, 17:41:43 time: 0.9967 data_time: 0.0042 memory: 8462 loss: 0.0336 decode.loss_ce: 0.0210 decode.acc_seg: 99.1304 aux.loss_ce: 0.0125 aux.acc_seg: 98.6174 +04/16 10:52:23 - mmengine - INFO - Iter(train) [ 9750/160000] base_lr: 9.4796e-05 lr: 3.5048e-07 eta: 1 day, 17:40:52 time: 0.9964 data_time: 0.0042 memory: 8462 loss: 0.0250 decode.loss_ce: 0.0143 decode.acc_seg: 99.3227 aux.loss_ce: 0.0107 aux.acc_seg: 98.9199 +04/16 10:53:13 - mmengine - INFO - Iter(train) [ 9800/160000] base_lr: 9.4764e-05 lr: 3.5036e-07 eta: 1 day, 17:40:00 time: 0.9967 data_time: 0.0042 memory: 8462 loss: 0.0249 decode.loss_ce: 0.0142 decode.acc_seg: 99.2334 aux.loss_ce: 0.0107 aux.acc_seg: 98.5453 +04/16 10:54:03 - mmengine - INFO - Iter(train) [ 9850/160000] base_lr: 9.4732e-05 lr: 3.5025e-07 eta: 1 day, 17:39:08 time: 0.9969 data_time: 0.0045 memory: 8462 loss: 0.0288 decode.loss_ce: 0.0172 decode.acc_seg: 99.1528 aux.loss_ce: 0.0115 aux.acc_seg: 98.8901 +04/16 10:54:53 - mmengine - INFO - Iter(train) [ 9900/160000] base_lr: 9.4701e-05 lr: 3.5013e-07 eta: 1 day, 17:38:16 time: 0.9961 data_time: 0.0046 memory: 8462 loss: 0.0258 decode.loss_ce: 0.0146 decode.acc_seg: 99.2846 aux.loss_ce: 0.0112 aux.acc_seg: 98.7749 +04/16 10:55:43 - mmengine - INFO - Iter(train) [ 9950/160000] base_lr: 9.4669e-05 lr: 3.5001e-07 eta: 1 day, 17:37:24 time: 0.9963 data_time: 0.0043 memory: 8462 loss: 0.0276 decode.loss_ce: 0.0161 decode.acc_seg: 99.2397 aux.loss_ce: 0.0116 aux.acc_seg: 98.6727 +04/16 10:56:32 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 10:56:32 - mmengine - INFO - Iter(train) [ 10000/160000] base_lr: 9.4638e-05 lr: 3.4990e-07 eta: 1 day, 17:36:32 time: 0.9959 data_time: 0.0044 memory: 8462 loss: 0.0283 decode.loss_ce: 0.0167 decode.acc_seg: 98.9981 aux.loss_ce: 0.0115 aux.acc_seg: 98.5960 +04/16 10:56:32 - mmengine - INFO - Saving checkpoint at 10000 iterations +04/16 10:56:44 - mmengine - INFO - Iter(val) [ 50/355] eta: 0:00:43 time: 0.1151 data_time: 0.0014 memory: 7125 +04/16 10:56:50 - mmengine - INFO - Iter(val) [100/355] eta: 0:00:33 time: 0.1154 data_time: 0.0015 memory: 4004 +04/16 10:56:55 - mmengine - INFO - Iter(val) [150/355] eta: 0:00:25 time: 0.1157 data_time: 0.0014 memory: 4004 +04/16 10:57:01 - mmengine - INFO - Iter(val) [200/355] eta: 0:00:19 time: 0.1153 data_time: 0.0013 memory: 4004 +04/16 10:57:07 - mmengine - INFO - Iter(val) [250/355] eta: 0:00:12 time: 0.1155 data_time: 0.0015 memory: 4004 +04/16 10:57:13 - mmengine - INFO - Iter(val) [300/355] eta: 0:00:06 time: 0.1156 data_time: 0.0015 memory: 4004 +04/16 10:57:18 - mmengine - INFO - Iter(val) [350/355] eta: 0:00:00 time: 0.1153 data_time: 0.0014 memory: 4004 +04/16 10:57:19 - mmengine - INFO - per class results: +04/16 10:57:19 - mmengine - INFO - ++------------+-------+-------+ +| Class | IoU | Acc | ++------------+-------+-------+ +| background | 99.1 | 99.54 | +| contrast | 80.58 | 89.43 | ++------------+-------+-------+ +04/16 10:57:19 - mmengine - INFO - Iter(val) [355/355] aAcc: 99.1300 mIoU: 89.8400 mAcc: 94.4800 data_time: 0.0021 time: 0.1196 +04/16 10:58:09 - mmengine - INFO - Iter(train) [ 10050/160000] base_lr: 9.4606e-05 lr: 3.4978e-07 eta: 1 day, 17:35:45 time: 0.9948 data_time: 0.0042 memory: 8462 loss: 0.0320 decode.loss_ce: 0.0194 decode.acc_seg: 98.7612 aux.loss_ce: 0.0126 aux.acc_seg: 98.2853 +04/16 10:58:59 - mmengine - INFO - Iter(train) [ 10100/160000] base_lr: 9.4575e-05 lr: 3.4966e-07 eta: 1 day, 17:34:52 time: 0.9946 data_time: 0.0044 memory: 8462 loss: 0.0302 decode.loss_ce: 0.0176 decode.acc_seg: 99.2897 aux.loss_ce: 0.0126 aux.acc_seg: 98.5819 +04/16 10:59:49 - mmengine - INFO - Iter(train) [ 10150/160000] base_lr: 9.4543e-05 lr: 3.4955e-07 eta: 1 day, 17:33:59 time: 0.9944 data_time: 0.0045 memory: 8462 loss: 0.0380 decode.loss_ce: 0.0233 decode.acc_seg: 99.3277 aux.loss_ce: 0.0147 aux.acc_seg: 98.6393 +04/16 11:00:38 - mmengine - INFO - Iter(train) [ 10200/160000] base_lr: 9.4512e-05 lr: 3.4943e-07 eta: 1 day, 17:33:07 time: 0.9957 data_time: 0.0046 memory: 8462 loss: 0.0314 decode.loss_ce: 0.0186 decode.acc_seg: 99.6262 aux.loss_ce: 0.0128 aux.acc_seg: 99.1808 +04/16 11:01:28 - mmengine - INFO - Iter(train) [ 10250/160000] base_lr: 9.4480e-05 lr: 3.4931e-07 eta: 1 day, 17:32:14 time: 0.9943 data_time: 0.0044 memory: 8462 loss: 0.0305 decode.loss_ce: 0.0182 decode.acc_seg: 99.1917 aux.loss_ce: 0.0123 aux.acc_seg: 98.9431 +04/16 11:02:18 - mmengine - INFO - Iter(train) [ 10300/160000] base_lr: 9.4449e-05 lr: 3.4920e-07 eta: 1 day, 17:31:21 time: 0.9958 data_time: 0.0042 memory: 8462 loss: 0.0281 decode.loss_ce: 0.0170 decode.acc_seg: 98.8918 aux.loss_ce: 0.0111 aux.acc_seg: 98.5752 +04/16 11:03:08 - mmengine - INFO - Iter(train) [ 10350/160000] base_lr: 9.4417e-05 lr: 3.4908e-07 eta: 1 day, 17:30:28 time: 0.9944 data_time: 0.0044 memory: 8462 loss: 0.0278 decode.loss_ce: 0.0161 decode.acc_seg: 99.0158 aux.loss_ce: 0.0117 aux.acc_seg: 98.1525 +04/16 11:03:57 - mmengine - INFO - Iter(train) [ 10400/160000] base_lr: 9.4385e-05 lr: 3.4896e-07 eta: 1 day, 17:29:36 time: 0.9941 data_time: 0.0044 memory: 8462 loss: 0.0324 decode.loss_ce: 0.0188 decode.acc_seg: 98.5664 aux.loss_ce: 0.0135 aux.acc_seg: 97.8355 +04/16 11:04:47 - mmengine - INFO - Iter(train) [ 10450/160000] base_lr: 9.4354e-05 lr: 3.4885e-07 eta: 1 day, 17:28:43 time: 0.9949 data_time: 0.0043 memory: 8462 loss: 0.0288 decode.loss_ce: 0.0179 decode.acc_seg: 99.4333 aux.loss_ce: 0.0109 aux.acc_seg: 98.8943 +04/16 11:05:37 - mmengine - INFO - Iter(train) [ 10500/160000] base_lr: 9.4322e-05 lr: 3.4873e-07 eta: 1 day, 17:27:51 time: 0.9961 data_time: 0.0041 memory: 8462 loss: 0.0295 decode.loss_ce: 0.0182 decode.acc_seg: 99.4612 aux.loss_ce: 0.0113 aux.acc_seg: 99.0013 +04/16 11:06:27 - mmengine - INFO - Iter(train) [ 10550/160000] base_lr: 9.4291e-05 lr: 3.4861e-07 eta: 1 day, 17:26:58 time: 0.9948 data_time: 0.0042 memory: 8462 loss: 0.0325 decode.loss_ce: 0.0198 decode.acc_seg: 99.4640 aux.loss_ce: 0.0127 aux.acc_seg: 98.8817 +04/16 11:07:16 - mmengine - INFO - Iter(train) [ 10600/160000] base_lr: 9.4259e-05 lr: 3.4850e-07 eta: 1 day, 17:26:05 time: 0.9941 data_time: 0.0042 memory: 8462 loss: 0.0332 decode.loss_ce: 0.0205 decode.acc_seg: 99.3015 aux.loss_ce: 0.0127 aux.acc_seg: 98.8506 +04/16 11:08:06 - mmengine - INFO - Iter(train) [ 10650/160000] base_lr: 9.4228e-05 lr: 3.4838e-07 eta: 1 day, 17:25:13 time: 0.9958 data_time: 0.0044 memory: 8462 loss: 0.0316 decode.loss_ce: 0.0187 decode.acc_seg: 99.4198 aux.loss_ce: 0.0129 aux.acc_seg: 98.6265 +04/16 11:08:56 - mmengine - INFO - Iter(train) [ 10700/160000] base_lr: 9.4196e-05 lr: 3.4826e-07 eta: 1 day, 17:24:20 time: 0.9951 data_time: 0.0047 memory: 8462 loss: 0.0340 decode.loss_ce: 0.0209 decode.acc_seg: 98.6649 aux.loss_ce: 0.0131 aux.acc_seg: 98.0890 +04/16 11:09:45 - mmengine - INFO - Iter(train) [ 10750/160000] base_lr: 9.4165e-05 lr: 3.4815e-07 eta: 1 day, 17:23:28 time: 0.9944 data_time: 0.0043 memory: 8462 loss: 0.0256 decode.loss_ce: 0.0151 decode.acc_seg: 99.4286 aux.loss_ce: 0.0105 aux.acc_seg: 99.2020 +04/16 11:10:35 - mmengine - INFO - Iter(train) [ 10800/160000] base_lr: 9.4133e-05 lr: 3.4803e-07 eta: 1 day, 17:22:35 time: 0.9947 data_time: 0.0044 memory: 8462 loss: 0.0267 decode.loss_ce: 0.0156 decode.acc_seg: 99.4251 aux.loss_ce: 0.0111 aux.acc_seg: 98.9100 +04/16 11:11:25 - mmengine - INFO - Iter(train) [ 10850/160000] base_lr: 9.4102e-05 lr: 3.4791e-07 eta: 1 day, 17:21:43 time: 0.9950 data_time: 0.0045 memory: 8462 loss: 0.0330 decode.loss_ce: 0.0200 decode.acc_seg: 99.2067 aux.loss_ce: 0.0130 aux.acc_seg: 98.6345 +04/16 11:12:15 - mmengine - INFO - Iter(train) [ 10900/160000] base_lr: 9.4070e-05 lr: 3.4780e-07 eta: 1 day, 17:20:50 time: 0.9961 data_time: 0.0047 memory: 8462 loss: 0.0331 decode.loss_ce: 0.0201 decode.acc_seg: 99.0942 aux.loss_ce: 0.0130 aux.acc_seg: 98.8865 +04/16 11:13:04 - mmengine - INFO - Iter(train) [ 10950/160000] base_lr: 9.4038e-05 lr: 3.4768e-07 eta: 1 day, 17:19:58 time: 0.9935 data_time: 0.0043 memory: 8462 loss: 0.0336 decode.loss_ce: 0.0195 decode.acc_seg: 98.7394 aux.loss_ce: 0.0141 aux.acc_seg: 97.6879 +04/16 11:13:54 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 11:13:54 - mmengine - INFO - Iter(train) [ 11000/160000] base_lr: 9.4007e-05 lr: 3.4756e-07 eta: 1 day, 17:19:05 time: 0.9938 data_time: 0.0042 memory: 8462 loss: 0.0322 decode.loss_ce: 0.0194 decode.acc_seg: 99.1959 aux.loss_ce: 0.0128 aux.acc_seg: 98.8623 +04/16 11:14:44 - mmengine - INFO - Iter(train) [ 11050/160000] base_lr: 9.3975e-05 lr: 3.4745e-07 eta: 1 day, 17:18:13 time: 0.9953 data_time: 0.0046 memory: 8462 loss: 0.0272 decode.loss_ce: 0.0161 decode.acc_seg: 99.1056 aux.loss_ce: 0.0112 aux.acc_seg: 98.4341 +04/16 11:15:34 - mmengine - INFO - Iter(train) [ 11100/160000] base_lr: 9.3944e-05 lr: 3.4733e-07 eta: 1 day, 17:17:21 time: 0.9950 data_time: 0.0048 memory: 8462 loss: 0.0280 decode.loss_ce: 0.0165 decode.acc_seg: 99.5041 aux.loss_ce: 0.0114 aux.acc_seg: 99.0475 +04/16 11:16:23 - mmengine - INFO - Iter(train) [ 11150/160000] base_lr: 9.3912e-05 lr: 3.4721e-07 eta: 1 day, 17:16:28 time: 0.9954 data_time: 0.0043 memory: 8462 loss: 0.0342 decode.loss_ce: 0.0206 decode.acc_seg: 99.3380 aux.loss_ce: 0.0136 aux.acc_seg: 98.7478 +04/16 11:17:13 - mmengine - INFO - Iter(train) [ 11200/160000] base_lr: 9.3881e-05 lr: 3.4710e-07 eta: 1 day, 17:15:36 time: 0.9939 data_time: 0.0045 memory: 8462 loss: 0.0323 decode.loss_ce: 0.0196 decode.acc_seg: 98.4217 aux.loss_ce: 0.0127 aux.acc_seg: 98.0209 +04/16 11:18:03 - mmengine - INFO - Iter(train) [ 11250/160000] base_lr: 9.3849e-05 lr: 3.4698e-07 eta: 1 day, 17:14:43 time: 0.9943 data_time: 0.0043 memory: 8462 loss: 0.0288 decode.loss_ce: 0.0165 decode.acc_seg: 99.4074 aux.loss_ce: 0.0123 aux.acc_seg: 98.8914 +04/16 11:18:52 - mmengine - INFO - Iter(train) [ 11300/160000] base_lr: 9.3818e-05 lr: 3.4686e-07 eta: 1 day, 17:13:51 time: 0.9954 data_time: 0.0046 memory: 8462 loss: 0.0275 decode.loss_ce: 0.0164 decode.acc_seg: 99.5909 aux.loss_ce: 0.0111 aux.acc_seg: 99.1802 +04/16 11:19:42 - mmengine - INFO - Iter(train) [ 11350/160000] base_lr: 9.3786e-05 lr: 3.4675e-07 eta: 1 day, 17:12:58 time: 0.9951 data_time: 0.0047 memory: 8462 loss: 0.0365 decode.loss_ce: 0.0222 decode.acc_seg: 98.8888 aux.loss_ce: 0.0142 aux.acc_seg: 98.4755 +04/16 11:20:32 - mmengine - INFO - Iter(train) [ 11400/160000] base_lr: 9.3755e-05 lr: 3.4663e-07 eta: 1 day, 17:12:06 time: 0.9944 data_time: 0.0041 memory: 8462 loss: 0.0253 decode.loss_ce: 0.0143 decode.acc_seg: 99.4547 aux.loss_ce: 0.0110 aux.acc_seg: 99.2214 +04/16 11:21:22 - mmengine - INFO - Iter(train) [ 11450/160000] base_lr: 9.3723e-05 lr: 3.4651e-07 eta: 1 day, 17:11:14 time: 0.9953 data_time: 0.0045 memory: 8462 loss: 0.0300 decode.loss_ce: 0.0172 decode.acc_seg: 99.3704 aux.loss_ce: 0.0127 aux.acc_seg: 98.9031 +04/16 11:22:11 - mmengine - INFO - Iter(train) [ 11500/160000] base_lr: 9.3691e-05 lr: 3.4640e-07 eta: 1 day, 17:10:22 time: 0.9945 data_time: 0.0046 memory: 8462 loss: 0.0290 decode.loss_ce: 0.0171 decode.acc_seg: 99.1337 aux.loss_ce: 0.0119 aux.acc_seg: 98.4346 +04/16 11:23:01 - mmengine - INFO - Iter(train) [ 11550/160000] base_lr: 9.3660e-05 lr: 3.4628e-07 eta: 1 day, 17:09:29 time: 0.9945 data_time: 0.0041 memory: 8462 loss: 0.0346 decode.loss_ce: 0.0213 decode.acc_seg: 98.7728 aux.loss_ce: 0.0133 aux.acc_seg: 98.3486 +04/16 11:23:51 - mmengine - INFO - Iter(train) [ 11600/160000] base_lr: 9.3628e-05 lr: 3.4616e-07 eta: 1 day, 17:08:37 time: 0.9948 data_time: 0.0043 memory: 8462 loss: 0.0306 decode.loss_ce: 0.0183 decode.acc_seg: 99.3786 aux.loss_ce: 0.0122 aux.acc_seg: 98.8174 +04/16 11:24:41 - mmengine - INFO - Iter(train) [ 11650/160000] base_lr: 9.3597e-05 lr: 3.4605e-07 eta: 1 day, 17:07:45 time: 0.9939 data_time: 0.0041 memory: 8462 loss: 0.0272 decode.loss_ce: 0.0166 decode.acc_seg: 98.9901 aux.loss_ce: 0.0106 aux.acc_seg: 98.5510 +04/16 11:25:30 - mmengine - INFO - Iter(train) [ 11700/160000] base_lr: 9.3565e-05 lr: 3.4593e-07 eta: 1 day, 17:06:53 time: 0.9942 data_time: 0.0043 memory: 8462 loss: 0.0282 decode.loss_ce: 0.0169 decode.acc_seg: 98.5504 aux.loss_ce: 0.0113 aux.acc_seg: 98.2452 +04/16 11:26:20 - mmengine - INFO - Iter(train) [ 11750/160000] base_lr: 9.3534e-05 lr: 3.4581e-07 eta: 1 day, 17:06:01 time: 0.9935 data_time: 0.0044 memory: 8462 loss: 0.0272 decode.loss_ce: 0.0155 decode.acc_seg: 99.5529 aux.loss_ce: 0.0117 aux.acc_seg: 98.8819 +04/16 11:27:10 - mmengine - INFO - Iter(train) [ 11800/160000] base_lr: 9.3502e-05 lr: 3.4570e-07 eta: 1 day, 17:05:08 time: 0.9942 data_time: 0.0051 memory: 8462 loss: 0.0256 decode.loss_ce: 0.0150 decode.acc_seg: 99.4202 aux.loss_ce: 0.0105 aux.acc_seg: 98.8888 +04/16 11:27:59 - mmengine - INFO - Iter(train) [ 11850/160000] base_lr: 9.3471e-05 lr: 3.4558e-07 eta: 1 day, 17:04:16 time: 0.9937 data_time: 0.0043 memory: 8462 loss: 0.0357 decode.loss_ce: 0.0216 decode.acc_seg: 99.3042 aux.loss_ce: 0.0141 aux.acc_seg: 98.7051 +04/16 11:28:49 - mmengine - INFO - Iter(train) [ 11900/160000] base_lr: 9.3439e-05 lr: 3.4546e-07 eta: 1 day, 17:03:24 time: 0.9942 data_time: 0.0044 memory: 8462 loss: 0.0263 decode.loss_ce: 0.0155 decode.acc_seg: 99.2916 aux.loss_ce: 0.0108 aux.acc_seg: 98.7413 +04/16 11:29:39 - mmengine - INFO - Iter(train) [ 11950/160000] base_lr: 9.3408e-05 lr: 3.4535e-07 eta: 1 day, 17:02:32 time: 0.9944 data_time: 0.0045 memory: 8462 loss: 0.0297 decode.loss_ce: 0.0176 decode.acc_seg: 99.4255 aux.loss_ce: 0.0121 aux.acc_seg: 98.8607 +04/16 11:30:29 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 11:30:29 - mmengine - INFO - Iter(train) [ 12000/160000] base_lr: 9.3376e-05 lr: 3.4523e-07 eta: 1 day, 17:01:39 time: 0.9935 data_time: 0.0042 memory: 8462 loss: 0.0265 decode.loss_ce: 0.0156 decode.acc_seg: 99.4846 aux.loss_ce: 0.0108 aux.acc_seg: 98.9838 +04/16 11:31:18 - mmengine - INFO - Iter(train) [ 12050/160000] base_lr: 9.3344e-05 lr: 3.4511e-07 eta: 1 day, 17:00:47 time: 0.9943 data_time: 0.0046 memory: 8462 loss: 0.0276 decode.loss_ce: 0.0161 decode.acc_seg: 99.1394 aux.loss_ce: 0.0115 aux.acc_seg: 98.3816 +04/16 11:32:08 - mmengine - INFO - Iter(train) [ 12100/160000] base_lr: 9.3313e-05 lr: 3.4500e-07 eta: 1 day, 16:59:55 time: 0.9947 data_time: 0.0047 memory: 8462 loss: 0.0287 decode.loss_ce: 0.0165 decode.acc_seg: 98.9498 aux.loss_ce: 0.0122 aux.acc_seg: 98.0978 +04/16 11:32:58 - mmengine - INFO - Iter(train) [ 12150/160000] base_lr: 9.3281e-05 lr: 3.4488e-07 eta: 1 day, 16:59:03 time: 0.9948 data_time: 0.0044 memory: 8462 loss: 0.0307 decode.loss_ce: 0.0181 decode.acc_seg: 99.2891 aux.loss_ce: 0.0126 aux.acc_seg: 98.4135 +04/16 11:33:47 - mmengine - INFO - Iter(train) [ 12200/160000] base_lr: 9.3250e-05 lr: 3.4476e-07 eta: 1 day, 16:58:11 time: 0.9930 data_time: 0.0043 memory: 8462 loss: 0.0251 decode.loss_ce: 0.0144 decode.acc_seg: 99.4017 aux.loss_ce: 0.0107 aux.acc_seg: 98.8657 +04/16 11:34:37 - mmengine - INFO - Iter(train) [ 12250/160000] base_lr: 9.3218e-05 lr: 3.4465e-07 eta: 1 day, 16:57:19 time: 0.9949 data_time: 0.0044 memory: 8462 loss: 0.0304 decode.loss_ce: 0.0180 decode.acc_seg: 99.2184 aux.loss_ce: 0.0124 aux.acc_seg: 98.4539 +04/16 11:35:27 - mmengine - INFO - Iter(train) [ 12300/160000] base_lr: 9.3187e-05 lr: 3.4453e-07 eta: 1 day, 16:56:27 time: 0.9947 data_time: 0.0047 memory: 8462 loss: 0.0266 decode.loss_ce: 0.0162 decode.acc_seg: 99.5399 aux.loss_ce: 0.0104 aux.acc_seg: 99.2146 +04/16 11:36:17 - mmengine - INFO - Iter(train) [ 12350/160000] base_lr: 9.3155e-05 lr: 3.4441e-07 eta: 1 day, 16:55:35 time: 0.9950 data_time: 0.0052 memory: 8462 loss: 0.0242 decode.loss_ce: 0.0142 decode.acc_seg: 99.5495 aux.loss_ce: 0.0101 aux.acc_seg: 99.3206 +04/16 11:37:06 - mmengine - INFO - Iter(train) [ 12400/160000] base_lr: 9.3124e-05 lr: 3.4430e-07 eta: 1 day, 16:54:43 time: 0.9944 data_time: 0.0044 memory: 8462 loss: 0.0243 decode.loss_ce: 0.0144 decode.acc_seg: 99.3097 aux.loss_ce: 0.0099 aux.acc_seg: 98.9460 +04/16 11:37:56 - mmengine - INFO - Iter(train) [ 12450/160000] base_lr: 9.3092e-05 lr: 3.4418e-07 eta: 1 day, 16:53:51 time: 0.9944 data_time: 0.0046 memory: 8462 loss: 0.0290 decode.loss_ce: 0.0171 decode.acc_seg: 99.2306 aux.loss_ce: 0.0119 aux.acc_seg: 98.2155 +04/16 11:38:46 - mmengine - INFO - Iter(train) [ 12500/160000] base_lr: 9.3061e-05 lr: 3.4406e-07 eta: 1 day, 16:52:59 time: 0.9961 data_time: 0.0047 memory: 8462 loss: 0.0296 decode.loss_ce: 0.0176 decode.acc_seg: 99.5613 aux.loss_ce: 0.0120 aux.acc_seg: 99.1644 +04/16 11:39:35 - mmengine - INFO - Iter(train) [ 12550/160000] base_lr: 9.3029e-05 lr: 3.4395e-07 eta: 1 day, 16:52:07 time: 0.9937 data_time: 0.0050 memory: 8462 loss: 0.0256 decode.loss_ce: 0.0157 decode.acc_seg: 98.9496 aux.loss_ce: 0.0099 aux.acc_seg: 98.8653 +04/16 11:40:25 - mmengine - INFO - Iter(train) [ 12600/160000] base_lr: 9.2997e-05 lr: 3.4383e-07 eta: 1 day, 16:51:15 time: 0.9946 data_time: 0.0047 memory: 8462 loss: 0.0275 decode.loss_ce: 0.0163 decode.acc_seg: 99.4997 aux.loss_ce: 0.0112 aux.acc_seg: 99.2428 +04/16 11:41:15 - mmengine - INFO - Iter(train) [ 12650/160000] base_lr: 9.2966e-05 lr: 3.4371e-07 eta: 1 day, 16:50:23 time: 0.9950 data_time: 0.0044 memory: 8462 loss: 0.0287 decode.loss_ce: 0.0168 decode.acc_seg: 99.2878 aux.loss_ce: 0.0118 aux.acc_seg: 98.7865 +04/16 11:42:05 - mmengine - INFO - Iter(train) [ 12700/160000] base_lr: 9.2934e-05 lr: 3.4360e-07 eta: 1 day, 16:49:31 time: 0.9945 data_time: 0.0044 memory: 8462 loss: 0.0293 decode.loss_ce: 0.0178 decode.acc_seg: 99.4368 aux.loss_ce: 0.0115 aux.acc_seg: 99.0410 +04/16 11:42:54 - mmengine - INFO - Iter(train) [ 12750/160000] base_lr: 9.2903e-05 lr: 3.4348e-07 eta: 1 day, 16:48:39 time: 0.9947 data_time: 0.0046 memory: 8462 loss: 0.0268 decode.loss_ce: 0.0154 decode.acc_seg: 99.3006 aux.loss_ce: 0.0115 aux.acc_seg: 98.5382 +04/16 11:43:44 - mmengine - INFO - Iter(train) [ 12800/160000] base_lr: 9.2871e-05 lr: 3.4336e-07 eta: 1 day, 16:47:47 time: 0.9946 data_time: 0.0047 memory: 8462 loss: 0.0293 decode.loss_ce: 0.0174 decode.acc_seg: 99.5275 aux.loss_ce: 0.0119 aux.acc_seg: 98.9567 +04/16 11:44:34 - mmengine - INFO - Iter(train) [ 12850/160000] base_lr: 9.2840e-05 lr: 3.4325e-07 eta: 1 day, 16:46:56 time: 0.9946 data_time: 0.0047 memory: 8462 loss: 0.0252 decode.loss_ce: 0.0146 decode.acc_seg: 99.4595 aux.loss_ce: 0.0106 aux.acc_seg: 99.2319 +04/16 11:45:24 - mmengine - INFO - Iter(train) [ 12900/160000] base_lr: 9.2808e-05 lr: 3.4313e-07 eta: 1 day, 16:46:04 time: 0.9946 data_time: 0.0043 memory: 8462 loss: 0.0278 decode.loss_ce: 0.0164 decode.acc_seg: 99.1468 aux.loss_ce: 0.0113 aux.acc_seg: 98.7642 +04/16 11:46:13 - mmengine - INFO - Iter(train) [ 12950/160000] base_lr: 9.2777e-05 lr: 3.4301e-07 eta: 1 day, 16:45:12 time: 0.9943 data_time: 0.0042 memory: 8462 loss: 0.0248 decode.loss_ce: 0.0147 decode.acc_seg: 99.4877 aux.loss_ce: 0.0101 aux.acc_seg: 99.1333 +04/16 11:47:03 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 11:47:03 - mmengine - INFO - Iter(train) [ 13000/160000] base_lr: 9.2745e-05 lr: 3.4290e-07 eta: 1 day, 16:44:21 time: 0.9950 data_time: 0.0043 memory: 8462 loss: 0.0273 decode.loss_ce: 0.0157 decode.acc_seg: 99.0637 aux.loss_ce: 0.0116 aux.acc_seg: 98.3202 +04/16 11:47:53 - mmengine - INFO - Iter(train) [ 13050/160000] base_lr: 9.2714e-05 lr: 3.4278e-07 eta: 1 day, 16:43:29 time: 0.9945 data_time: 0.0042 memory: 8462 loss: 0.0249 decode.loss_ce: 0.0142 decode.acc_seg: 99.5186 aux.loss_ce: 0.0108 aux.acc_seg: 98.9353 +04/16 11:48:42 - mmengine - INFO - Iter(train) [ 13100/160000] base_lr: 9.2682e-05 lr: 3.4266e-07 eta: 1 day, 16:42:38 time: 0.9956 data_time: 0.0051 memory: 8462 loss: 0.0273 decode.loss_ce: 0.0160 decode.acc_seg: 98.8503 aux.loss_ce: 0.0113 aux.acc_seg: 98.3902 +04/16 11:49:32 - mmengine - INFO - Iter(train) [ 13150/160000] base_lr: 9.2650e-05 lr: 3.4255e-07 eta: 1 day, 16:41:46 time: 0.9939 data_time: 0.0043 memory: 8462 loss: 0.0238 decode.loss_ce: 0.0137 decode.acc_seg: 99.2641 aux.loss_ce: 0.0101 aux.acc_seg: 98.5931 +04/16 11:50:22 - mmengine - INFO - Iter(train) [ 13200/160000] base_lr: 9.2619e-05 lr: 3.4243e-07 eta: 1 day, 16:40:54 time: 0.9940 data_time: 0.0044 memory: 8462 loss: 0.0275 decode.loss_ce: 0.0168 decode.acc_seg: 99.5317 aux.loss_ce: 0.0106 aux.acc_seg: 99.2304 +04/16 11:51:12 - mmengine - INFO - Iter(train) [ 13250/160000] base_lr: 9.2587e-05 lr: 3.4231e-07 eta: 1 day, 16:40:03 time: 0.9957 data_time: 0.0048 memory: 8462 loss: 0.0237 decode.loss_ce: 0.0135 decode.acc_seg: 99.4019 aux.loss_ce: 0.0102 aux.acc_seg: 99.0068 +04/16 11:52:01 - mmengine - INFO - Iter(train) [ 13300/160000] base_lr: 9.2556e-05 lr: 3.4220e-07 eta: 1 day, 16:39:11 time: 0.9942 data_time: 0.0041 memory: 8462 loss: 0.0237 decode.loss_ce: 0.0134 decode.acc_seg: 99.4535 aux.loss_ce: 0.0103 aux.acc_seg: 99.1028 +04/16 11:52:51 - mmengine - INFO - Iter(train) [ 13350/160000] base_lr: 9.2524e-05 lr: 3.4208e-07 eta: 1 day, 16:38:20 time: 0.9946 data_time: 0.0043 memory: 8462 loss: 0.0283 decode.loss_ce: 0.0164 decode.acc_seg: 99.1343 aux.loss_ce: 0.0119 aux.acc_seg: 98.5346 +04/16 11:53:41 - mmengine - INFO - Iter(train) [ 13400/160000] base_lr: 9.2493e-05 lr: 3.4196e-07 eta: 1 day, 16:37:28 time: 0.9935 data_time: 0.0045 memory: 8462 loss: 0.0240 decode.loss_ce: 0.0137 decode.acc_seg: 99.4165 aux.loss_ce: 0.0103 aux.acc_seg: 98.7953 +04/16 11:54:31 - mmengine - INFO - Iter(train) [ 13450/160000] base_lr: 9.2461e-05 lr: 3.4185e-07 eta: 1 day, 16:36:36 time: 0.9943 data_time: 0.0047 memory: 8462 loss: 0.0217 decode.loss_ce: 0.0126 decode.acc_seg: 99.5184 aux.loss_ce: 0.0091 aux.acc_seg: 99.1095 +04/16 11:55:20 - mmengine - INFO - Iter(train) [ 13500/160000] base_lr: 9.2430e-05 lr: 3.4173e-07 eta: 1 day, 16:35:45 time: 0.9945 data_time: 0.0047 memory: 8462 loss: 0.0259 decode.loss_ce: 0.0152 decode.acc_seg: 99.6181 aux.loss_ce: 0.0107 aux.acc_seg: 99.2376 +04/16 11:56:10 - mmengine - INFO - Iter(train) [ 13550/160000] base_lr: 9.2398e-05 lr: 3.4161e-07 eta: 1 day, 16:34:53 time: 0.9947 data_time: 0.0050 memory: 8462 loss: 0.0275 decode.loss_ce: 0.0159 decode.acc_seg: 99.5602 aux.loss_ce: 0.0116 aux.acc_seg: 99.0578 +04/16 11:57:00 - mmengine - INFO - Iter(train) [ 13600/160000] base_lr: 9.2367e-05 lr: 3.4150e-07 eta: 1 day, 16:34:02 time: 0.9954 data_time: 0.0044 memory: 8462 loss: 0.0230 decode.loss_ce: 0.0135 decode.acc_seg: 99.0862 aux.loss_ce: 0.0094 aux.acc_seg: 98.6214 +04/16 11:57:49 - mmengine - INFO - Iter(train) [ 13650/160000] base_lr: 9.2335e-05 lr: 3.4138e-07 eta: 1 day, 16:33:10 time: 0.9946 data_time: 0.0043 memory: 8462 loss: 0.0286 decode.loss_ce: 0.0169 decode.acc_seg: 98.9897 aux.loss_ce: 0.0117 aux.acc_seg: 98.2307 +04/16 11:58:39 - mmengine - INFO - Iter(train) [ 13700/160000] base_lr: 9.2303e-05 lr: 3.4126e-07 eta: 1 day, 16:32:18 time: 0.9947 data_time: 0.0049 memory: 8462 loss: 0.0233 decode.loss_ce: 0.0135 decode.acc_seg: 99.2607 aux.loss_ce: 0.0099 aux.acc_seg: 98.7852 +04/16 11:59:29 - mmengine - INFO - Iter(train) [ 13750/160000] base_lr: 9.2272e-05 lr: 3.4115e-07 eta: 1 day, 16:31:27 time: 0.9950 data_time: 0.0045 memory: 8462 loss: 0.0214 decode.loss_ce: 0.0119 decode.acc_seg: 99.5897 aux.loss_ce: 0.0095 aux.acc_seg: 99.0618 +04/16 12:00:19 - mmengine - INFO - Iter(train) [ 13800/160000] base_lr: 9.2240e-05 lr: 3.4103e-07 eta: 1 day, 16:30:35 time: 0.9938 data_time: 0.0044 memory: 8462 loss: 0.0342 decode.loss_ce: 0.0200 decode.acc_seg: 99.3093 aux.loss_ce: 0.0143 aux.acc_seg: 98.5952 +04/16 12:01:08 - mmengine - INFO - Iter(train) [ 13850/160000] base_lr: 9.2209e-05 lr: 3.4091e-07 eta: 1 day, 16:29:43 time: 0.9951 data_time: 0.0046 memory: 8462 loss: 0.0277 decode.loss_ce: 0.0163 decode.acc_seg: 99.1318 aux.loss_ce: 0.0114 aux.acc_seg: 98.6946 +04/16 12:01:58 - mmengine - INFO - Iter(train) [ 13900/160000] base_lr: 9.2177e-05 lr: 3.4080e-07 eta: 1 day, 16:28:52 time: 0.9937 data_time: 0.0044 memory: 8462 loss: 0.0281 decode.loss_ce: 0.0166 decode.acc_seg: 99.5358 aux.loss_ce: 0.0115 aux.acc_seg: 99.1428 +04/16 12:02:48 - mmengine - INFO - Iter(train) [ 13950/160000] base_lr: 9.2146e-05 lr: 3.4068e-07 eta: 1 day, 16:28:00 time: 0.9948 data_time: 0.0042 memory: 8462 loss: 0.0272 decode.loss_ce: 0.0156 decode.acc_seg: 99.3177 aux.loss_ce: 0.0116 aux.acc_seg: 98.7206 +04/16 12:03:37 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 12:03:37 - mmengine - INFO - Iter(train) [ 14000/160000] base_lr: 9.2114e-05 lr: 3.4056e-07 eta: 1 day, 16:27:09 time: 0.9954 data_time: 0.0040 memory: 8462 loss: 0.0240 decode.loss_ce: 0.0141 decode.acc_seg: 99.4854 aux.loss_ce: 0.0099 aux.acc_seg: 99.0868 +04/16 12:04:27 - mmengine - INFO - Iter(train) [ 14050/160000] base_lr: 9.2083e-05 lr: 3.4045e-07 eta: 1 day, 16:26:18 time: 0.9941 data_time: 0.0044 memory: 8462 loss: 0.0268 decode.loss_ce: 0.0153 decode.acc_seg: 99.2085 aux.loss_ce: 0.0115 aux.acc_seg: 98.5888 +04/16 12:05:17 - mmengine - INFO - Iter(train) [ 14100/160000] base_lr: 9.2051e-05 lr: 3.4033e-07 eta: 1 day, 16:25:27 time: 0.9947 data_time: 0.0044 memory: 8462 loss: 0.0273 decode.loss_ce: 0.0160 decode.acc_seg: 98.7940 aux.loss_ce: 0.0113 aux.acc_seg: 97.9073 +04/16 12:06:07 - mmengine - INFO - Iter(train) [ 14150/160000] base_lr: 9.2020e-05 lr: 3.4022e-07 eta: 1 day, 16:24:36 time: 0.9947 data_time: 0.0045 memory: 8462 loss: 0.0269 decode.loss_ce: 0.0162 decode.acc_seg: 99.2874 aux.loss_ce: 0.0108 aux.acc_seg: 98.6807 +04/16 12:06:56 - mmengine - INFO - Iter(train) [ 14200/160000] base_lr: 9.1988e-05 lr: 3.4010e-07 eta: 1 day, 16:23:44 time: 0.9949 data_time: 0.0043 memory: 8462 loss: 0.0257 decode.loss_ce: 0.0147 decode.acc_seg: 99.4144 aux.loss_ce: 0.0109 aux.acc_seg: 98.7972 +04/16 12:07:46 - mmengine - INFO - Iter(train) [ 14250/160000] base_lr: 9.1956e-05 lr: 3.3998e-07 eta: 1 day, 16:22:53 time: 0.9943 data_time: 0.0041 memory: 8462 loss: 0.0231 decode.loss_ce: 0.0131 decode.acc_seg: 99.4913 aux.loss_ce: 0.0100 aux.acc_seg: 99.1066 +04/16 12:08:36 - mmengine - INFO - Iter(train) [ 14300/160000] base_lr: 9.1925e-05 lr: 3.3987e-07 eta: 1 day, 16:22:02 time: 0.9946 data_time: 0.0044 memory: 8462 loss: 0.0310 decode.loss_ce: 0.0182 decode.acc_seg: 98.8054 aux.loss_ce: 0.0129 aux.acc_seg: 98.5731 +04/16 12:09:26 - mmengine - INFO - Iter(train) [ 14350/160000] base_lr: 9.1893e-05 lr: 3.3975e-07 eta: 1 day, 16:21:10 time: 0.9946 data_time: 0.0042 memory: 8462 loss: 0.0281 decode.loss_ce: 0.0166 decode.acc_seg: 99.5039 aux.loss_ce: 0.0115 aux.acc_seg: 98.9126 +04/16 12:10:15 - mmengine - INFO - Iter(train) [ 14400/160000] base_lr: 9.1862e-05 lr: 3.3963e-07 eta: 1 day, 16:20:19 time: 0.9938 data_time: 0.0045 memory: 8462 loss: 0.0256 decode.loss_ce: 0.0146 decode.acc_seg: 99.3481 aux.loss_ce: 0.0110 aux.acc_seg: 98.9779 +04/16 12:11:05 - mmengine - INFO - Iter(train) [ 14450/160000] base_lr: 9.1830e-05 lr: 3.3952e-07 eta: 1 day, 16:19:28 time: 0.9941 data_time: 0.0048 memory: 8462 loss: 0.0207 decode.loss_ce: 0.0115 decode.acc_seg: 99.1724 aux.loss_ce: 0.0091 aux.acc_seg: 98.6012 +04/16 12:11:55 - mmengine - INFO - Iter(train) [ 14500/160000] base_lr: 9.1799e-05 lr: 3.3940e-07 eta: 1 day, 16:18:37 time: 0.9937 data_time: 0.0042 memory: 8462 loss: 0.0259 decode.loss_ce: 0.0149 decode.acc_seg: 99.7185 aux.loss_ce: 0.0110 aux.acc_seg: 99.2022 +04/16 12:12:45 - mmengine - INFO - Iter(train) [ 14550/160000] base_lr: 9.1767e-05 lr: 3.3928e-07 eta: 1 day, 16:17:45 time: 0.9941 data_time: 0.0046 memory: 8462 loss: 0.0257 decode.loss_ce: 0.0151 decode.acc_seg: 99.2897 aux.loss_ce: 0.0106 aux.acc_seg: 98.7783 +04/16 12:13:34 - mmengine - INFO - Iter(train) [ 14600/160000] base_lr: 9.1736e-05 lr: 3.3917e-07 eta: 1 day, 16:16:54 time: 0.9946 data_time: 0.0045 memory: 8462 loss: 0.0285 decode.loss_ce: 0.0163 decode.acc_seg: 99.3795 aux.loss_ce: 0.0122 aux.acc_seg: 98.6528 +04/16 12:14:24 - mmengine - INFO - Iter(train) [ 14650/160000] base_lr: 9.1704e-05 lr: 3.3905e-07 eta: 1 day, 16:16:03 time: 0.9949 data_time: 0.0046 memory: 8462 loss: 0.0238 decode.loss_ce: 0.0137 decode.acc_seg: 99.6058 aux.loss_ce: 0.0101 aux.acc_seg: 99.0711 +04/16 12:15:14 - mmengine - INFO - Iter(train) [ 14700/160000] base_lr: 9.1673e-05 lr: 3.3893e-07 eta: 1 day, 16:15:12 time: 0.9943 data_time: 0.0043 memory: 8462 loss: 0.0237 decode.loss_ce: 0.0133 decode.acc_seg: 99.6855 aux.loss_ce: 0.0103 aux.acc_seg: 99.2374 +04/16 12:16:04 - mmengine - INFO - Iter(train) [ 14750/160000] base_lr: 9.1641e-05 lr: 3.3882e-07 eta: 1 day, 16:14:21 time: 0.9943 data_time: 0.0048 memory: 8462 loss: 0.0272 decode.loss_ce: 0.0160 decode.acc_seg: 99.1642 aux.loss_ce: 0.0111 aux.acc_seg: 98.5998 +04/16 12:16:53 - mmengine - INFO - Iter(train) [ 14800/160000] base_lr: 9.1609e-05 lr: 3.3870e-07 eta: 1 day, 16:13:29 time: 0.9947 data_time: 0.0049 memory: 8462 loss: 0.0255 decode.loss_ce: 0.0145 decode.acc_seg: 99.4167 aux.loss_ce: 0.0110 aux.acc_seg: 98.8081 +04/16 12:17:43 - mmengine - INFO - Iter(train) [ 14850/160000] base_lr: 9.1578e-05 lr: 3.3858e-07 eta: 1 day, 16:12:38 time: 0.9937 data_time: 0.0048 memory: 8462 loss: 0.0204 decode.loss_ce: 0.0116 decode.acc_seg: 99.6367 aux.loss_ce: 0.0088 aux.acc_seg: 99.2485 +04/16 12:18:33 - mmengine - INFO - Iter(train) [ 14900/160000] base_lr: 9.1546e-05 lr: 3.3847e-07 eta: 1 day, 16:11:46 time: 0.9945 data_time: 0.0043 memory: 8462 loss: 0.0241 decode.loss_ce: 0.0134 decode.acc_seg: 99.7013 aux.loss_ce: 0.0106 aux.acc_seg: 99.1682 +04/16 12:19:22 - mmengine - INFO - Iter(train) [ 14950/160000] base_lr: 9.1515e-05 lr: 3.3835e-07 eta: 1 day, 16:10:55 time: 0.9937 data_time: 0.0043 memory: 8462 loss: 0.0234 decode.loss_ce: 0.0134 decode.acc_seg: 99.4577 aux.loss_ce: 0.0100 aux.acc_seg: 98.9719 +04/16 12:20:12 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 12:20:12 - mmengine - INFO - Iter(train) [ 15000/160000] base_lr: 9.1483e-05 lr: 3.3823e-07 eta: 1 day, 16:10:03 time: 0.9944 data_time: 0.0045 memory: 8462 loss: 0.0269 decode.loss_ce: 0.0150 decode.acc_seg: 99.3587 aux.loss_ce: 0.0120 aux.acc_seg: 98.9525 +04/16 12:21:02 - mmengine - INFO - Iter(train) [ 15050/160000] base_lr: 9.1452e-05 lr: 3.3812e-07 eta: 1 day, 16:09:12 time: 0.9936 data_time: 0.0046 memory: 8462 loss: 0.0275 decode.loss_ce: 0.0164 decode.acc_seg: 98.5355 aux.loss_ce: 0.0111 aux.acc_seg: 98.1300 +04/16 12:21:51 - mmengine - INFO - Iter(train) [ 15100/160000] base_lr: 9.1420e-05 lr: 3.3800e-07 eta: 1 day, 16:08:20 time: 0.9952 data_time: 0.0045 memory: 8462 loss: 0.0267 decode.loss_ce: 0.0151 decode.acc_seg: 99.1245 aux.loss_ce: 0.0117 aux.acc_seg: 98.4695 +04/16 12:22:41 - mmengine - INFO - Iter(train) [ 15150/160000] base_lr: 9.1389e-05 lr: 3.3788e-07 eta: 1 day, 16:07:29 time: 0.9931 data_time: 0.0042 memory: 8462 loss: 0.0282 decode.loss_ce: 0.0168 decode.acc_seg: 99.2170 aux.loss_ce: 0.0115 aux.acc_seg: 98.6338 +04/16 12:23:31 - mmengine - INFO - Iter(train) [ 15200/160000] base_lr: 9.1357e-05 lr: 3.3777e-07 eta: 1 day, 16:06:37 time: 0.9934 data_time: 0.0043 memory: 8462 loss: 0.0281 decode.loss_ce: 0.0162 decode.acc_seg: 99.2514 aux.loss_ce: 0.0118 aux.acc_seg: 98.8026 +04/16 12:24:21 - mmengine - INFO - Iter(train) [ 15250/160000] base_lr: 9.1325e-05 lr: 3.3765e-07 eta: 1 day, 16:05:46 time: 0.9944 data_time: 0.0042 memory: 8462 loss: 0.0270 decode.loss_ce: 0.0156 decode.acc_seg: 99.1510 aux.loss_ce: 0.0114 aux.acc_seg: 98.5109 +04/16 12:25:10 - mmengine - INFO - Iter(train) [ 15300/160000] base_lr: 9.1294e-05 lr: 3.3753e-07 eta: 1 day, 16:04:55 time: 0.9937 data_time: 0.0043 memory: 8462 loss: 0.0269 decode.loss_ce: 0.0160 decode.acc_seg: 99.3240 aux.loss_ce: 0.0110 aux.acc_seg: 98.5308 +04/16 12:26:00 - mmengine - INFO - Iter(train) [ 15350/160000] base_lr: 9.1262e-05 lr: 3.3742e-07 eta: 1 day, 16:04:04 time: 0.9943 data_time: 0.0043 memory: 8462 loss: 0.0271 decode.loss_ce: 0.0154 decode.acc_seg: 99.4852 aux.loss_ce: 0.0117 aux.acc_seg: 98.7577 +04/16 12:26:50 - mmengine - INFO - Iter(train) [ 15400/160000] base_lr: 9.1231e-05 lr: 3.3730e-07 eta: 1 day, 16:03:13 time: 0.9947 data_time: 0.0046 memory: 8462 loss: 0.0249 decode.loss_ce: 0.0147 decode.acc_seg: 99.4936 aux.loss_ce: 0.0102 aux.acc_seg: 99.1734 +04/16 12:27:39 - mmengine - INFO - Iter(train) [ 15450/160000] base_lr: 9.1199e-05 lr: 3.3718e-07 eta: 1 day, 16:02:22 time: 0.9952 data_time: 0.0047 memory: 8462 loss: 0.0233 decode.loss_ce: 0.0139 decode.acc_seg: 99.0118 aux.loss_ce: 0.0094 aux.acc_seg: 98.8165 +04/16 12:28:29 - mmengine - INFO - Iter(train) [ 15500/160000] base_lr: 9.1168e-05 lr: 3.3707e-07 eta: 1 day, 16:01:30 time: 0.9940 data_time: 0.0045 memory: 8462 loss: 0.0258 decode.loss_ce: 0.0150 decode.acc_seg: 99.4658 aux.loss_ce: 0.0108 aux.acc_seg: 98.7791 +04/16 12:29:19 - mmengine - INFO - Iter(train) [ 15550/160000] base_lr: 9.1136e-05 lr: 3.3695e-07 eta: 1 day, 16:00:39 time: 0.9945 data_time: 0.0044 memory: 8462 loss: 0.0274 decode.loss_ce: 0.0157 decode.acc_seg: 98.7837 aux.loss_ce: 0.0116 aux.acc_seg: 97.6768 +04/16 12:30:09 - mmengine - INFO - Iter(train) [ 15600/160000] base_lr: 9.1105e-05 lr: 3.3683e-07 eta: 1 day, 15:59:48 time: 0.9941 data_time: 0.0043 memory: 8462 loss: 0.0265 decode.loss_ce: 0.0149 decode.acc_seg: 99.6019 aux.loss_ce: 0.0116 aux.acc_seg: 98.9788 +04/16 12:30:58 - mmengine - INFO - Iter(train) [ 15650/160000] base_lr: 9.1073e-05 lr: 3.3672e-07 eta: 1 day, 15:58:57 time: 0.9956 data_time: 0.0040 memory: 8462 loss: 0.0234 decode.loss_ce: 0.0128 decode.acc_seg: 99.5907 aux.loss_ce: 0.0106 aux.acc_seg: 99.0166 +04/16 12:31:48 - mmengine - INFO - Iter(train) [ 15700/160000] base_lr: 9.1042e-05 lr: 3.3660e-07 eta: 1 day, 15:58:06 time: 0.9950 data_time: 0.0043 memory: 8462 loss: 0.0314 decode.loss_ce: 0.0188 decode.acc_seg: 99.2195 aux.loss_ce: 0.0126 aux.acc_seg: 98.7101 +04/16 12:32:38 - mmengine - INFO - Iter(train) [ 15750/160000] base_lr: 9.1010e-05 lr: 3.3648e-07 eta: 1 day, 15:57:15 time: 0.9938 data_time: 0.0042 memory: 8462 loss: 0.0252 decode.loss_ce: 0.0149 decode.acc_seg: 99.3534 aux.loss_ce: 0.0102 aux.acc_seg: 98.4583 +04/16 12:33:27 - mmengine - INFO - Iter(train) [ 15800/160000] base_lr: 9.0978e-05 lr: 3.3637e-07 eta: 1 day, 15:56:24 time: 0.9938 data_time: 0.0044 memory: 8462 loss: 0.0218 decode.loss_ce: 0.0121 decode.acc_seg: 99.6956 aux.loss_ce: 0.0097 aux.acc_seg: 99.3204 +04/16 12:34:17 - mmengine - INFO - Iter(train) [ 15850/160000] base_lr: 9.0947e-05 lr: 3.3625e-07 eta: 1 day, 15:55:33 time: 0.9939 data_time: 0.0044 memory: 8462 loss: 0.0235 decode.loss_ce: 0.0128 decode.acc_seg: 99.3752 aux.loss_ce: 0.0107 aux.acc_seg: 98.9956 +04/16 12:35:07 - mmengine - INFO - Iter(train) [ 15900/160000] base_lr: 9.0915e-05 lr: 3.3613e-07 eta: 1 day, 15:54:41 time: 0.9955 data_time: 0.0047 memory: 8462 loss: 0.0247 decode.loss_ce: 0.0144 decode.acc_seg: 99.2025 aux.loss_ce: 0.0103 aux.acc_seg: 98.9445 +04/16 12:35:57 - mmengine - INFO - Iter(train) [ 15950/160000] base_lr: 9.0884e-05 lr: 3.3602e-07 eta: 1 day, 15:53:51 time: 0.9944 data_time: 0.0045 memory: 8462 loss: 0.0289 decode.loss_ce: 0.0168 decode.acc_seg: 99.4741 aux.loss_ce: 0.0121 aux.acc_seg: 99.0055 +04/16 12:36:46 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 12:36:46 - mmengine - INFO - Iter(train) [ 16000/160000] base_lr: 9.0852e-05 lr: 3.3590e-07 eta: 1 day, 15:52:59 time: 0.9950 data_time: 0.0045 memory: 8462 loss: 0.0251 decode.loss_ce: 0.0143 decode.acc_seg: 99.3509 aux.loss_ce: 0.0108 aux.acc_seg: 98.8085 +04/16 12:37:36 - mmengine - INFO - Iter(train) [ 16050/160000] base_lr: 9.0821e-05 lr: 3.3578e-07 eta: 1 day, 15:52:09 time: 0.9947 data_time: 0.0047 memory: 8462 loss: 0.0339 decode.loss_ce: 0.0205 decode.acc_seg: 98.8548 aux.loss_ce: 0.0134 aux.acc_seg: 98.2023 +04/16 12:38:26 - mmengine - INFO - Iter(train) [ 16100/160000] base_lr: 9.0789e-05 lr: 3.3567e-07 eta: 1 day, 15:51:18 time: 0.9947 data_time: 0.0044 memory: 8462 loss: 0.0333 decode.loss_ce: 0.0198 decode.acc_seg: 98.9243 aux.loss_ce: 0.0135 aux.acc_seg: 98.3068 +04/16 12:39:16 - mmengine - INFO - Iter(train) [ 16150/160000] base_lr: 9.0758e-05 lr: 3.3555e-07 eta: 1 day, 15:50:27 time: 0.9947 data_time: 0.0045 memory: 8462 loss: 0.0289 decode.loss_ce: 0.0170 decode.acc_seg: 99.2973 aux.loss_ce: 0.0119 aux.acc_seg: 98.9229 +04/16 12:40:05 - mmengine - INFO - Iter(train) [ 16200/160000] base_lr: 9.0726e-05 lr: 3.3543e-07 eta: 1 day, 15:49:35 time: 0.9933 data_time: 0.0043 memory: 8462 loss: 0.0289 decode.loss_ce: 0.0165 decode.acc_seg: 99.2540 aux.loss_ce: 0.0124 aux.acc_seg: 98.7711 +04/16 12:40:55 - mmengine - INFO - Iter(train) [ 16250/160000] base_lr: 9.0695e-05 lr: 3.3532e-07 eta: 1 day, 15:48:44 time: 0.9938 data_time: 0.0045 memory: 8462 loss: 0.0229 decode.loss_ce: 0.0135 decode.acc_seg: 99.6256 aux.loss_ce: 0.0093 aux.acc_seg: 99.3422 +04/16 12:41:45 - mmengine - INFO - Iter(train) [ 16300/160000] base_lr: 9.0663e-05 lr: 3.3520e-07 eta: 1 day, 15:47:53 time: 0.9942 data_time: 0.0042 memory: 8462 loss: 0.0241 decode.loss_ce: 0.0139 decode.acc_seg: 99.5008 aux.loss_ce: 0.0102 aux.acc_seg: 99.0074 +04/16 12:42:34 - mmengine - INFO - Iter(train) [ 16350/160000] base_lr: 9.0631e-05 lr: 3.3508e-07 eta: 1 day, 15:47:02 time: 0.9941 data_time: 0.0045 memory: 8462 loss: 0.0244 decode.loss_ce: 0.0141 decode.acc_seg: 99.3250 aux.loss_ce: 0.0102 aux.acc_seg: 98.9412 +04/16 12:43:24 - mmengine - INFO - Iter(train) [ 16400/160000] base_lr: 9.0600e-05 lr: 3.3497e-07 eta: 1 day, 15:46:11 time: 0.9948 data_time: 0.0041 memory: 8462 loss: 0.0240 decode.loss_ce: 0.0136 decode.acc_seg: 98.9716 aux.loss_ce: 0.0104 aux.acc_seg: 98.4613 +04/16 12:44:14 - mmengine - INFO - Iter(train) [ 16450/160000] base_lr: 9.0568e-05 lr: 3.3485e-07 eta: 1 day, 15:45:20 time: 0.9952 data_time: 0.0043 memory: 8462 loss: 0.0307 decode.loss_ce: 0.0182 decode.acc_seg: 99.5695 aux.loss_ce: 0.0125 aux.acc_seg: 98.6841 +04/16 12:45:04 - mmengine - INFO - Iter(train) [ 16500/160000] base_lr: 9.0537e-05 lr: 3.3473e-07 eta: 1 day, 15:44:30 time: 0.9952 data_time: 0.0043 memory: 8462 loss: 0.0269 decode.loss_ce: 0.0154 decode.acc_seg: 99.6611 aux.loss_ce: 0.0115 aux.acc_seg: 99.2939 +04/16 12:45:53 - mmengine - INFO - Iter(train) [ 16550/160000] base_lr: 9.0505e-05 lr: 3.3462e-07 eta: 1 day, 15:43:39 time: 0.9945 data_time: 0.0047 memory: 8462 loss: 0.0230 decode.loss_ce: 0.0127 decode.acc_seg: 99.2729 aux.loss_ce: 0.0103 aux.acc_seg: 98.8014 +04/16 12:46:43 - mmengine - INFO - Iter(train) [ 16600/160000] base_lr: 9.0474e-05 lr: 3.3450e-07 eta: 1 day, 15:42:48 time: 0.9943 data_time: 0.0049 memory: 8462 loss: 0.0271 decode.loss_ce: 0.0158 decode.acc_seg: 99.4303 aux.loss_ce: 0.0113 aux.acc_seg: 98.9616 +04/16 12:47:33 - mmengine - INFO - Iter(train) [ 16650/160000] base_lr: 9.0442e-05 lr: 3.3438e-07 eta: 1 day, 15:41:57 time: 0.9940 data_time: 0.0044 memory: 8462 loss: 0.0253 decode.loss_ce: 0.0146 decode.acc_seg: 99.6136 aux.loss_ce: 0.0107 aux.acc_seg: 99.3130 +04/16 12:48:23 - mmengine - INFO - Iter(train) [ 16700/160000] base_lr: 9.0411e-05 lr: 3.3427e-07 eta: 1 day, 15:41:06 time: 0.9938 data_time: 0.0043 memory: 8462 loss: 0.0289 decode.loss_ce: 0.0167 decode.acc_seg: 99.1507 aux.loss_ce: 0.0121 aux.acc_seg: 98.4793 +04/16 12:49:12 - mmengine - INFO - Iter(train) [ 16750/160000] base_lr: 9.0379e-05 lr: 3.3415e-07 eta: 1 day, 15:40:15 time: 0.9950 data_time: 0.0047 memory: 8462 loss: 0.0268 decode.loss_ce: 0.0155 decode.acc_seg: 99.3820 aux.loss_ce: 0.0113 aux.acc_seg: 98.8203 +04/16 12:50:02 - mmengine - INFO - Iter(train) [ 16800/160000] base_lr: 9.0348e-05 lr: 3.3403e-07 eta: 1 day, 15:39:25 time: 0.9943 data_time: 0.0041 memory: 8462 loss: 0.0265 decode.loss_ce: 0.0158 decode.acc_seg: 99.6620 aux.loss_ce: 0.0107 aux.acc_seg: 99.1005 +04/16 12:50:52 - mmengine - INFO - Iter(train) [ 16850/160000] base_lr: 9.0316e-05 lr: 3.3392e-07 eta: 1 day, 15:38:34 time: 0.9947 data_time: 0.0042 memory: 8462 loss: 0.0254 decode.loss_ce: 0.0147 decode.acc_seg: 99.4753 aux.loss_ce: 0.0107 aux.acc_seg: 98.9782 +04/16 12:51:41 - mmengine - INFO - Iter(train) [ 16900/160000] base_lr: 9.0284e-05 lr: 3.3380e-07 eta: 1 day, 15:37:43 time: 0.9941 data_time: 0.0042 memory: 8462 loss: 0.0260 decode.loss_ce: 0.0151 decode.acc_seg: 99.6439 aux.loss_ce: 0.0109 aux.acc_seg: 99.1451 +04/16 12:52:31 - mmengine - INFO - Iter(train) [ 16950/160000] base_lr: 9.0253e-05 lr: 3.3368e-07 eta: 1 day, 15:36:52 time: 0.9949 data_time: 0.0046 memory: 8462 loss: 0.0249 decode.loss_ce: 0.0146 decode.acc_seg: 99.0032 aux.loss_ce: 0.0103 aux.acc_seg: 98.3549 +04/16 12:53:21 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 12:53:21 - mmengine - INFO - Iter(train) [ 17000/160000] base_lr: 9.0221e-05 lr: 3.3357e-07 eta: 1 day, 15:36:01 time: 0.9946 data_time: 0.0045 memory: 8462 loss: 0.0250 decode.loss_ce: 0.0144 decode.acc_seg: 99.5878 aux.loss_ce: 0.0106 aux.acc_seg: 99.1272 +04/16 12:54:11 - mmengine - INFO - Iter(train) [ 17050/160000] base_lr: 9.0190e-05 lr: 3.3345e-07 eta: 1 day, 15:35:10 time: 0.9936 data_time: 0.0044 memory: 8462 loss: 0.0237 decode.loss_ce: 0.0135 decode.acc_seg: 99.3713 aux.loss_ce: 0.0102 aux.acc_seg: 98.7221 +04/16 12:55:00 - mmengine - INFO - Iter(train) [ 17100/160000] base_lr: 9.0158e-05 lr: 3.3333e-07 eta: 1 day, 15:34:19 time: 0.9948 data_time: 0.0046 memory: 8462 loss: 0.0292 decode.loss_ce: 0.0171 decode.acc_seg: 99.2281 aux.loss_ce: 0.0122 aux.acc_seg: 99.0976 +04/16 12:55:50 - mmengine - INFO - Iter(train) [ 17150/160000] base_lr: 9.0127e-05 lr: 3.3322e-07 eta: 1 day, 15:33:28 time: 0.9938 data_time: 0.0045 memory: 8462 loss: 0.0280 decode.loss_ce: 0.0165 decode.acc_seg: 99.3313 aux.loss_ce: 0.0115 aux.acc_seg: 98.8781 +04/16 12:56:40 - mmengine - INFO - Iter(train) [ 17200/160000] base_lr: 9.0095e-05 lr: 3.3310e-07 eta: 1 day, 15:32:37 time: 0.9943 data_time: 0.0043 memory: 8462 loss: 0.0246 decode.loss_ce: 0.0146 decode.acc_seg: 99.5783 aux.loss_ce: 0.0100 aux.acc_seg: 99.2655 +04/16 12:57:30 - mmengine - INFO - Iter(train) [ 17250/160000] base_lr: 9.0064e-05 lr: 3.3298e-07 eta: 1 day, 15:31:47 time: 0.9956 data_time: 0.0044 memory: 8462 loss: 0.0232 decode.loss_ce: 0.0131 decode.acc_seg: 99.7149 aux.loss_ce: 0.0101 aux.acc_seg: 99.2603 +04/16 12:58:19 - mmengine - INFO - Iter(train) [ 17300/160000] base_lr: 9.0032e-05 lr: 3.3287e-07 eta: 1 day, 15:30:56 time: 0.9933 data_time: 0.0043 memory: 8462 loss: 0.0240 decode.loss_ce: 0.0141 decode.acc_seg: 99.4154 aux.loss_ce: 0.0098 aux.acc_seg: 98.8899 +04/16 12:59:09 - mmengine - INFO - Iter(train) [ 17350/160000] base_lr: 9.0001e-05 lr: 3.3275e-07 eta: 1 day, 15:30:05 time: 0.9952 data_time: 0.0042 memory: 8462 loss: 0.0260 decode.loss_ce: 0.0150 decode.acc_seg: 99.5564 aux.loss_ce: 0.0110 aux.acc_seg: 99.2086 +04/16 12:59:59 - mmengine - INFO - Iter(train) [ 17400/160000] base_lr: 8.9969e-05 lr: 3.3263e-07 eta: 1 day, 15:29:14 time: 0.9945 data_time: 0.0044 memory: 8462 loss: 0.0249 decode.loss_ce: 0.0138 decode.acc_seg: 99.5892 aux.loss_ce: 0.0111 aux.acc_seg: 98.9199 +04/16 13:00:48 - mmengine - INFO - Iter(train) [ 17450/160000] base_lr: 8.9937e-05 lr: 3.3252e-07 eta: 1 day, 15:28:23 time: 0.9939 data_time: 0.0042 memory: 8462 loss: 0.0223 decode.loss_ce: 0.0130 decode.acc_seg: 99.6269 aux.loss_ce: 0.0093 aux.acc_seg: 99.3746 +04/16 13:01:38 - mmengine - INFO - Iter(train) [ 17500/160000] base_lr: 8.9906e-05 lr: 3.3240e-07 eta: 1 day, 15:27:33 time: 0.9939 data_time: 0.0042 memory: 8462 loss: 0.0239 decode.loss_ce: 0.0136 decode.acc_seg: 99.3824 aux.loss_ce: 0.0103 aux.acc_seg: 98.6378 +04/16 13:02:28 - mmengine - INFO - Iter(train) [ 17550/160000] base_lr: 8.9874e-05 lr: 3.3228e-07 eta: 1 day, 15:26:42 time: 0.9943 data_time: 0.0044 memory: 8462 loss: 0.0239 decode.loss_ce: 0.0136 decode.acc_seg: 99.5794 aux.loss_ce: 0.0103 aux.acc_seg: 99.0973 +04/16 13:03:18 - mmengine - INFO - Iter(train) [ 17600/160000] base_lr: 8.9843e-05 lr: 3.3217e-07 eta: 1 day, 15:25:51 time: 0.9961 data_time: 0.0044 memory: 8462 loss: 0.0247 decode.loss_ce: 0.0140 decode.acc_seg: 99.5371 aux.loss_ce: 0.0107 aux.acc_seg: 99.2817 +04/16 13:04:07 - mmengine - INFO - Iter(train) [ 17650/160000] base_lr: 8.9811e-05 lr: 3.3205e-07 eta: 1 day, 15:25:00 time: 0.9952 data_time: 0.0046 memory: 8462 loss: 0.0285 decode.loss_ce: 0.0169 decode.acc_seg: 99.0692 aux.loss_ce: 0.0116 aux.acc_seg: 98.6195 +04/16 13:04:57 - mmengine - INFO - Iter(train) [ 17700/160000] base_lr: 8.9780e-05 lr: 3.3193e-07 eta: 1 day, 15:24:10 time: 0.9953 data_time: 0.0046 memory: 8462 loss: 0.0239 decode.loss_ce: 0.0142 decode.acc_seg: 99.5008 aux.loss_ce: 0.0097 aux.acc_seg: 98.8279 +04/16 13:05:47 - mmengine - INFO - Iter(train) [ 17750/160000] base_lr: 8.9748e-05 lr: 3.3182e-07 eta: 1 day, 15:23:19 time: 0.9949 data_time: 0.0046 memory: 8462 loss: 0.0279 decode.loss_ce: 0.0164 decode.acc_seg: 99.3298 aux.loss_ce: 0.0115 aux.acc_seg: 98.9090 +04/16 13:06:37 - mmengine - INFO - Iter(train) [ 17800/160000] base_lr: 8.9717e-05 lr: 3.3170e-07 eta: 1 day, 15:22:28 time: 0.9940 data_time: 0.0041 memory: 8462 loss: 0.0266 decode.loss_ce: 0.0154 decode.acc_seg: 99.4667 aux.loss_ce: 0.0112 aux.acc_seg: 98.8150 +04/16 13:07:26 - mmengine - INFO - Iter(train) [ 17850/160000] base_lr: 8.9685e-05 lr: 3.3158e-07 eta: 1 day, 15:21:37 time: 0.9951 data_time: 0.0044 memory: 8462 loss: 0.0263 decode.loss_ce: 0.0162 decode.acc_seg: 98.9483 aux.loss_ce: 0.0100 aux.acc_seg: 98.5121 +04/16 13:08:16 - mmengine - INFO - Iter(train) [ 17900/160000] base_lr: 8.9654e-05 lr: 3.3147e-07 eta: 1 day, 15:20:47 time: 0.9957 data_time: 0.0045 memory: 8462 loss: 0.0261 decode.loss_ce: 0.0151 decode.acc_seg: 99.7086 aux.loss_ce: 0.0110 aux.acc_seg: 99.5508 +04/16 13:09:06 - mmengine - INFO - Iter(train) [ 17950/160000] base_lr: 8.9622e-05 lr: 3.3135e-07 eta: 1 day, 15:19:56 time: 0.9954 data_time: 0.0044 memory: 8462 loss: 0.0242 decode.loss_ce: 0.0139 decode.acc_seg: 99.3830 aux.loss_ce: 0.0103 aux.acc_seg: 98.6149 +04/16 13:09:56 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 13:09:56 - mmengine - INFO - Iter(train) [ 18000/160000] base_lr: 8.9590e-05 lr: 3.3123e-07 eta: 1 day, 15:19:06 time: 0.9942 data_time: 0.0047 memory: 8462 loss: 0.0242 decode.loss_ce: 0.0132 decode.acc_seg: 99.1476 aux.loss_ce: 0.0109 aux.acc_seg: 98.5649 +04/16 13:10:45 - mmengine - INFO - Iter(train) [ 18050/160000] base_lr: 8.9559e-05 lr: 3.3112e-07 eta: 1 day, 15:18:15 time: 0.9946 data_time: 0.0043 memory: 8462 loss: 0.0243 decode.loss_ce: 0.0138 decode.acc_seg: 99.5646 aux.loss_ce: 0.0105 aux.acc_seg: 98.5634 +04/16 13:11:35 - mmengine - INFO - Iter(train) [ 18100/160000] base_lr: 8.9527e-05 lr: 3.3100e-07 eta: 1 day, 15:17:25 time: 0.9949 data_time: 0.0046 memory: 8462 loss: 0.0230 decode.loss_ce: 0.0125 decode.acc_seg: 99.3540 aux.loss_ce: 0.0105 aux.acc_seg: 98.6666 +04/16 13:12:25 - mmengine - INFO - Iter(train) [ 18150/160000] base_lr: 8.9496e-05 lr: 3.3088e-07 eta: 1 day, 15:16:34 time: 0.9943 data_time: 0.0045 memory: 8462 loss: 0.0263 decode.loss_ce: 0.0153 decode.acc_seg: 99.7435 aux.loss_ce: 0.0110 aux.acc_seg: 99.3992 +04/16 13:13:15 - mmengine - INFO - Iter(train) [ 18200/160000] base_lr: 8.9464e-05 lr: 3.3077e-07 eta: 1 day, 15:15:43 time: 0.9945 data_time: 0.0043 memory: 8462 loss: 0.0252 decode.loss_ce: 0.0144 decode.acc_seg: 99.2609 aux.loss_ce: 0.0108 aux.acc_seg: 98.9428 +04/16 13:14:04 - mmengine - INFO - Iter(train) [ 18250/160000] base_lr: 8.9433e-05 lr: 3.3065e-07 eta: 1 day, 15:14:52 time: 0.9944 data_time: 0.0042 memory: 8462 loss: 0.0253 decode.loss_ce: 0.0142 decode.acc_seg: 99.3624 aux.loss_ce: 0.0111 aux.acc_seg: 99.0913 +04/16 13:14:54 - mmengine - INFO - Iter(train) [ 18300/160000] base_lr: 8.9401e-05 lr: 3.3053e-07 eta: 1 day, 15:14:02 time: 0.9963 data_time: 0.0044 memory: 8462 loss: 0.0258 decode.loss_ce: 0.0145 decode.acc_seg: 99.6546 aux.loss_ce: 0.0113 aux.acc_seg: 99.3649 +04/16 13:15:44 - mmengine - INFO - Iter(train) [ 18350/160000] base_lr: 8.9370e-05 lr: 3.3042e-07 eta: 1 day, 15:13:11 time: 0.9934 data_time: 0.0042 memory: 8462 loss: 0.0205 decode.loss_ce: 0.0115 decode.acc_seg: 99.4238 aux.loss_ce: 0.0090 aux.acc_seg: 98.7648 +04/16 13:16:34 - mmengine - INFO - Iter(train) [ 18400/160000] base_lr: 8.9338e-05 lr: 3.3030e-07 eta: 1 day, 15:12:21 time: 0.9954 data_time: 0.0042 memory: 8462 loss: 0.0319 decode.loss_ce: 0.0187 decode.acc_seg: 99.6195 aux.loss_ce: 0.0133 aux.acc_seg: 99.2496 +04/16 13:17:23 - mmengine - INFO - Iter(train) [ 18450/160000] base_lr: 8.9307e-05 lr: 3.3018e-07 eta: 1 day, 15:11:30 time: 0.9944 data_time: 0.0042 memory: 8462 loss: 0.0211 decode.loss_ce: 0.0118 decode.acc_seg: 99.5138 aux.loss_ce: 0.0094 aux.acc_seg: 99.0393 +04/16 13:18:13 - mmengine - INFO - Iter(train) [ 18500/160000] base_lr: 8.9275e-05 lr: 3.3007e-07 eta: 1 day, 15:10:39 time: 0.9942 data_time: 0.0043 memory: 8462 loss: 0.0227 decode.loss_ce: 0.0125 decode.acc_seg: 99.7066 aux.loss_ce: 0.0101 aux.acc_seg: 99.0858 +04/16 13:19:03 - mmengine - INFO - Iter(train) [ 18550/160000] base_lr: 8.9243e-05 lr: 3.2995e-07 eta: 1 day, 15:09:48 time: 0.9949 data_time: 0.0045 memory: 8462 loss: 0.0200 decode.loss_ce: 0.0113 decode.acc_seg: 99.6571 aux.loss_ce: 0.0087 aux.acc_seg: 99.1039 +04/16 13:19:52 - mmengine - INFO - Iter(train) [ 18600/160000] base_lr: 8.9212e-05 lr: 3.2983e-07 eta: 1 day, 15:08:58 time: 0.9944 data_time: 0.0043 memory: 8462 loss: 0.0249 decode.loss_ce: 0.0140 decode.acc_seg: 99.3109 aux.loss_ce: 0.0109 aux.acc_seg: 98.7213 +04/16 13:20:42 - mmengine - INFO - Iter(train) [ 18650/160000] base_lr: 8.9180e-05 lr: 3.2972e-07 eta: 1 day, 15:08:07 time: 0.9938 data_time: 0.0042 memory: 8462 loss: 0.0236 decode.loss_ce: 0.0132 decode.acc_seg: 99.6332 aux.loss_ce: 0.0104 aux.acc_seg: 98.9868 +04/16 13:21:32 - mmengine - INFO - Iter(train) [ 18700/160000] base_lr: 8.9149e-05 lr: 3.2960e-07 eta: 1 day, 15:07:16 time: 0.9950 data_time: 0.0046 memory: 8462 loss: 0.0223 decode.loss_ce: 0.0125 decode.acc_seg: 99.3567 aux.loss_ce: 0.0098 aux.acc_seg: 98.7898 +04/16 13:22:22 - mmengine - INFO - Iter(train) [ 18750/160000] base_lr: 8.9117e-05 lr: 3.2948e-07 eta: 1 day, 15:06:25 time: 0.9939 data_time: 0.0043 memory: 8462 loss: 0.0244 decode.loss_ce: 0.0139 decode.acc_seg: 99.5955 aux.loss_ce: 0.0105 aux.acc_seg: 99.0759 +04/16 13:23:11 - mmengine - INFO - Iter(train) [ 18800/160000] base_lr: 8.9086e-05 lr: 3.2937e-07 eta: 1 day, 15:05:35 time: 0.9947 data_time: 0.0044 memory: 8462 loss: 0.0246 decode.loss_ce: 0.0140 decode.acc_seg: 99.1087 aux.loss_ce: 0.0106 aux.acc_seg: 98.5701 +04/16 13:24:01 - mmengine - INFO - Iter(train) [ 18850/160000] base_lr: 8.9054e-05 lr: 3.2925e-07 eta: 1 day, 15:04:44 time: 0.9952 data_time: 0.0048 memory: 8462 loss: 0.0241 decode.loss_ce: 0.0138 decode.acc_seg: 99.3853 aux.loss_ce: 0.0104 aux.acc_seg: 98.9166 +04/16 13:24:51 - mmengine - INFO - Iter(train) [ 18900/160000] base_lr: 8.9023e-05 lr: 3.2914e-07 eta: 1 day, 15:03:54 time: 0.9946 data_time: 0.0043 memory: 8462 loss: 0.0270 decode.loss_ce: 0.0159 decode.acc_seg: 99.3099 aux.loss_ce: 0.0111 aux.acc_seg: 98.9237 +04/16 13:25:41 - mmengine - INFO - Iter(train) [ 18950/160000] base_lr: 8.8991e-05 lr: 3.2902e-07 eta: 1 day, 15:03:03 time: 0.9949 data_time: 0.0046 memory: 8462 loss: 0.0261 decode.loss_ce: 0.0152 decode.acc_seg: 99.2718 aux.loss_ce: 0.0109 aux.acc_seg: 98.5825 +04/16 13:26:30 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 13:26:30 - mmengine - INFO - Iter(train) [ 19000/160000] base_lr: 8.8960e-05 lr: 3.2890e-07 eta: 1 day, 15:02:13 time: 0.9959 data_time: 0.0047 memory: 8462 loss: 0.0269 decode.loss_ce: 0.0155 decode.acc_seg: 99.3855 aux.loss_ce: 0.0114 aux.acc_seg: 98.9611 +04/16 13:27:20 - mmengine - INFO - Iter(train) [ 19050/160000] base_lr: 8.8928e-05 lr: 3.2879e-07 eta: 1 day, 15:01:23 time: 0.9954 data_time: 0.0045 memory: 8462 loss: 0.0267 decode.loss_ce: 0.0158 decode.acc_seg: 99.3736 aux.loss_ce: 0.0108 aux.acc_seg: 98.9206 +04/16 13:28:10 - mmengine - INFO - Iter(train) [ 19100/160000] base_lr: 8.8896e-05 lr: 3.2867e-07 eta: 1 day, 15:00:32 time: 0.9939 data_time: 0.0042 memory: 8462 loss: 0.0258 decode.loss_ce: 0.0151 decode.acc_seg: 99.3345 aux.loss_ce: 0.0108 aux.acc_seg: 98.7389 +04/16 13:29:00 - mmengine - INFO - Iter(train) [ 19150/160000] base_lr: 8.8865e-05 lr: 3.2855e-07 eta: 1 day, 14:59:42 time: 0.9954 data_time: 0.0042 memory: 8462 loss: 0.0285 decode.loss_ce: 0.0166 decode.acc_seg: 99.3147 aux.loss_ce: 0.0119 aux.acc_seg: 98.7106 +04/16 13:29:49 - mmengine - INFO - Iter(train) [ 19200/160000] base_lr: 8.8833e-05 lr: 3.2844e-07 eta: 1 day, 14:58:51 time: 0.9954 data_time: 0.0046 memory: 8462 loss: 0.0236 decode.loss_ce: 0.0136 decode.acc_seg: 99.3536 aux.loss_ce: 0.0100 aux.acc_seg: 98.9351 +04/16 13:30:39 - mmengine - INFO - Iter(train) [ 19250/160000] base_lr: 8.8802e-05 lr: 3.2832e-07 eta: 1 day, 14:58:01 time: 0.9949 data_time: 0.0047 memory: 8462 loss: 0.0250 decode.loss_ce: 0.0145 decode.acc_seg: 99.7335 aux.loss_ce: 0.0105 aux.acc_seg: 99.4125 +04/16 13:31:29 - mmengine - INFO - Iter(train) [ 19300/160000] base_lr: 8.8770e-05 lr: 3.2820e-07 eta: 1 day, 14:57:11 time: 0.9966 data_time: 0.0045 memory: 8462 loss: 0.0275 decode.loss_ce: 0.0157 decode.acc_seg: 99.1705 aux.loss_ce: 0.0117 aux.acc_seg: 98.4364 +04/16 13:32:19 - mmengine - INFO - Iter(train) [ 19350/160000] base_lr: 8.8739e-05 lr: 3.2809e-07 eta: 1 day, 14:56:20 time: 0.9969 data_time: 0.0044 memory: 8462 loss: 0.0224 decode.loss_ce: 0.0130 decode.acc_seg: 99.3145 aux.loss_ce: 0.0094 aux.acc_seg: 98.8594 +04/16 13:33:08 - mmengine - INFO - Iter(train) [ 19400/160000] base_lr: 8.8707e-05 lr: 3.2797e-07 eta: 1 day, 14:55:30 time: 0.9945 data_time: 0.0047 memory: 8462 loss: 0.0228 decode.loss_ce: 0.0131 decode.acc_seg: 99.2586 aux.loss_ce: 0.0097 aux.acc_seg: 98.9151 +04/16 13:33:58 - mmengine - INFO - Iter(train) [ 19450/160000] base_lr: 8.8676e-05 lr: 3.2785e-07 eta: 1 day, 14:54:40 time: 0.9957 data_time: 0.0045 memory: 8462 loss: 0.0242 decode.loss_ce: 0.0132 decode.acc_seg: 99.3841 aux.loss_ce: 0.0109 aux.acc_seg: 98.5268 +04/16 13:34:48 - mmengine - INFO - Iter(train) [ 19500/160000] base_lr: 8.8644e-05 lr: 3.2774e-07 eta: 1 day, 14:53:50 time: 0.9953 data_time: 0.0045 memory: 8462 loss: 0.0228 decode.loss_ce: 0.0127 decode.acc_seg: 99.5329 aux.loss_ce: 0.0101 aux.acc_seg: 99.0875 +04/16 13:35:38 - mmengine - INFO - Iter(train) [ 19550/160000] base_lr: 8.8613e-05 lr: 3.2762e-07 eta: 1 day, 14:52:59 time: 0.9963 data_time: 0.0046 memory: 8462 loss: 0.0267 decode.loss_ce: 0.0148 decode.acc_seg: 99.3374 aux.loss_ce: 0.0119 aux.acc_seg: 98.7682 +04/16 13:36:28 - mmengine - INFO - Iter(train) [ 19600/160000] base_lr: 8.8581e-05 lr: 3.2750e-07 eta: 1 day, 14:52:09 time: 0.9957 data_time: 0.0046 memory: 8462 loss: 0.0239 decode.loss_ce: 0.0137 decode.acc_seg: 99.4041 aux.loss_ce: 0.0103 aux.acc_seg: 98.5903 +04/16 13:37:17 - mmengine - INFO - Iter(train) [ 19650/160000] base_lr: 8.8549e-05 lr: 3.2739e-07 eta: 1 day, 14:51:19 time: 0.9951 data_time: 0.0047 memory: 8462 loss: 0.0216 decode.loss_ce: 0.0122 decode.acc_seg: 99.5720 aux.loss_ce: 0.0094 aux.acc_seg: 99.1405 +04/16 13:38:07 - mmengine - INFO - Iter(train) [ 19700/160000] base_lr: 8.8518e-05 lr: 3.2727e-07 eta: 1 day, 14:50:28 time: 0.9953 data_time: 0.0051 memory: 8462 loss: 0.0281 decode.loss_ce: 0.0164 decode.acc_seg: 98.9395 aux.loss_ce: 0.0118 aux.acc_seg: 98.3028 +04/16 13:38:57 - mmengine - INFO - Iter(train) [ 19750/160000] base_lr: 8.8486e-05 lr: 3.2715e-07 eta: 1 day, 14:49:38 time: 0.9940 data_time: 0.0044 memory: 8462 loss: 0.0215 decode.loss_ce: 0.0119 decode.acc_seg: 99.4602 aux.loss_ce: 0.0096 aux.acc_seg: 99.1415 +04/16 13:39:47 - mmengine - INFO - Iter(train) [ 19800/160000] base_lr: 8.8455e-05 lr: 3.2704e-07 eta: 1 day, 14:48:47 time: 0.9945 data_time: 0.0045 memory: 8462 loss: 0.0227 decode.loss_ce: 0.0132 decode.acc_seg: 99.2502 aux.loss_ce: 0.0095 aux.acc_seg: 98.8523 +04/16 13:40:36 - mmengine - INFO - Iter(train) [ 19850/160000] base_lr: 8.8423e-05 lr: 3.2692e-07 eta: 1 day, 14:47:58 time: 0.9951 data_time: 0.0045 memory: 8462 loss: 0.0255 decode.loss_ce: 0.0144 decode.acc_seg: 99.4022 aux.loss_ce: 0.0112 aux.acc_seg: 98.5693 +04/16 13:41:26 - mmengine - INFO - Iter(train) [ 19900/160000] base_lr: 8.8392e-05 lr: 3.2680e-07 eta: 1 day, 14:47:07 time: 0.9962 data_time: 0.0044 memory: 8462 loss: 0.0227 decode.loss_ce: 0.0129 decode.acc_seg: 99.5338 aux.loss_ce: 0.0098 aux.acc_seg: 98.9962 +04/16 13:42:16 - mmengine - INFO - Iter(train) [ 19950/160000] base_lr: 8.8360e-05 lr: 3.2669e-07 eta: 1 day, 14:46:17 time: 0.9960 data_time: 0.0045 memory: 8462 loss: 0.0232 decode.loss_ce: 0.0131 decode.acc_seg: 99.4450 aux.loss_ce: 0.0101 aux.acc_seg: 99.0223 +04/16 13:43:06 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 13:43:06 - mmengine - INFO - Iter(train) [ 20000/160000] base_lr: 8.8329e-05 lr: 3.2657e-07 eta: 1 day, 14:45:27 time: 0.9964 data_time: 0.0046 memory: 8462 loss: 0.0244 decode.loss_ce: 0.0140 decode.acc_seg: 99.5642 aux.loss_ce: 0.0104 aux.acc_seg: 99.0322 +04/16 13:43:06 - mmengine - INFO - Saving checkpoint at 20000 iterations +04/16 13:43:16 - mmengine - INFO - Iter(val) [ 50/355] eta: 0:00:35 time: 0.1152 data_time: 0.0014 memory: 4004 +04/16 13:43:21 - mmengine - INFO - Iter(val) [100/355] eta: 0:00:29 time: 0.1157 data_time: 0.0017 memory: 4004 +04/16 13:43:27 - mmengine - INFO - Iter(val) [150/355] eta: 0:00:23 time: 0.1154 data_time: 0.0015 memory: 4004 +04/16 13:43:33 - mmengine - INFO - Iter(val) [200/355] eta: 0:00:17 time: 0.1154 data_time: 0.0014 memory: 4004 +04/16 13:43:39 - mmengine - INFO - Iter(val) [250/355] eta: 0:00:12 time: 0.1157 data_time: 0.0016 memory: 4004 +04/16 13:43:44 - mmengine - INFO - Iter(val) [300/355] eta: 0:00:06 time: 0.1155 data_time: 0.0015 memory: 4004 +04/16 13:43:50 - mmengine - INFO - Iter(val) [350/355] eta: 0:00:00 time: 0.1157 data_time: 0.0016 memory: 4004 +04/16 13:43:51 - mmengine - INFO - per class results: +04/16 13:43:51 - mmengine - INFO - ++------------+-------+-------+ +| Class | IoU | Acc | ++------------+-------+-------+ +| background | 99.18 | 99.59 | +| contrast | 82.13 | 90.09 | ++------------+-------+-------+ +04/16 13:43:51 - mmengine - INFO - Iter(val) [355/355] aAcc: 99.2100 mIoU: 90.6500 mAcc: 94.8400 data_time: 0.0017 time: 0.1157 +04/16 13:44:41 - mmengine - INFO - Iter(train) [ 20050/160000] base_lr: 8.8297e-05 lr: 3.2645e-07 eta: 1 day, 14:44:39 time: 0.9948 data_time: 0.0044 memory: 8462 loss: 0.0237 decode.loss_ce: 0.0136 decode.acc_seg: 99.4276 aux.loss_ce: 0.0102 aux.acc_seg: 99.0868 +04/16 13:45:31 - mmengine - INFO - Iter(train) [ 20100/160000] base_lr: 8.8266e-05 lr: 3.2634e-07 eta: 1 day, 14:43:48 time: 0.9965 data_time: 0.0043 memory: 8462 loss: 0.0308 decode.loss_ce: 0.0174 decode.acc_seg: 99.2613 aux.loss_ce: 0.0134 aux.acc_seg: 98.5760 +04/16 13:46:20 - mmengine - INFO - Iter(train) [ 20150/160000] base_lr: 8.8234e-05 lr: 3.2622e-07 eta: 1 day, 14:42:58 time: 0.9954 data_time: 0.0045 memory: 8462 loss: 0.0247 decode.loss_ce: 0.0141 decode.acc_seg: 99.6801 aux.loss_ce: 0.0105 aux.acc_seg: 99.2981 +04/16 13:47:10 - mmengine - INFO - Iter(train) [ 20200/160000] base_lr: 8.8202e-05 lr: 3.2610e-07 eta: 1 day, 14:42:08 time: 0.9946 data_time: 0.0043 memory: 8462 loss: 0.0231 decode.loss_ce: 0.0127 decode.acc_seg: 99.7377 aux.loss_ce: 0.0105 aux.acc_seg: 99.4703 +04/16 13:48:00 - mmengine - INFO - Iter(train) [ 20250/160000] base_lr: 8.8171e-05 lr: 3.2599e-07 eta: 1 day, 14:41:18 time: 0.9958 data_time: 0.0044 memory: 8462 loss: 0.0211 decode.loss_ce: 0.0116 decode.acc_seg: 99.5728 aux.loss_ce: 0.0095 aux.acc_seg: 99.2064 +04/16 13:48:50 - mmengine - INFO - Iter(train) [ 20300/160000] base_lr: 8.8139e-05 lr: 3.2587e-07 eta: 1 day, 14:40:28 time: 0.9955 data_time: 0.0047 memory: 8462 loss: 0.0222 decode.loss_ce: 0.0122 decode.acc_seg: 99.5043 aux.loss_ce: 0.0100 aux.acc_seg: 98.6240 +04/16 13:49:40 - mmengine - INFO - Iter(train) [ 20350/160000] base_lr: 8.8108e-05 lr: 3.2575e-07 eta: 1 day, 14:39:38 time: 0.9963 data_time: 0.0045 memory: 8462 loss: 0.0226 decode.loss_ce: 0.0126 decode.acc_seg: 99.4381 aux.loss_ce: 0.0100 aux.acc_seg: 98.7467 +04/16 13:50:29 - mmengine - INFO - Iter(train) [ 20400/160000] base_lr: 8.8076e-05 lr: 3.2564e-07 eta: 1 day, 14:38:48 time: 0.9958 data_time: 0.0042 memory: 8462 loss: 0.0234 decode.loss_ce: 0.0130 decode.acc_seg: 99.5968 aux.loss_ce: 0.0104 aux.acc_seg: 99.2714 +04/16 13:51:19 - mmengine - INFO - Iter(train) [ 20450/160000] base_lr: 8.8045e-05 lr: 3.2552e-07 eta: 1 day, 14:37:58 time: 0.9963 data_time: 0.0043 memory: 8462 loss: 0.0220 decode.loss_ce: 0.0124 decode.acc_seg: 99.3410 aux.loss_ce: 0.0096 aux.acc_seg: 98.7488 +04/16 13:52:09 - mmengine - INFO - Iter(train) [ 20500/160000] base_lr: 8.8013e-05 lr: 3.2540e-07 eta: 1 day, 14:37:08 time: 0.9954 data_time: 0.0043 memory: 8462 loss: 0.0272 decode.loss_ce: 0.0159 decode.acc_seg: 98.9864 aux.loss_ce: 0.0114 aux.acc_seg: 98.5086 +04/16 13:52:59 - mmengine - INFO - Iter(train) [ 20550/160000] base_lr: 8.7982e-05 lr: 3.2529e-07 eta: 1 day, 14:36:18 time: 0.9956 data_time: 0.0042 memory: 8462 loss: 0.0261 decode.loss_ce: 0.0151 decode.acc_seg: 99.5337 aux.loss_ce: 0.0110 aux.acc_seg: 98.8176 +04/16 13:53:49 - mmengine - INFO - Iter(train) [ 20600/160000] base_lr: 8.7950e-05 lr: 3.2517e-07 eta: 1 day, 14:35:28 time: 0.9963 data_time: 0.0046 memory: 8462 loss: 0.0235 decode.loss_ce: 0.0128 decode.acc_seg: 99.7038 aux.loss_ce: 0.0108 aux.acc_seg: 99.2350 +04/16 13:54:38 - mmengine - INFO - Iter(train) [ 20650/160000] base_lr: 8.7919e-05 lr: 3.2505e-07 eta: 1 day, 14:34:38 time: 0.9973 data_time: 0.0042 memory: 8462 loss: 0.0199 decode.loss_ce: 0.0114 decode.acc_seg: 99.1310 aux.loss_ce: 0.0084 aux.acc_seg: 98.4112 +04/16 13:55:28 - mmengine - INFO - Iter(train) [ 20700/160000] base_lr: 8.7887e-05 lr: 3.2494e-07 eta: 1 day, 14:33:48 time: 0.9956 data_time: 0.0044 memory: 8462 loss: 0.0261 decode.loss_ce: 0.0154 decode.acc_seg: 99.6151 aux.loss_ce: 0.0108 aux.acc_seg: 99.3122 +04/16 13:56:18 - mmengine - INFO - Iter(train) [ 20750/160000] base_lr: 8.7855e-05 lr: 3.2482e-07 eta: 1 day, 14:32:58 time: 0.9951 data_time: 0.0046 memory: 8462 loss: 0.0284 decode.loss_ce: 0.0161 decode.acc_seg: 99.7086 aux.loss_ce: 0.0122 aux.acc_seg: 99.2863 +04/16 13:57:08 - mmengine - INFO - Iter(train) [ 20800/160000] base_lr: 8.7824e-05 lr: 3.2470e-07 eta: 1 day, 14:32:08 time: 0.9958 data_time: 0.0041 memory: 8462 loss: 0.0230 decode.loss_ce: 0.0133 decode.acc_seg: 99.6466 aux.loss_ce: 0.0097 aux.acc_seg: 98.8922 +04/16 13:57:58 - mmengine - INFO - Iter(train) [ 20850/160000] base_lr: 8.7792e-05 lr: 3.2459e-07 eta: 1 day, 14:31:17 time: 0.9961 data_time: 0.0045 memory: 8462 loss: 0.0219 decode.loss_ce: 0.0125 decode.acc_seg: 99.4942 aux.loss_ce: 0.0094 aux.acc_seg: 98.9948 +04/16 13:58:47 - mmengine - INFO - Iter(train) [ 20900/160000] base_lr: 8.7761e-05 lr: 3.2447e-07 eta: 1 day, 14:30:28 time: 0.9965 data_time: 0.0044 memory: 8462 loss: 0.0258 decode.loss_ce: 0.0144 decode.acc_seg: 99.0498 aux.loss_ce: 0.0115 aux.acc_seg: 98.2670 +04/16 13:59:37 - mmengine - INFO - Iter(train) [ 20950/160000] base_lr: 8.7729e-05 lr: 3.2435e-07 eta: 1 day, 14:29:38 time: 0.9974 data_time: 0.0047 memory: 8462 loss: 0.0242 decode.loss_ce: 0.0140 decode.acc_seg: 99.5800 aux.loss_ce: 0.0102 aux.acc_seg: 99.3896 +04/16 14:00:27 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 14:00:27 - mmengine - INFO - Iter(train) [ 21000/160000] base_lr: 8.7698e-05 lr: 3.2424e-07 eta: 1 day, 14:28:48 time: 0.9980 data_time: 0.0053 memory: 8462 loss: 0.0248 decode.loss_ce: 0.0142 decode.acc_seg: 98.9771 aux.loss_ce: 0.0105 aux.acc_seg: 98.5836 +04/16 14:01:17 - mmengine - INFO - Iter(train) [ 21050/160000] base_lr: 8.7666e-05 lr: 3.2412e-07 eta: 1 day, 14:27:58 time: 0.9961 data_time: 0.0044 memory: 8462 loss: 0.0297 decode.loss_ce: 0.0173 decode.acc_seg: 99.3839 aux.loss_ce: 0.0124 aux.acc_seg: 98.7478 +04/16 14:02:07 - mmengine - INFO - Iter(train) [ 21100/160000] base_lr: 8.7635e-05 lr: 3.2400e-07 eta: 1 day, 14:27:08 time: 0.9969 data_time: 0.0050 memory: 8462 loss: 0.0233 decode.loss_ce: 0.0132 decode.acc_seg: 99.4356 aux.loss_ce: 0.0101 aux.acc_seg: 99.1446 +04/16 14:02:57 - mmengine - INFO - Iter(train) [ 21150/160000] base_lr: 8.7603e-05 lr: 3.2389e-07 eta: 1 day, 14:26:19 time: 0.9967 data_time: 0.0043 memory: 8462 loss: 0.0227 decode.loss_ce: 0.0133 decode.acc_seg: 99.4202 aux.loss_ce: 0.0094 aux.acc_seg: 98.9567 +04/16 14:03:47 - mmengine - INFO - Iter(train) [ 21200/160000] base_lr: 8.7572e-05 lr: 3.2377e-07 eta: 1 day, 14:25:29 time: 0.9969 data_time: 0.0047 memory: 8462 loss: 0.0217 decode.loss_ce: 0.0123 decode.acc_seg: 99.7370 aux.loss_ce: 0.0095 aux.acc_seg: 99.4749 +04/16 14:04:36 - mmengine - INFO - Iter(train) [ 21250/160000] base_lr: 8.7540e-05 lr: 3.2365e-07 eta: 1 day, 14:24:39 time: 0.9969 data_time: 0.0046 memory: 8462 loss: 0.0236 decode.loss_ce: 0.0130 decode.acc_seg: 99.6424 aux.loss_ce: 0.0106 aux.acc_seg: 99.0780 +04/16 14:05:26 - mmengine - INFO - Iter(train) [ 21300/160000] base_lr: 8.7508e-05 lr: 3.2354e-07 eta: 1 day, 14:23:49 time: 0.9962 data_time: 0.0044 memory: 8462 loss: 0.0228 decode.loss_ce: 0.0129 decode.acc_seg: 99.6460 aux.loss_ce: 0.0099 aux.acc_seg: 99.1171 +04/16 14:06:16 - mmengine - INFO - Iter(train) [ 21350/160000] base_lr: 8.7477e-05 lr: 3.2342e-07 eta: 1 day, 14:22:59 time: 0.9978 data_time: 0.0043 memory: 8462 loss: 0.0257 decode.loss_ce: 0.0149 decode.acc_seg: 99.1144 aux.loss_ce: 0.0108 aux.acc_seg: 98.3063 +04/16 14:07:06 - mmengine - INFO - Iter(train) [ 21400/160000] base_lr: 8.7445e-05 lr: 3.2330e-07 eta: 1 day, 14:22:10 time: 0.9970 data_time: 0.0046 memory: 8462 loss: 0.0260 decode.loss_ce: 0.0144 decode.acc_seg: 99.5527 aux.loss_ce: 0.0116 aux.acc_seg: 98.5941 +04/16 14:07:56 - mmengine - INFO - Iter(train) [ 21450/160000] base_lr: 8.7414e-05 lr: 3.2319e-07 eta: 1 day, 14:21:20 time: 0.9966 data_time: 0.0044 memory: 8462 loss: 0.0214 decode.loss_ce: 0.0123 decode.acc_seg: 99.2031 aux.loss_ce: 0.0091 aux.acc_seg: 98.9466 +04/16 14:08:45 - mmengine - INFO - Iter(train) [ 21500/160000] base_lr: 8.7382e-05 lr: 3.2307e-07 eta: 1 day, 14:20:30 time: 0.9976 data_time: 0.0047 memory: 8462 loss: 0.0261 decode.loss_ce: 0.0150 decode.acc_seg: 99.6807 aux.loss_ce: 0.0111 aux.acc_seg: 99.3599 +04/16 14:09:35 - mmengine - INFO - Iter(train) [ 21550/160000] base_lr: 8.7351e-05 lr: 3.2295e-07 eta: 1 day, 14:19:40 time: 0.9962 data_time: 0.0050 memory: 8462 loss: 0.0239 decode.loss_ce: 0.0134 decode.acc_seg: 99.3223 aux.loss_ce: 0.0105 aux.acc_seg: 98.9008 +04/16 14:10:25 - mmengine - INFO - Iter(train) [ 21600/160000] base_lr: 8.7319e-05 lr: 3.2284e-07 eta: 1 day, 14:18:50 time: 0.9974 data_time: 0.0045 memory: 8462 loss: 0.0219 decode.loss_ce: 0.0122 decode.acc_seg: 99.5829 aux.loss_ce: 0.0098 aux.acc_seg: 99.2950 +04/16 14:11:15 - mmengine - INFO - Iter(train) [ 21650/160000] base_lr: 8.7288e-05 lr: 3.2272e-07 eta: 1 day, 14:18:00 time: 0.9974 data_time: 0.0046 memory: 8462 loss: 0.0248 decode.loss_ce: 0.0135 decode.acc_seg: 99.1840 aux.loss_ce: 0.0113 aux.acc_seg: 98.6668 +04/16 14:12:05 - mmengine - INFO - Iter(train) [ 21700/160000] base_lr: 8.7256e-05 lr: 3.2260e-07 eta: 1 day, 14:17:11 time: 0.9964 data_time: 0.0043 memory: 8462 loss: 0.0257 decode.loss_ce: 0.0143 decode.acc_seg: 99.3885 aux.loss_ce: 0.0114 aux.acc_seg: 98.5630 +04/16 14:12:55 - mmengine - INFO - Iter(train) [ 21750/160000] base_lr: 8.7225e-05 lr: 3.2249e-07 eta: 1 day, 14:16:21 time: 0.9972 data_time: 0.0046 memory: 8462 loss: 0.0261 decode.loss_ce: 0.0154 decode.acc_seg: 99.4652 aux.loss_ce: 0.0107 aux.acc_seg: 99.2119 +04/16 14:13:44 - mmengine - INFO - Iter(train) [ 21800/160000] base_lr: 8.7193e-05 lr: 3.2237e-07 eta: 1 day, 14:15:31 time: 0.9969 data_time: 0.0043 memory: 8462 loss: 0.0227 decode.loss_ce: 0.0128 decode.acc_seg: 99.5832 aux.loss_ce: 0.0098 aux.acc_seg: 98.6763 +04/16 14:14:34 - mmengine - INFO - Iter(train) [ 21850/160000] base_lr: 8.7161e-05 lr: 3.2225e-07 eta: 1 day, 14:14:41 time: 0.9957 data_time: 0.0047 memory: 8462 loss: 0.0224 decode.loss_ce: 0.0125 decode.acc_seg: 99.3341 aux.loss_ce: 0.0100 aux.acc_seg: 98.8565 +04/16 14:15:24 - mmengine - INFO - Iter(train) [ 21900/160000] base_lr: 8.7130e-05 lr: 3.2214e-07 eta: 1 day, 14:13:51 time: 0.9981 data_time: 0.0044 memory: 8462 loss: 0.0205 decode.loss_ce: 0.0117 decode.acc_seg: 99.6178 aux.loss_ce: 0.0089 aux.acc_seg: 99.0368 +04/16 14:16:14 - mmengine - INFO - Iter(train) [ 21950/160000] base_lr: 8.7098e-05 lr: 3.2202e-07 eta: 1 day, 14:13:02 time: 0.9969 data_time: 0.0043 memory: 8462 loss: 0.0278 decode.loss_ce: 0.0158 decode.acc_seg: 98.3555 aux.loss_ce: 0.0120 aux.acc_seg: 97.7139 +04/16 14:17:04 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 14:17:04 - mmengine - INFO - Iter(train) [ 22000/160000] base_lr: 8.7067e-05 lr: 3.2190e-07 eta: 1 day, 14:12:13 time: 0.9964 data_time: 0.0045 memory: 8462 loss: 0.0204 decode.loss_ce: 0.0116 decode.acc_seg: 99.6542 aux.loss_ce: 0.0088 aux.acc_seg: 99.0362 +04/16 14:17:54 - mmengine - INFO - Iter(train) [ 22050/160000] base_lr: 8.7035e-05 lr: 3.2179e-07 eta: 1 day, 14:11:23 time: 0.9976 data_time: 0.0048 memory: 8462 loss: 0.0223 decode.loss_ce: 0.0124 decode.acc_seg: 99.2950 aux.loss_ce: 0.0098 aux.acc_seg: 98.9662 +04/16 14:18:44 - mmengine - INFO - Iter(train) [ 22100/160000] base_lr: 8.7004e-05 lr: 3.2167e-07 eta: 1 day, 14:10:33 time: 0.9961 data_time: 0.0043 memory: 8462 loss: 0.0235 decode.loss_ce: 0.0133 decode.acc_seg: 99.2270 aux.loss_ce: 0.0103 aux.acc_seg: 98.7913 +04/16 14:19:34 - mmengine - INFO - Iter(train) [ 22150/160000] base_lr: 8.6972e-05 lr: 3.2155e-07 eta: 1 day, 14:09:43 time: 0.9974 data_time: 0.0046 memory: 8462 loss: 0.0232 decode.loss_ce: 0.0128 decode.acc_seg: 99.4585 aux.loss_ce: 0.0104 aux.acc_seg: 98.9992 +04/16 14:20:23 - mmengine - INFO - Iter(train) [ 22200/160000] base_lr: 8.6941e-05 lr: 3.2144e-07 eta: 1 day, 14:08:53 time: 0.9968 data_time: 0.0043 memory: 8462 loss: 0.0211 decode.loss_ce: 0.0113 decode.acc_seg: 99.6887 aux.loss_ce: 0.0098 aux.acc_seg: 99.2519 +04/16 14:21:13 - mmengine - INFO - Iter(train) [ 22250/160000] base_lr: 8.6909e-05 lr: 3.2132e-07 eta: 1 day, 14:08:04 time: 0.9971 data_time: 0.0043 memory: 8462 loss: 0.0220 decode.loss_ce: 0.0120 decode.acc_seg: 99.6883 aux.loss_ce: 0.0100 aux.acc_seg: 99.4377 +04/16 14:22:03 - mmengine - INFO - Iter(train) [ 22300/160000] base_lr: 8.6878e-05 lr: 3.2120e-07 eta: 1 day, 14:07:14 time: 0.9976 data_time: 0.0047 memory: 8462 loss: 0.0231 decode.loss_ce: 0.0137 decode.acc_seg: 99.3752 aux.loss_ce: 0.0094 aux.acc_seg: 98.8932 +04/16 14:22:53 - mmengine - INFO - Iter(train) [ 22350/160000] base_lr: 8.6846e-05 lr: 3.2109e-07 eta: 1 day, 14:06:24 time: 0.9970 data_time: 0.0046 memory: 8462 loss: 0.0255 decode.loss_ce: 0.0146 decode.acc_seg: 99.4810 aux.loss_ce: 0.0109 aux.acc_seg: 98.9985 +04/16 14:23:43 - mmengine - INFO - Iter(train) [ 22400/160000] base_lr: 8.6814e-05 lr: 3.2097e-07 eta: 1 day, 14:05:34 time: 0.9965 data_time: 0.0044 memory: 8462 loss: 0.0228 decode.loss_ce: 0.0125 decode.acc_seg: 99.4696 aux.loss_ce: 0.0103 aux.acc_seg: 99.0412 +04/16 14:24:33 - mmengine - INFO - Iter(train) [ 22450/160000] base_lr: 8.6783e-05 lr: 3.2085e-07 eta: 1 day, 14:04:44 time: 0.9965 data_time: 0.0042 memory: 8462 loss: 0.0241 decode.loss_ce: 0.0134 decode.acc_seg: 99.5745 aux.loss_ce: 0.0107 aux.acc_seg: 99.1385 +04/16 14:25:22 - mmengine - INFO - Iter(train) [ 22500/160000] base_lr: 8.6751e-05 lr: 3.2074e-07 eta: 1 day, 14:03:55 time: 0.9967 data_time: 0.0041 memory: 8462 loss: 0.0225 decode.loss_ce: 0.0126 decode.acc_seg: 99.4711 aux.loss_ce: 0.0099 aux.acc_seg: 99.0034 +04/16 14:26:12 - mmengine - INFO - Iter(train) [ 22550/160000] base_lr: 8.6720e-05 lr: 3.2062e-07 eta: 1 day, 14:03:05 time: 0.9976 data_time: 0.0051 memory: 8462 loss: 0.0253 decode.loss_ce: 0.0149 decode.acc_seg: 99.3826 aux.loss_ce: 0.0104 aux.acc_seg: 98.5819 +04/16 14:27:02 - mmengine - INFO - Iter(train) [ 22600/160000] base_lr: 8.6688e-05 lr: 3.2050e-07 eta: 1 day, 14:02:15 time: 0.9971 data_time: 0.0044 memory: 8462 loss: 0.0222 decode.loss_ce: 0.0123 decode.acc_seg: 99.5838 aux.loss_ce: 0.0100 aux.acc_seg: 99.1564 +04/16 14:27:52 - mmengine - INFO - Iter(train) [ 22650/160000] base_lr: 8.6657e-05 lr: 3.2039e-07 eta: 1 day, 14:01:25 time: 0.9963 data_time: 0.0046 memory: 8462 loss: 0.0223 decode.loss_ce: 0.0124 decode.acc_seg: 99.5415 aux.loss_ce: 0.0100 aux.acc_seg: 98.6750 +04/16 14:28:42 - mmengine - INFO - Iter(train) [ 22700/160000] base_lr: 8.6625e-05 lr: 3.2027e-07 eta: 1 day, 14:00:36 time: 0.9968 data_time: 0.0045 memory: 8462 loss: 0.0222 decode.loss_ce: 0.0123 decode.acc_seg: 99.6025 aux.loss_ce: 0.0098 aux.acc_seg: 99.0717 +04/16 14:29:32 - mmengine - INFO - Iter(train) [ 22750/160000] base_lr: 8.6594e-05 lr: 3.2015e-07 eta: 1 day, 13:59:46 time: 0.9963 data_time: 0.0043 memory: 8462 loss: 0.0257 decode.loss_ce: 0.0146 decode.acc_seg: 99.0210 aux.loss_ce: 0.0111 aux.acc_seg: 98.5020 +04/16 14:30:21 - mmengine - INFO - Iter(train) [ 22800/160000] base_lr: 8.6562e-05 lr: 3.2004e-07 eta: 1 day, 13:58:56 time: 0.9976 data_time: 0.0049 memory: 8462 loss: 0.0238 decode.loss_ce: 0.0131 decode.acc_seg: 99.3656 aux.loss_ce: 0.0107 aux.acc_seg: 98.9552 +04/16 14:31:11 - mmengine - INFO - Iter(train) [ 22850/160000] base_lr: 8.6531e-05 lr: 3.1992e-07 eta: 1 day, 13:58:06 time: 0.9968 data_time: 0.0046 memory: 8462 loss: 0.0266 decode.loss_ce: 0.0149 decode.acc_seg: 99.5199 aux.loss_ce: 0.0117 aux.acc_seg: 98.6288 +04/16 14:32:01 - mmengine - INFO - Iter(train) [ 22900/160000] base_lr: 8.6499e-05 lr: 3.1980e-07 eta: 1 day, 13:57:16 time: 0.9954 data_time: 0.0044 memory: 8462 loss: 0.0187 decode.loss_ce: 0.0101 decode.acc_seg: 99.6670 aux.loss_ce: 0.0086 aux.acc_seg: 99.4059 +04/16 14:32:51 - mmengine - INFO - Iter(train) [ 22950/160000] base_lr: 8.6467e-05 lr: 3.1969e-07 eta: 1 day, 13:56:27 time: 0.9961 data_time: 0.0044 memory: 8462 loss: 0.0249 decode.loss_ce: 0.0138 decode.acc_seg: 99.1858 aux.loss_ce: 0.0111 aux.acc_seg: 98.5119 +04/16 14:33:41 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 14:33:41 - mmengine - INFO - Iter(train) [ 23000/160000] base_lr: 8.6436e-05 lr: 3.1957e-07 eta: 1 day, 13:55:37 time: 0.9963 data_time: 0.0046 memory: 8462 loss: 0.0245 decode.loss_ce: 0.0136 decode.acc_seg: 99.4303 aux.loss_ce: 0.0109 aux.acc_seg: 98.9567 +04/16 14:34:31 - mmengine - INFO - Iter(train) [ 23050/160000] base_lr: 8.6404e-05 lr: 3.1945e-07 eta: 1 day, 13:54:47 time: 0.9973 data_time: 0.0046 memory: 8462 loss: 0.0214 decode.loss_ce: 0.0117 decode.acc_seg: 99.4234 aux.loss_ce: 0.0097 aux.acc_seg: 98.9040 +04/16 14:35:20 - mmengine - INFO - Iter(train) [ 23100/160000] base_lr: 8.6373e-05 lr: 3.1934e-07 eta: 1 day, 13:53:57 time: 0.9962 data_time: 0.0043 memory: 8462 loss: 0.0206 decode.loss_ce: 0.0116 decode.acc_seg: 99.3435 aux.loss_ce: 0.0090 aux.acc_seg: 99.0650 +04/16 14:36:10 - mmengine - INFO - Iter(train) [ 23150/160000] base_lr: 8.6341e-05 lr: 3.1922e-07 eta: 1 day, 13:53:07 time: 0.9965 data_time: 0.0042 memory: 8462 loss: 0.0214 decode.loss_ce: 0.0115 decode.acc_seg: 99.4799 aux.loss_ce: 0.0100 aux.acc_seg: 98.9277 +04/16 14:37:00 - mmengine - INFO - Iter(train) [ 23200/160000] base_lr: 8.6310e-05 lr: 3.1910e-07 eta: 1 day, 13:52:18 time: 0.9983 data_time: 0.0044 memory: 8462 loss: 0.0297 decode.loss_ce: 0.0155 decode.acc_seg: 99.3931 aux.loss_ce: 0.0142 aux.acc_seg: 98.6179 +04/16 14:37:50 - mmengine - INFO - Iter(train) [ 23250/160000] base_lr: 8.6278e-05 lr: 3.1899e-07 eta: 1 day, 13:51:29 time: 0.9978 data_time: 0.0045 memory: 8462 loss: 0.0223 decode.loss_ce: 0.0127 decode.acc_seg: 99.5281 aux.loss_ce: 0.0096 aux.acc_seg: 98.7810 +04/16 14:38:40 - mmengine - INFO - Iter(train) [ 23300/160000] base_lr: 8.6247e-05 lr: 3.1887e-07 eta: 1 day, 13:50:39 time: 0.9974 data_time: 0.0043 memory: 8462 loss: 0.0237 decode.loss_ce: 0.0129 decode.acc_seg: 99.6302 aux.loss_ce: 0.0108 aux.acc_seg: 98.9265 +04/16 14:39:30 - mmengine - INFO - Iter(train) [ 23350/160000] base_lr: 8.6215e-05 lr: 3.1875e-07 eta: 1 day, 13:49:49 time: 0.9983 data_time: 0.0043 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0103 decode.acc_seg: 99.5640 aux.loss_ce: 0.0087 aux.acc_seg: 98.7467 +04/16 14:40:20 - mmengine - INFO - Iter(train) [ 23400/160000] base_lr: 8.6184e-05 lr: 3.1864e-07 eta: 1 day, 13:49:00 time: 0.9982 data_time: 0.0043 memory: 8462 loss: 0.0215 decode.loss_ce: 0.0118 decode.acc_seg: 99.4350 aux.loss_ce: 0.0097 aux.acc_seg: 98.9246 +04/16 14:41:10 - mmengine - INFO - Iter(train) [ 23450/160000] base_lr: 8.6152e-05 lr: 3.1852e-07 eta: 1 day, 13:48:10 time: 0.9973 data_time: 0.0047 memory: 8462 loss: 0.0239 decode.loss_ce: 0.0135 decode.acc_seg: 99.6048 aux.loss_ce: 0.0104 aux.acc_seg: 99.3065 +04/16 14:41:59 - mmengine - INFO - Iter(train) [ 23500/160000] base_lr: 8.6120e-05 lr: 3.1840e-07 eta: 1 day, 13:47:20 time: 0.9975 data_time: 0.0044 memory: 8462 loss: 0.0195 decode.loss_ce: 0.0104 decode.acc_seg: 99.5548 aux.loss_ce: 0.0091 aux.acc_seg: 99.1333 +04/16 14:42:49 - mmengine - INFO - Iter(train) [ 23550/160000] base_lr: 8.6089e-05 lr: 3.1829e-07 eta: 1 day, 13:46:31 time: 0.9973 data_time: 0.0044 memory: 8462 loss: 0.0206 decode.loss_ce: 0.0113 decode.acc_seg: 99.6290 aux.loss_ce: 0.0093 aux.acc_seg: 99.0456 +04/16 14:43:39 - mmengine - INFO - Iter(train) [ 23600/160000] base_lr: 8.6057e-05 lr: 3.1817e-07 eta: 1 day, 13:45:41 time: 0.9973 data_time: 0.0043 memory: 8462 loss: 0.0236 decode.loss_ce: 0.0132 decode.acc_seg: 99.5073 aux.loss_ce: 0.0105 aux.acc_seg: 99.1844 +04/16 14:44:29 - mmengine - INFO - Iter(train) [ 23650/160000] base_lr: 8.6026e-05 lr: 3.1806e-07 eta: 1 day, 13:44:52 time: 0.9975 data_time: 0.0046 memory: 8462 loss: 0.0267 decode.loss_ce: 0.0154 decode.acc_seg: 99.3158 aux.loss_ce: 0.0113 aux.acc_seg: 98.8556 +04/16 14:45:19 - mmengine - INFO - Iter(train) [ 23700/160000] base_lr: 8.5994e-05 lr: 3.1794e-07 eta: 1 day, 13:44:02 time: 0.9964 data_time: 0.0046 memory: 8462 loss: 0.0226 decode.loss_ce: 0.0121 decode.acc_seg: 99.5440 aux.loss_ce: 0.0104 aux.acc_seg: 98.9670 +04/16 14:46:09 - mmengine - INFO - Iter(train) [ 23750/160000] base_lr: 8.5963e-05 lr: 3.1782e-07 eta: 1 day, 13:43:12 time: 0.9973 data_time: 0.0045 memory: 8462 loss: 0.0221 decode.loss_ce: 0.0123 decode.acc_seg: 99.6754 aux.loss_ce: 0.0098 aux.acc_seg: 99.3891 +04/16 14:46:59 - mmengine - INFO - Iter(train) [ 23800/160000] base_lr: 8.5931e-05 lr: 3.1771e-07 eta: 1 day, 13:42:23 time: 0.9976 data_time: 0.0042 memory: 8462 loss: 0.0253 decode.loss_ce: 0.0143 decode.acc_seg: 99.6031 aux.loss_ce: 0.0109 aux.acc_seg: 98.8400 +04/16 14:47:49 - mmengine - INFO - Iter(train) [ 23850/160000] base_lr: 8.5900e-05 lr: 3.1759e-07 eta: 1 day, 13:41:33 time: 0.9988 data_time: 0.0047 memory: 8462 loss: 0.0225 decode.loss_ce: 0.0122 decode.acc_seg: 99.3511 aux.loss_ce: 0.0103 aux.acc_seg: 98.2054 +04/16 14:48:38 - mmengine - INFO - Iter(train) [ 23900/160000] base_lr: 8.5868e-05 lr: 3.1747e-07 eta: 1 day, 13:40:44 time: 0.9977 data_time: 0.0045 memory: 8462 loss: 0.0225 decode.loss_ce: 0.0124 decode.acc_seg: 99.4944 aux.loss_ce: 0.0101 aux.acc_seg: 98.6906 +04/16 14:49:28 - mmengine - INFO - Iter(train) [ 23950/160000] base_lr: 8.5837e-05 lr: 3.1736e-07 eta: 1 day, 13:39:54 time: 0.9983 data_time: 0.0043 memory: 8462 loss: 0.0229 decode.loss_ce: 0.0126 decode.acc_seg: 99.4253 aux.loss_ce: 0.0103 aux.acc_seg: 98.7873 +04/16 14:50:18 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 14:50:18 - mmengine - INFO - Iter(train) [ 24000/160000] base_lr: 8.5805e-05 lr: 3.1724e-07 eta: 1 day, 13:39:05 time: 0.9982 data_time: 0.0044 memory: 8462 loss: 0.0232 decode.loss_ce: 0.0121 decode.acc_seg: 99.3252 aux.loss_ce: 0.0111 aux.acc_seg: 98.3219 +04/16 14:51:08 - mmengine - INFO - Iter(train) [ 24050/160000] base_lr: 8.5773e-05 lr: 3.1712e-07 eta: 1 day, 13:38:16 time: 0.9982 data_time: 0.0045 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0095 decode.acc_seg: 99.6187 aux.loss_ce: 0.0083 aux.acc_seg: 99.1653 +04/16 14:51:58 - mmengine - INFO - Iter(train) [ 24100/160000] base_lr: 8.5742e-05 lr: 3.1701e-07 eta: 1 day, 13:37:26 time: 0.9977 data_time: 0.0045 memory: 8462 loss: 0.0226 decode.loss_ce: 0.0124 decode.acc_seg: 99.3769 aux.loss_ce: 0.0102 aux.acc_seg: 98.9765 +04/16 14:52:48 - mmengine - INFO - Iter(train) [ 24150/160000] base_lr: 8.5710e-05 lr: 3.1689e-07 eta: 1 day, 13:36:37 time: 0.9972 data_time: 0.0051 memory: 8462 loss: 0.0222 decode.loss_ce: 0.0121 decode.acc_seg: 99.5529 aux.loss_ce: 0.0101 aux.acc_seg: 98.9731 +04/16 14:53:38 - mmengine - INFO - Iter(train) [ 24200/160000] base_lr: 8.5679e-05 lr: 3.1677e-07 eta: 1 day, 13:35:47 time: 0.9972 data_time: 0.0050 memory: 8462 loss: 0.0230 decode.loss_ce: 0.0127 decode.acc_seg: 98.8163 aux.loss_ce: 0.0102 aux.acc_seg: 97.9467 +04/16 14:54:28 - mmengine - INFO - Iter(train) [ 24250/160000] base_lr: 8.5647e-05 lr: 3.1666e-07 eta: 1 day, 13:34:58 time: 0.9983 data_time: 0.0043 memory: 8462 loss: 0.0217 decode.loss_ce: 0.0118 decode.acc_seg: 99.3769 aux.loss_ce: 0.0099 aux.acc_seg: 98.8825 +04/16 14:55:18 - mmengine - INFO - Iter(train) [ 24300/160000] base_lr: 8.5616e-05 lr: 3.1654e-07 eta: 1 day, 13:34:08 time: 0.9964 data_time: 0.0046 memory: 8462 loss: 0.0231 decode.loss_ce: 0.0127 decode.acc_seg: 99.4455 aux.loss_ce: 0.0104 aux.acc_seg: 98.9311 +04/16 14:56:08 - mmengine - INFO - Iter(train) [ 24350/160000] base_lr: 8.5584e-05 lr: 3.1642e-07 eta: 1 day, 13:33:19 time: 0.9990 data_time: 0.0050 memory: 8462 loss: 0.0217 decode.loss_ce: 0.0119 decode.acc_seg: 99.4951 aux.loss_ce: 0.0098 aux.acc_seg: 98.9597 +04/16 14:56:58 - mmengine - INFO - Iter(train) [ 24400/160000] base_lr: 8.5553e-05 lr: 3.1631e-07 eta: 1 day, 13:32:29 time: 0.9988 data_time: 0.0044 memory: 8462 loss: 0.0231 decode.loss_ce: 0.0128 decode.acc_seg: 99.5234 aux.loss_ce: 0.0103 aux.acc_seg: 98.9836 +04/16 14:57:47 - mmengine - INFO - Iter(train) [ 24450/160000] base_lr: 8.5521e-05 lr: 3.1619e-07 eta: 1 day, 13:31:40 time: 0.9968 data_time: 0.0043 memory: 8462 loss: 0.0225 decode.loss_ce: 0.0121 decode.acc_seg: 99.2128 aux.loss_ce: 0.0104 aux.acc_seg: 98.7537 +04/16 14:58:37 - mmengine - INFO - Iter(train) [ 24500/160000] base_lr: 8.5489e-05 lr: 3.1607e-07 eta: 1 day, 13:30:50 time: 0.9977 data_time: 0.0044 memory: 8462 loss: 0.0211 decode.loss_ce: 0.0116 decode.acc_seg: 99.5535 aux.loss_ce: 0.0095 aux.acc_seg: 99.1903 +04/16 14:59:27 - mmengine - INFO - Iter(train) [ 24550/160000] base_lr: 8.5458e-05 lr: 3.1596e-07 eta: 1 day, 13:30:01 time: 0.9969 data_time: 0.0046 memory: 8462 loss: 0.0248 decode.loss_ce: 0.0129 decode.acc_seg: 99.3225 aux.loss_ce: 0.0119 aux.acc_seg: 98.6050 +04/16 15:00:17 - mmengine - INFO - Iter(train) [ 24600/160000] base_lr: 8.5426e-05 lr: 3.1584e-07 eta: 1 day, 13:29:11 time: 0.9979 data_time: 0.0044 memory: 8462 loss: 0.0233 decode.loss_ce: 0.0132 decode.acc_seg: 99.6574 aux.loss_ce: 0.0101 aux.acc_seg: 99.1491 +04/16 15:01:07 - mmengine - INFO - Iter(train) [ 24650/160000] base_lr: 8.5395e-05 lr: 3.1572e-07 eta: 1 day, 13:28:22 time: 0.9986 data_time: 0.0041 memory: 8462 loss: 0.0204 decode.loss_ce: 0.0109 decode.acc_seg: 99.5369 aux.loss_ce: 0.0095 aux.acc_seg: 99.2842 +04/16 15:01:57 - mmengine - INFO - Iter(train) [ 24700/160000] base_lr: 8.5363e-05 lr: 3.1561e-07 eta: 1 day, 13:27:32 time: 0.9979 data_time: 0.0049 memory: 8462 loss: 0.0221 decode.loss_ce: 0.0124 decode.acc_seg: 99.7446 aux.loss_ce: 0.0096 aux.acc_seg: 99.3711 +04/16 15:02:47 - mmengine - INFO - Iter(train) [ 24750/160000] base_lr: 8.5332e-05 lr: 3.1549e-07 eta: 1 day, 13:26:43 time: 0.9980 data_time: 0.0045 memory: 8462 loss: 0.0205 decode.loss_ce: 0.0113 decode.acc_seg: 99.5089 aux.loss_ce: 0.0093 aux.acc_seg: 99.0139 +04/16 15:03:37 - mmengine - INFO - Iter(train) [ 24800/160000] base_lr: 8.5300e-05 lr: 3.1537e-07 eta: 1 day, 13:25:53 time: 0.9985 data_time: 0.0048 memory: 8462 loss: 0.0224 decode.loss_ce: 0.0123 decode.acc_seg: 99.3095 aux.loss_ce: 0.0101 aux.acc_seg: 98.6654 +04/16 15:04:27 - mmengine - INFO - Iter(train) [ 24850/160000] base_lr: 8.5269e-05 lr: 3.1526e-07 eta: 1 day, 13:25:04 time: 0.9996 data_time: 0.0046 memory: 8462 loss: 0.0263 decode.loss_ce: 0.0137 decode.acc_seg: 99.4915 aux.loss_ce: 0.0125 aux.acc_seg: 98.6107 +04/16 15:05:17 - mmengine - INFO - Iter(train) [ 24900/160000] base_lr: 8.5237e-05 lr: 3.1514e-07 eta: 1 day, 13:24:15 time: 0.9976 data_time: 0.0043 memory: 8462 loss: 0.0315 decode.loss_ce: 0.0190 decode.acc_seg: 99.6281 aux.loss_ce: 0.0124 aux.acc_seg: 99.4452 +04/16 15:06:07 - mmengine - INFO - Iter(train) [ 24950/160000] base_lr: 8.5206e-05 lr: 3.1502e-07 eta: 1 day, 13:23:26 time: 0.9985 data_time: 0.0046 memory: 8462 loss: 0.0228 decode.loss_ce: 0.0127 decode.acc_seg: 99.5264 aux.loss_ce: 0.0101 aux.acc_seg: 98.7545 +04/16 15:06:57 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 15:06:57 - mmengine - INFO - Iter(train) [ 25000/160000] base_lr: 8.5174e-05 lr: 3.1491e-07 eta: 1 day, 13:22:37 time: 0.9990 data_time: 0.0051 memory: 8462 loss: 0.0223 decode.loss_ce: 0.0123 decode.acc_seg: 99.5230 aux.loss_ce: 0.0100 aux.acc_seg: 98.9000 +04/16 15:07:46 - mmengine - INFO - Iter(train) [ 25050/160000] base_lr: 8.5142e-05 lr: 3.1479e-07 eta: 1 day, 13:21:47 time: 0.9994 data_time: 0.0046 memory: 8462 loss: 0.0261 decode.loss_ce: 0.0138 decode.acc_seg: 98.9008 aux.loss_ce: 0.0124 aux.acc_seg: 98.2924 +04/16 15:08:36 - mmengine - INFO - Iter(train) [ 25100/160000] base_lr: 8.5111e-05 lr: 3.1467e-07 eta: 1 day, 13:20:58 time: 0.9984 data_time: 0.0048 memory: 8462 loss: 0.0200 decode.loss_ce: 0.0111 decode.acc_seg: 99.5007 aux.loss_ce: 0.0089 aux.acc_seg: 99.1041 +04/16 15:09:26 - mmengine - INFO - Iter(train) [ 25150/160000] base_lr: 8.5079e-05 lr: 3.1456e-07 eta: 1 day, 13:20:08 time: 0.9994 data_time: 0.0044 memory: 8462 loss: 0.0200 decode.loss_ce: 0.0103 decode.acc_seg: 99.6618 aux.loss_ce: 0.0096 aux.acc_seg: 99.2456 +04/16 15:10:16 - mmengine - INFO - Iter(train) [ 25200/160000] base_lr: 8.5048e-05 lr: 3.1444e-07 eta: 1 day, 13:19:19 time: 0.9974 data_time: 0.0043 memory: 8462 loss: 0.0228 decode.loss_ce: 0.0127 decode.acc_seg: 99.2535 aux.loss_ce: 0.0100 aux.acc_seg: 98.7143 +04/16 15:11:06 - mmengine - INFO - Iter(train) [ 25250/160000] base_lr: 8.5016e-05 lr: 3.1432e-07 eta: 1 day, 13:18:29 time: 0.9972 data_time: 0.0046 memory: 8462 loss: 0.0202 decode.loss_ce: 0.0105 decode.acc_seg: 99.4162 aux.loss_ce: 0.0098 aux.acc_seg: 99.0307 +04/16 15:11:56 - mmengine - INFO - Iter(train) [ 25300/160000] base_lr: 8.4985e-05 lr: 3.1421e-07 eta: 1 day, 13:17:40 time: 0.9987 data_time: 0.0045 memory: 8462 loss: 0.0214 decode.loss_ce: 0.0116 decode.acc_seg: 99.5352 aux.loss_ce: 0.0098 aux.acc_seg: 99.0580 +04/16 15:12:46 - mmengine - INFO - Iter(train) [ 25350/160000] base_lr: 8.4953e-05 lr: 3.1409e-07 eta: 1 day, 13:16:51 time: 0.9987 data_time: 0.0050 memory: 8462 loss: 0.0200 decode.loss_ce: 0.0108 decode.acc_seg: 99.6500 aux.loss_ce: 0.0092 aux.acc_seg: 99.0580 +04/16 15:13:36 - mmengine - INFO - Iter(train) [ 25400/160000] base_lr: 8.4922e-05 lr: 3.1397e-07 eta: 1 day, 13:16:01 time: 0.9988 data_time: 0.0044 memory: 8462 loss: 0.0226 decode.loss_ce: 0.0130 decode.acc_seg: 99.6540 aux.loss_ce: 0.0096 aux.acc_seg: 99.1856 +04/16 15:14:26 - mmengine - INFO - Iter(train) [ 25450/160000] base_lr: 8.4890e-05 lr: 3.1386e-07 eta: 1 day, 13:15:12 time: 0.9985 data_time: 0.0044 memory: 8462 loss: 0.0219 decode.loss_ce: 0.0122 decode.acc_seg: 99.4015 aux.loss_ce: 0.0097 aux.acc_seg: 98.8869 +04/16 15:15:16 - mmengine - INFO - Iter(train) [ 25500/160000] base_lr: 8.4859e-05 lr: 3.1374e-07 eta: 1 day, 13:14:22 time: 0.9982 data_time: 0.0047 memory: 8462 loss: 0.0220 decode.loss_ce: 0.0115 decode.acc_seg: 99.6840 aux.loss_ce: 0.0105 aux.acc_seg: 99.2466 +04/16 15:16:06 - mmengine - INFO - Iter(train) [ 25550/160000] base_lr: 8.4827e-05 lr: 3.1362e-07 eta: 1 day, 13:13:33 time: 0.9979 data_time: 0.0046 memory: 8462 loss: 0.0258 decode.loss_ce: 0.0139 decode.acc_seg: 99.6510 aux.loss_ce: 0.0119 aux.acc_seg: 99.0093 +04/16 15:16:56 - mmengine - INFO - Iter(train) [ 25600/160000] base_lr: 8.4795e-05 lr: 3.1351e-07 eta: 1 day, 13:12:44 time: 0.9991 data_time: 0.0044 memory: 8462 loss: 0.0200 decode.loss_ce: 0.0105 decode.acc_seg: 99.4869 aux.loss_ce: 0.0095 aux.acc_seg: 98.7789 +04/16 15:17:45 - mmengine - INFO - Iter(train) [ 25650/160000] base_lr: 8.4764e-05 lr: 3.1339e-07 eta: 1 day, 13:11:54 time: 0.9993 data_time: 0.0046 memory: 8462 loss: 0.0213 decode.loss_ce: 0.0117 decode.acc_seg: 99.4572 aux.loss_ce: 0.0097 aux.acc_seg: 98.7028 +04/16 15:18:35 - mmengine - INFO - Iter(train) [ 25700/160000] base_lr: 8.4732e-05 lr: 3.1327e-07 eta: 1 day, 13:11:05 time: 0.9985 data_time: 0.0046 memory: 8462 loss: 0.0230 decode.loss_ce: 0.0129 decode.acc_seg: 99.4556 aux.loss_ce: 0.0101 aux.acc_seg: 98.9740 +04/16 15:19:25 - mmengine - INFO - Iter(train) [ 25750/160000] base_lr: 8.4701e-05 lr: 3.1316e-07 eta: 1 day, 13:10:15 time: 0.9978 data_time: 0.0047 memory: 8462 loss: 0.0223 decode.loss_ce: 0.0120 decode.acc_seg: 99.6185 aux.loss_ce: 0.0102 aux.acc_seg: 99.2373 +04/16 15:20:15 - mmengine - INFO - Iter(train) [ 25800/160000] base_lr: 8.4669e-05 lr: 3.1304e-07 eta: 1 day, 13:09:26 time: 0.9973 data_time: 0.0045 memory: 8462 loss: 0.0229 decode.loss_ce: 0.0126 decode.acc_seg: 99.5419 aux.loss_ce: 0.0103 aux.acc_seg: 99.0185 +04/16 15:21:05 - mmengine - INFO - Iter(train) [ 25850/160000] base_lr: 8.4638e-05 lr: 3.1292e-07 eta: 1 day, 13:08:36 time: 0.9979 data_time: 0.0046 memory: 8462 loss: 0.0196 decode.loss_ce: 0.0107 decode.acc_seg: 99.6773 aux.loss_ce: 0.0090 aux.acc_seg: 99.1724 +04/16 15:21:55 - mmengine - INFO - Iter(train) [ 25900/160000] base_lr: 8.4606e-05 lr: 3.1281e-07 eta: 1 day, 13:07:47 time: 0.9980 data_time: 0.0044 memory: 8462 loss: 0.0210 decode.loss_ce: 0.0115 decode.acc_seg: 99.6471 aux.loss_ce: 0.0095 aux.acc_seg: 99.1873 +04/16 15:22:45 - mmengine - INFO - Iter(train) [ 25950/160000] base_lr: 8.4575e-05 lr: 3.1269e-07 eta: 1 day, 13:06:58 time: 0.9990 data_time: 0.0043 memory: 8462 loss: 0.0230 decode.loss_ce: 0.0125 decode.acc_seg: 99.7402 aux.loss_ce: 0.0105 aux.acc_seg: 99.2634 +04/16 15:23:35 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 15:23:35 - mmengine - INFO - Iter(train) [ 26000/160000] base_lr: 8.4543e-05 lr: 3.1257e-07 eta: 1 day, 13:06:08 time: 0.9976 data_time: 0.0045 memory: 8462 loss: 0.0235 decode.loss_ce: 0.0133 decode.acc_seg: 99.7139 aux.loss_ce: 0.0103 aux.acc_seg: 99.3311 +04/16 15:24:25 - mmengine - INFO - Iter(train) [ 26050/160000] base_lr: 8.4512e-05 lr: 3.1246e-07 eta: 1 day, 13:05:18 time: 0.9977 data_time: 0.0045 memory: 8462 loss: 0.0238 decode.loss_ce: 0.0136 decode.acc_seg: 99.5901 aux.loss_ce: 0.0102 aux.acc_seg: 99.0496 +04/16 15:25:15 - mmengine - INFO - Iter(train) [ 26100/160000] base_lr: 8.4480e-05 lr: 3.1234e-07 eta: 1 day, 13:04:29 time: 0.9992 data_time: 0.0043 memory: 8462 loss: 0.0218 decode.loss_ce: 0.0117 decode.acc_seg: 99.5930 aux.loss_ce: 0.0101 aux.acc_seg: 99.1802 +04/16 15:26:05 - mmengine - INFO - Iter(train) [ 26150/160000] base_lr: 8.4448e-05 lr: 3.1222e-07 eta: 1 day, 13:03:40 time: 0.9985 data_time: 0.0049 memory: 8462 loss: 0.0203 decode.loss_ce: 0.0109 decode.acc_seg: 99.6428 aux.loss_ce: 0.0094 aux.acc_seg: 99.0723 +04/16 15:26:55 - mmengine - INFO - Iter(train) [ 26200/160000] base_lr: 8.4417e-05 lr: 3.1211e-07 eta: 1 day, 13:02:51 time: 0.9999 data_time: 0.0044 memory: 8462 loss: 0.0221 decode.loss_ce: 0.0119 decode.acc_seg: 99.4940 aux.loss_ce: 0.0102 aux.acc_seg: 99.0963 +04/16 15:27:45 - mmengine - INFO - Iter(train) [ 26250/160000] base_lr: 8.4385e-05 lr: 3.1199e-07 eta: 1 day, 13:02:01 time: 0.9977 data_time: 0.0045 memory: 8462 loss: 0.0201 decode.loss_ce: 0.0111 decode.acc_seg: 99.3864 aux.loss_ce: 0.0090 aux.acc_seg: 98.7890 +04/16 15:28:35 - mmengine - INFO - Iter(train) [ 26300/160000] base_lr: 8.4354e-05 lr: 3.1187e-07 eta: 1 day, 13:01:12 time: 0.9996 data_time: 0.0043 memory: 8462 loss: 0.0210 decode.loss_ce: 0.0111 decode.acc_seg: 99.6614 aux.loss_ce: 0.0099 aux.acc_seg: 99.0555 +04/16 15:29:24 - mmengine - INFO - Iter(train) [ 26350/160000] base_lr: 8.4322e-05 lr: 3.1176e-07 eta: 1 day, 13:00:22 time: 0.9977 data_time: 0.0047 memory: 8462 loss: 0.0213 decode.loss_ce: 0.0115 decode.acc_seg: 99.6431 aux.loss_ce: 0.0098 aux.acc_seg: 99.2271 +04/16 15:30:14 - mmengine - INFO - Iter(train) [ 26400/160000] base_lr: 8.4291e-05 lr: 3.1164e-07 eta: 1 day, 12:59:33 time: 0.9981 data_time: 0.0045 memory: 8462 loss: 0.0212 decode.loss_ce: 0.0118 decode.acc_seg: 99.6040 aux.loss_ce: 0.0094 aux.acc_seg: 98.8987 +04/16 15:31:04 - mmengine - INFO - Iter(train) [ 26450/160000] base_lr: 8.4259e-05 lr: 3.1152e-07 eta: 1 day, 12:58:44 time: 0.9987 data_time: 0.0042 memory: 8462 loss: 0.0203 decode.loss_ce: 0.0112 decode.acc_seg: 99.5342 aux.loss_ce: 0.0091 aux.acc_seg: 99.0847 +04/16 15:31:54 - mmengine - INFO - Iter(train) [ 26500/160000] base_lr: 8.4228e-05 lr: 3.1141e-07 eta: 1 day, 12:57:54 time: 1.0006 data_time: 0.0047 memory: 8462 loss: 0.0214 decode.loss_ce: 0.0121 decode.acc_seg: 99.5789 aux.loss_ce: 0.0093 aux.acc_seg: 99.3631 +04/16 15:32:44 - mmengine - INFO - Iter(train) [ 26550/160000] base_lr: 8.4196e-05 lr: 3.1129e-07 eta: 1 day, 12:57:05 time: 0.9983 data_time: 0.0044 memory: 8462 loss: 0.0231 decode.loss_ce: 0.0126 decode.acc_seg: 99.3521 aux.loss_ce: 0.0105 aux.acc_seg: 99.0625 +04/16 15:33:34 - mmengine - INFO - Iter(train) [ 26600/160000] base_lr: 8.4165e-05 lr: 3.1117e-07 eta: 1 day, 12:56:15 time: 0.9991 data_time: 0.0042 memory: 8462 loss: 0.0192 decode.loss_ce: 0.0105 decode.acc_seg: 99.3067 aux.loss_ce: 0.0087 aux.acc_seg: 98.6843 +04/16 15:34:24 - mmengine - INFO - Iter(train) [ 26650/160000] base_lr: 8.4133e-05 lr: 3.1106e-07 eta: 1 day, 12:55:26 time: 0.9978 data_time: 0.0043 memory: 8462 loss: 0.0221 decode.loss_ce: 0.0119 decode.acc_seg: 99.5811 aux.loss_ce: 0.0102 aux.acc_seg: 99.1932 +04/16 15:35:14 - mmengine - INFO - Iter(train) [ 26700/160000] base_lr: 8.4101e-05 lr: 3.1094e-07 eta: 1 day, 12:54:37 time: 0.9995 data_time: 0.0048 memory: 8462 loss: 0.0243 decode.loss_ce: 0.0137 decode.acc_seg: 99.4766 aux.loss_ce: 0.0107 aux.acc_seg: 98.7249 +04/16 15:36:04 - mmengine - INFO - Iter(train) [ 26750/160000] base_lr: 8.4070e-05 lr: 3.1082e-07 eta: 1 day, 12:53:47 time: 0.9989 data_time: 0.0048 memory: 8462 loss: 0.0192 decode.loss_ce: 0.0105 decode.acc_seg: 99.5125 aux.loss_ce: 0.0087 aux.acc_seg: 98.9407 +04/16 15:36:54 - mmengine - INFO - Iter(train) [ 26800/160000] base_lr: 8.4038e-05 lr: 3.1071e-07 eta: 1 day, 12:52:58 time: 0.9988 data_time: 0.0046 memory: 8462 loss: 0.0237 decode.loss_ce: 0.0132 decode.acc_seg: 99.5655 aux.loss_ce: 0.0104 aux.acc_seg: 98.9796 +04/16 15:37:44 - mmengine - INFO - Iter(train) [ 26850/160000] base_lr: 8.4007e-05 lr: 3.1059e-07 eta: 1 day, 12:52:09 time: 0.9972 data_time: 0.0042 memory: 8462 loss: 0.0216 decode.loss_ce: 0.0119 decode.acc_seg: 99.6017 aux.loss_ce: 0.0097 aux.acc_seg: 99.2483 +04/16 15:38:34 - mmengine - INFO - Iter(train) [ 26900/160000] base_lr: 8.3975e-05 lr: 3.1047e-07 eta: 1 day, 12:51:19 time: 0.9973 data_time: 0.0045 memory: 8462 loss: 0.0196 decode.loss_ce: 0.0106 decode.acc_seg: 99.6767 aux.loss_ce: 0.0090 aux.acc_seg: 99.3874 +04/16 15:39:24 - mmengine - INFO - Iter(train) [ 26950/160000] base_lr: 8.3944e-05 lr: 3.1036e-07 eta: 1 day, 12:50:30 time: 0.9977 data_time: 0.0043 memory: 8462 loss: 0.0213 decode.loss_ce: 0.0115 decode.acc_seg: 99.6105 aux.loss_ce: 0.0098 aux.acc_seg: 98.9498 +04/16 15:40:14 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 15:40:14 - mmengine - INFO - Iter(train) [ 27000/160000] base_lr: 8.3912e-05 lr: 3.1024e-07 eta: 1 day, 12:49:40 time: 0.9982 data_time: 0.0043 memory: 8462 loss: 0.0239 decode.loss_ce: 0.0132 decode.acc_seg: 99.3624 aux.loss_ce: 0.0107 aux.acc_seg: 98.8785 +04/16 15:41:04 - mmengine - INFO - Iter(train) [ 27050/160000] base_lr: 8.3881e-05 lr: 3.1012e-07 eta: 1 day, 12:48:51 time: 0.9987 data_time: 0.0046 memory: 8462 loss: 0.0214 decode.loss_ce: 0.0112 decode.acc_seg: 99.6258 aux.loss_ce: 0.0102 aux.acc_seg: 99.1985 +04/16 15:41:53 - mmengine - INFO - Iter(train) [ 27100/160000] base_lr: 8.3849e-05 lr: 3.1001e-07 eta: 1 day, 12:48:02 time: 0.9979 data_time: 0.0046 memory: 8462 loss: 0.0257 decode.loss_ce: 0.0140 decode.acc_seg: 99.5079 aux.loss_ce: 0.0117 aux.acc_seg: 99.1341 +04/16 15:42:43 - mmengine - INFO - Iter(train) [ 27150/160000] base_lr: 8.3818e-05 lr: 3.0989e-07 eta: 1 day, 12:47:12 time: 0.9989 data_time: 0.0048 memory: 8462 loss: 0.0194 decode.loss_ce: 0.0108 decode.acc_seg: 99.4413 aux.loss_ce: 0.0086 aux.acc_seg: 98.9935 +04/16 15:43:33 - mmengine - INFO - Iter(train) [ 27200/160000] base_lr: 8.3786e-05 lr: 3.0977e-07 eta: 1 day, 12:46:23 time: 0.9989 data_time: 0.0043 memory: 8462 loss: 0.0222 decode.loss_ce: 0.0119 decode.acc_seg: 99.2996 aux.loss_ce: 0.0103 aux.acc_seg: 98.4583 +04/16 15:44:23 - mmengine - INFO - Iter(train) [ 27250/160000] base_lr: 8.3754e-05 lr: 3.0966e-07 eta: 1 day, 12:45:34 time: 0.9994 data_time: 0.0044 memory: 8462 loss: 0.0209 decode.loss_ce: 0.0110 decode.acc_seg: 99.5111 aux.loss_ce: 0.0099 aux.acc_seg: 98.9071 +04/16 15:45:13 - mmengine - INFO - Iter(train) [ 27300/160000] base_lr: 8.3723e-05 lr: 3.0954e-07 eta: 1 day, 12:44:44 time: 0.9976 data_time: 0.0043 memory: 8462 loss: 0.0232 decode.loss_ce: 0.0123 decode.acc_seg: 99.5283 aux.loss_ce: 0.0110 aux.acc_seg: 98.9485 +04/16 15:46:03 - mmengine - INFO - Iter(train) [ 27350/160000] base_lr: 8.3691e-05 lr: 3.0942e-07 eta: 1 day, 12:43:55 time: 0.9994 data_time: 0.0044 memory: 8462 loss: 0.0216 decode.loss_ce: 0.0113 decode.acc_seg: 99.7271 aux.loss_ce: 0.0103 aux.acc_seg: 99.2552 +04/16 15:46:53 - mmengine - INFO - Iter(train) [ 27400/160000] base_lr: 8.3660e-05 lr: 3.0931e-07 eta: 1 day, 12:43:05 time: 0.9980 data_time: 0.0042 memory: 8462 loss: 0.0192 decode.loss_ce: 0.0105 decode.acc_seg: 99.6401 aux.loss_ce: 0.0087 aux.acc_seg: 99.2111 +04/16 15:47:43 - mmengine - INFO - Iter(train) [ 27450/160000] base_lr: 8.3628e-05 lr: 3.0919e-07 eta: 1 day, 12:42:16 time: 0.9990 data_time: 0.0043 memory: 8462 loss: 0.0186 decode.loss_ce: 0.0101 decode.acc_seg: 99.7450 aux.loss_ce: 0.0085 aux.acc_seg: 99.3895 +04/16 15:48:33 - mmengine - INFO - Iter(train) [ 27500/160000] base_lr: 8.3597e-05 lr: 3.0907e-07 eta: 1 day, 12:41:26 time: 0.9986 data_time: 0.0041 memory: 8462 loss: 0.0212 decode.loss_ce: 0.0110 decode.acc_seg: 99.5293 aux.loss_ce: 0.0102 aux.acc_seg: 98.3387 +04/16 15:49:23 - mmengine - INFO - Iter(train) [ 27550/160000] base_lr: 8.3565e-05 lr: 3.0896e-07 eta: 1 day, 12:40:37 time: 0.9984 data_time: 0.0047 memory: 8462 loss: 0.0226 decode.loss_ce: 0.0123 decode.acc_seg: 99.7822 aux.loss_ce: 0.0103 aux.acc_seg: 99.4503 +04/16 15:50:13 - mmengine - INFO - Iter(train) [ 27600/160000] base_lr: 8.3534e-05 lr: 3.0884e-07 eta: 1 day, 12:39:48 time: 0.9984 data_time: 0.0045 memory: 8462 loss: 0.0252 decode.loss_ce: 0.0140 decode.acc_seg: 99.2548 aux.loss_ce: 0.0112 aux.acc_seg: 98.5176 +04/16 15:51:03 - mmengine - INFO - Iter(train) [ 27650/160000] base_lr: 8.3502e-05 lr: 3.0872e-07 eta: 1 day, 12:38:58 time: 0.9982 data_time: 0.0045 memory: 8462 loss: 0.0257 decode.loss_ce: 0.0153 decode.acc_seg: 99.4677 aux.loss_ce: 0.0105 aux.acc_seg: 98.7925 +04/16 15:51:53 - mmengine - INFO - Iter(train) [ 27700/160000] base_lr: 8.3471e-05 lr: 3.0861e-07 eta: 1 day, 12:38:09 time: 0.9994 data_time: 0.0048 memory: 8462 loss: 0.0206 decode.loss_ce: 0.0112 decode.acc_seg: 99.8020 aux.loss_ce: 0.0095 aux.acc_seg: 99.4364 +04/16 15:52:43 - mmengine - INFO - Iter(train) [ 27750/160000] base_lr: 8.3439e-05 lr: 3.0849e-07 eta: 1 day, 12:37:20 time: 0.9992 data_time: 0.0043 memory: 8462 loss: 0.0203 decode.loss_ce: 0.0105 decode.acc_seg: 99.5148 aux.loss_ce: 0.0098 aux.acc_seg: 99.0330 +04/16 15:53:33 - mmengine - INFO - Iter(train) [ 27800/160000] base_lr: 8.3407e-05 lr: 3.0837e-07 eta: 1 day, 12:36:30 time: 0.9991 data_time: 0.0044 memory: 8462 loss: 0.0213 decode.loss_ce: 0.0112 decode.acc_seg: 99.4135 aux.loss_ce: 0.0102 aux.acc_seg: 98.6155 +04/16 15:54:23 - mmengine - INFO - Iter(train) [ 27850/160000] base_lr: 8.3376e-05 lr: 3.0826e-07 eta: 1 day, 12:35:41 time: 0.9990 data_time: 0.0047 memory: 8462 loss: 0.0209 decode.loss_ce: 0.0109 decode.acc_seg: 99.5617 aux.loss_ce: 0.0100 aux.acc_seg: 99.0740 +04/16 15:55:13 - mmengine - INFO - Iter(train) [ 27900/160000] base_lr: 8.3344e-05 lr: 3.0814e-07 eta: 1 day, 12:34:52 time: 0.9997 data_time: 0.0049 memory: 8462 loss: 0.0212 decode.loss_ce: 0.0115 decode.acc_seg: 99.6744 aux.loss_ce: 0.0096 aux.acc_seg: 99.0763 +04/16 15:56:03 - mmengine - INFO - Iter(train) [ 27950/160000] base_lr: 8.3313e-05 lr: 3.0802e-07 eta: 1 day, 12:34:02 time: 0.9996 data_time: 0.0047 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0097 decode.acc_seg: 99.7650 aux.loss_ce: 0.0077 aux.acc_seg: 99.2826 +04/16 15:56:52 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 15:56:52 - mmengine - INFO - Iter(train) [ 28000/160000] base_lr: 8.3281e-05 lr: 3.0791e-07 eta: 1 day, 12:33:13 time: 0.9996 data_time: 0.0046 memory: 8462 loss: 0.0236 decode.loss_ce: 0.0131 decode.acc_seg: 99.3725 aux.loss_ce: 0.0106 aux.acc_seg: 98.1825 +04/16 15:57:42 - mmengine - INFO - Iter(train) [ 28050/160000] base_lr: 8.3250e-05 lr: 3.0779e-07 eta: 1 day, 12:32:24 time: 0.9979 data_time: 0.0045 memory: 8462 loss: 0.0210 decode.loss_ce: 0.0116 decode.acc_seg: 99.7627 aux.loss_ce: 0.0093 aux.acc_seg: 99.5249 +04/16 15:58:32 - mmengine - INFO - Iter(train) [ 28100/160000] base_lr: 8.3218e-05 lr: 3.0767e-07 eta: 1 day, 12:31:34 time: 0.9990 data_time: 0.0048 memory: 8462 loss: 0.0209 decode.loss_ce: 0.0107 decode.acc_seg: 99.4638 aux.loss_ce: 0.0102 aux.acc_seg: 98.6311 +04/16 15:59:22 - mmengine - INFO - Iter(train) [ 28150/160000] base_lr: 8.3187e-05 lr: 3.0756e-07 eta: 1 day, 12:30:45 time: 0.9989 data_time: 0.0044 memory: 8462 loss: 0.0202 decode.loss_ce: 0.0109 decode.acc_seg: 99.5501 aux.loss_ce: 0.0093 aux.acc_seg: 98.6595 +04/16 16:00:12 - mmengine - INFO - Iter(train) [ 28200/160000] base_lr: 8.3155e-05 lr: 3.0744e-07 eta: 1 day, 12:29:55 time: 0.9978 data_time: 0.0047 memory: 8462 loss: 0.0199 decode.loss_ce: 0.0104 decode.acc_seg: 99.5430 aux.loss_ce: 0.0094 aux.acc_seg: 98.7978 +04/16 16:01:02 - mmengine - INFO - Iter(train) [ 28250/160000] base_lr: 8.3124e-05 lr: 3.0732e-07 eta: 1 day, 12:29:06 time: 0.9990 data_time: 0.0045 memory: 8462 loss: 0.0199 decode.loss_ce: 0.0103 decode.acc_seg: 99.5480 aux.loss_ce: 0.0095 aux.acc_seg: 99.3269 +04/16 16:01:52 - mmengine - INFO - Iter(train) [ 28300/160000] base_lr: 8.3092e-05 lr: 3.0721e-07 eta: 1 day, 12:28:17 time: 0.9994 data_time: 0.0045 memory: 8462 loss: 0.0217 decode.loss_ce: 0.0116 decode.acc_seg: 99.4350 aux.loss_ce: 0.0102 aux.acc_seg: 98.5794 +04/16 16:02:42 - mmengine - INFO - Iter(train) [ 28350/160000] base_lr: 8.3060e-05 lr: 3.0709e-07 eta: 1 day, 12:27:27 time: 0.9992 data_time: 0.0046 memory: 8462 loss: 0.0191 decode.loss_ce: 0.0104 decode.acc_seg: 99.5749 aux.loss_ce: 0.0087 aux.acc_seg: 99.1053 +04/16 16:03:32 - mmengine - INFO - Iter(train) [ 28400/160000] base_lr: 8.3029e-05 lr: 3.0697e-07 eta: 1 day, 12:26:38 time: 0.9986 data_time: 0.0044 memory: 8462 loss: 0.0205 decode.loss_ce: 0.0109 decode.acc_seg: 99.4873 aux.loss_ce: 0.0096 aux.acc_seg: 98.4062 +04/16 16:04:22 - mmengine - INFO - Iter(train) [ 28450/160000] base_lr: 8.2997e-05 lr: 3.0686e-07 eta: 1 day, 12:25:48 time: 0.9976 data_time: 0.0044 memory: 8462 loss: 0.0224 decode.loss_ce: 0.0116 decode.acc_seg: 99.2525 aux.loss_ce: 0.0109 aux.acc_seg: 98.8028 +04/16 16:05:12 - mmengine - INFO - Iter(train) [ 28500/160000] base_lr: 8.2966e-05 lr: 3.0674e-07 eta: 1 day, 12:24:59 time: 0.9990 data_time: 0.0044 memory: 8462 loss: 0.0223 decode.loss_ce: 0.0124 decode.acc_seg: 99.4305 aux.loss_ce: 0.0099 aux.acc_seg: 98.6626 +04/16 16:06:02 - mmengine - INFO - Iter(train) [ 28550/160000] base_lr: 8.2934e-05 lr: 3.0663e-07 eta: 1 day, 12:24:09 time: 0.9995 data_time: 0.0045 memory: 8462 loss: 0.0209 decode.loss_ce: 0.0115 decode.acc_seg: 99.7007 aux.loss_ce: 0.0094 aux.acc_seg: 99.1093 +04/16 16:06:52 - mmengine - INFO - Iter(train) [ 28600/160000] base_lr: 8.2903e-05 lr: 3.0651e-07 eta: 1 day, 12:23:20 time: 0.9982 data_time: 0.0045 memory: 8462 loss: 0.0208 decode.loss_ce: 0.0113 decode.acc_seg: 99.5255 aux.loss_ce: 0.0096 aux.acc_seg: 98.9964 +04/16 16:07:42 - mmengine - INFO - Iter(train) [ 28650/160000] base_lr: 8.2871e-05 lr: 3.0639e-07 eta: 1 day, 12:22:31 time: 0.9989 data_time: 0.0045 memory: 8462 loss: 0.0204 decode.loss_ce: 0.0108 decode.acc_seg: 99.5737 aux.loss_ce: 0.0095 aux.acc_seg: 98.8533 +04/16 16:08:32 - mmengine - INFO - Iter(train) [ 28700/160000] base_lr: 8.2840e-05 lr: 3.0628e-07 eta: 1 day, 12:21:41 time: 0.9995 data_time: 0.0043 memory: 8462 loss: 0.0214 decode.loss_ce: 0.0116 decode.acc_seg: 99.5186 aux.loss_ce: 0.0098 aux.acc_seg: 98.9225 +04/16 16:09:22 - mmengine - INFO - Iter(train) [ 28750/160000] base_lr: 8.2808e-05 lr: 3.0616e-07 eta: 1 day, 12:20:52 time: 0.9987 data_time: 0.0047 memory: 8462 loss: 0.0247 decode.loss_ce: 0.0135 decode.acc_seg: 99.5628 aux.loss_ce: 0.0112 aux.acc_seg: 98.9569 +04/16 16:10:12 - mmengine - INFO - Iter(train) [ 28800/160000] base_lr: 8.2777e-05 lr: 3.0604e-07 eta: 1 day, 12:20:03 time: 0.9998 data_time: 0.0046 memory: 8462 loss: 0.0196 decode.loss_ce: 0.0103 decode.acc_seg: 99.4392 aux.loss_ce: 0.0094 aux.acc_seg: 98.5386 +04/16 16:11:02 - mmengine - INFO - Iter(train) [ 28850/160000] base_lr: 8.2745e-05 lr: 3.0593e-07 eta: 1 day, 12:19:13 time: 0.9999 data_time: 0.0045 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0101 decode.acc_seg: 99.7364 aux.loss_ce: 0.0089 aux.acc_seg: 99.1697 +04/16 16:11:52 - mmengine - INFO - Iter(train) [ 28900/160000] base_lr: 8.2713e-05 lr: 3.0581e-07 eta: 1 day, 12:18:24 time: 0.9981 data_time: 0.0046 memory: 8462 loss: 0.0204 decode.loss_ce: 0.0105 decode.acc_seg: 99.4009 aux.loss_ce: 0.0099 aux.acc_seg: 98.6269 +04/16 16:12:41 - mmengine - INFO - Iter(train) [ 28950/160000] base_lr: 8.2682e-05 lr: 3.0569e-07 eta: 1 day, 12:17:34 time: 0.9990 data_time: 0.0047 memory: 8462 loss: 0.0218 decode.loss_ce: 0.0118 decode.acc_seg: 99.3357 aux.loss_ce: 0.0100 aux.acc_seg: 98.6673 +04/16 16:13:31 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 16:13:31 - mmengine - INFO - Iter(train) [ 29000/160000] base_lr: 8.2650e-05 lr: 3.0558e-07 eta: 1 day, 12:16:45 time: 0.9988 data_time: 0.0046 memory: 8462 loss: 0.0230 decode.loss_ce: 0.0123 decode.acc_seg: 99.5920 aux.loss_ce: 0.0107 aux.acc_seg: 99.0730 +04/16 16:14:21 - mmengine - INFO - Iter(train) [ 29050/160000] base_lr: 8.2619e-05 lr: 3.0546e-07 eta: 1 day, 12:15:56 time: 0.9994 data_time: 0.0041 memory: 8462 loss: 0.0212 decode.loss_ce: 0.0112 decode.acc_seg: 99.6086 aux.loss_ce: 0.0100 aux.acc_seg: 99.0261 +04/16 16:15:11 - mmengine - INFO - Iter(train) [ 29100/160000] base_lr: 8.2587e-05 lr: 3.0534e-07 eta: 1 day, 12:15:06 time: 0.9987 data_time: 0.0044 memory: 8462 loss: 0.0216 decode.loss_ce: 0.0115 decode.acc_seg: 99.7440 aux.loss_ce: 0.0101 aux.acc_seg: 99.4856 +04/16 16:16:01 - mmengine - INFO - Iter(train) [ 29150/160000] base_lr: 8.2556e-05 lr: 3.0523e-07 eta: 1 day, 12:14:17 time: 0.9984 data_time: 0.0043 memory: 8462 loss: 0.0202 decode.loss_ce: 0.0107 decode.acc_seg: 99.3132 aux.loss_ce: 0.0096 aux.acc_seg: 98.8047 +04/16 16:16:51 - mmengine - INFO - Iter(train) [ 29200/160000] base_lr: 8.2524e-05 lr: 3.0511e-07 eta: 1 day, 12:13:27 time: 0.9989 data_time: 0.0043 memory: 8462 loss: 0.0236 decode.loss_ce: 0.0133 decode.acc_seg: 99.7334 aux.loss_ce: 0.0103 aux.acc_seg: 99.4469 +04/16 16:17:41 - mmengine - INFO - Iter(train) [ 29250/160000] base_lr: 8.2493e-05 lr: 3.0499e-07 eta: 1 day, 12:12:38 time: 0.9984 data_time: 0.0045 memory: 8462 loss: 0.0209 decode.loss_ce: 0.0115 decode.acc_seg: 99.2943 aux.loss_ce: 0.0094 aux.acc_seg: 98.6359 +04/16 16:18:31 - mmengine - INFO - Iter(train) [ 29300/160000] base_lr: 8.2461e-05 lr: 3.0488e-07 eta: 1 day, 12:11:49 time: 0.9982 data_time: 0.0047 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0091 decode.acc_seg: 99.6403 aux.loss_ce: 0.0084 aux.acc_seg: 98.9340 +04/16 16:19:21 - mmengine - INFO - Iter(train) [ 29350/160000] base_lr: 8.2430e-05 lr: 3.0476e-07 eta: 1 day, 12:10:59 time: 0.9993 data_time: 0.0042 memory: 8462 loss: 0.0233 decode.loss_ce: 0.0126 decode.acc_seg: 99.5003 aux.loss_ce: 0.0107 aux.acc_seg: 99.0429 +04/16 16:20:11 - mmengine - INFO - Iter(train) [ 29400/160000] base_lr: 8.2398e-05 lr: 3.0464e-07 eta: 1 day, 12:10:10 time: 0.9985 data_time: 0.0047 memory: 8462 loss: 0.0200 decode.loss_ce: 0.0106 decode.acc_seg: 99.5972 aux.loss_ce: 0.0094 aux.acc_seg: 99.3414 +04/16 16:21:01 - mmengine - INFO - Iter(train) [ 29450/160000] base_lr: 8.2366e-05 lr: 3.0453e-07 eta: 1 day, 12:09:20 time: 0.9988 data_time: 0.0051 memory: 8462 loss: 0.0215 decode.loss_ce: 0.0117 decode.acc_seg: 99.4379 aux.loss_ce: 0.0098 aux.acc_seg: 98.8281 +04/16 16:21:51 - mmengine - INFO - Iter(train) [ 29500/160000] base_lr: 8.2335e-05 lr: 3.0441e-07 eta: 1 day, 12:08:31 time: 0.9987 data_time: 0.0046 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0100 decode.acc_seg: 99.5880 aux.loss_ce: 0.0078 aux.acc_seg: 99.0520 +04/16 16:22:41 - mmengine - INFO - Iter(train) [ 29550/160000] base_lr: 8.2303e-05 lr: 3.0429e-07 eta: 1 day, 12:07:41 time: 0.9985 data_time: 0.0042 memory: 8462 loss: 0.0215 decode.loss_ce: 0.0115 decode.acc_seg: 99.6632 aux.loss_ce: 0.0101 aux.acc_seg: 99.0210 +04/16 16:23:31 - mmengine - INFO - Iter(train) [ 29600/160000] base_lr: 8.2272e-05 lr: 3.0418e-07 eta: 1 day, 12:06:52 time: 0.9980 data_time: 0.0043 memory: 8462 loss: 0.0191 decode.loss_ce: 0.0099 decode.acc_seg: 99.5785 aux.loss_ce: 0.0092 aux.acc_seg: 99.1531 +04/16 16:24:21 - mmengine - INFO - Iter(train) [ 29650/160000] base_lr: 8.2240e-05 lr: 3.0406e-07 eta: 1 day, 12:06:02 time: 0.9999 data_time: 0.0044 memory: 8462 loss: 0.0191 decode.loss_ce: 0.0102 decode.acc_seg: 99.5041 aux.loss_ce: 0.0090 aux.acc_seg: 98.8997 +04/16 16:25:11 - mmengine - INFO - Iter(train) [ 29700/160000] base_lr: 8.2209e-05 lr: 3.0394e-07 eta: 1 day, 12:05:13 time: 0.9982 data_time: 0.0046 memory: 8462 loss: 0.0184 decode.loss_ce: 0.0099 decode.acc_seg: 99.7015 aux.loss_ce: 0.0085 aux.acc_seg: 99.2313 +04/16 16:26:01 - mmengine - INFO - Iter(train) [ 29750/160000] base_lr: 8.2177e-05 lr: 3.0383e-07 eta: 1 day, 12:04:23 time: 0.9992 data_time: 0.0046 memory: 8462 loss: 0.0187 decode.loss_ce: 0.0101 decode.acc_seg: 99.5207 aux.loss_ce: 0.0086 aux.acc_seg: 98.9408 +04/16 16:26:51 - mmengine - INFO - Iter(train) [ 29800/160000] base_lr: 8.2146e-05 lr: 3.0371e-07 eta: 1 day, 12:03:34 time: 0.9992 data_time: 0.0043 memory: 8462 loss: 0.0194 decode.loss_ce: 0.0101 decode.acc_seg: 99.5872 aux.loss_ce: 0.0093 aux.acc_seg: 98.7730 +04/16 16:27:41 - mmengine - INFO - Iter(train) [ 29850/160000] base_lr: 8.2114e-05 lr: 3.0359e-07 eta: 1 day, 12:02:45 time: 0.9988 data_time: 0.0045 memory: 8462 loss: 0.0213 decode.loss_ce: 0.0112 decode.acc_seg: 99.4427 aux.loss_ce: 0.0101 aux.acc_seg: 98.6961 +04/16 16:28:30 - mmengine - INFO - Iter(train) [ 29900/160000] base_lr: 8.2083e-05 lr: 3.0348e-07 eta: 1 day, 12:01:55 time: 0.9998 data_time: 0.0045 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0098 decode.acc_seg: 99.5665 aux.loss_ce: 0.0081 aux.acc_seg: 98.8626 +04/16 16:29:20 - mmengine - INFO - Iter(train) [ 29950/160000] base_lr: 8.2051e-05 lr: 3.0336e-07 eta: 1 day, 12:01:06 time: 0.9996 data_time: 0.0043 memory: 8462 loss: 0.0196 decode.loss_ce: 0.0107 decode.acc_seg: 99.7433 aux.loss_ce: 0.0089 aux.acc_seg: 99.2117 +04/16 16:30:10 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 16:30:10 - mmengine - INFO - Iter(train) [ 30000/160000] base_lr: 8.2019e-05 lr: 3.0324e-07 eta: 1 day, 12:00:17 time: 1.0001 data_time: 0.0047 memory: 8462 loss: 0.0210 decode.loss_ce: 0.0109 decode.acc_seg: 99.3954 aux.loss_ce: 0.0101 aux.acc_seg: 98.5298 +04/16 16:30:10 - mmengine - INFO - Saving checkpoint at 30000 iterations +04/16 16:30:20 - mmengine - INFO - Iter(val) [ 50/355] eta: 0:00:35 time: 0.1152 data_time: 0.0013 memory: 4004 +04/16 16:30:26 - mmengine - INFO - Iter(val) [100/355] eta: 0:00:29 time: 0.1154 data_time: 0.0016 memory: 4004 +04/16 16:30:32 - mmengine - INFO - Iter(val) [150/355] eta: 0:00:23 time: 0.1154 data_time: 0.0013 memory: 4004 +04/16 16:30:38 - mmengine - INFO - Iter(val) [200/355] eta: 0:00:17 time: 0.1155 data_time: 0.0015 memory: 4004 +04/16 16:30:43 - mmengine - INFO - Iter(val) [250/355] eta: 0:00:12 time: 0.1157 data_time: 0.0015 memory: 4004 +04/16 16:30:49 - mmengine - INFO - Iter(val) [300/355] eta: 0:00:06 time: 0.1157 data_time: 0.0014 memory: 4004 +04/16 16:30:55 - mmengine - INFO - Iter(val) [350/355] eta: 0:00:00 time: 0.1158 data_time: 0.0014 memory: 4004 +04/16 16:30:56 - mmengine - INFO - per class results: +04/16 16:30:56 - mmengine - INFO - ++------------+-------+-------+ +| Class | IoU | Acc | ++------------+-------+-------+ +| background | 99.2 | 99.61 | +| contrast | 82.37 | 89.96 | ++------------+-------+-------+ +04/16 16:30:56 - mmengine - INFO - Iter(val) [355/355] aAcc: 99.2200 mIoU: 90.7800 mAcc: 94.7900 data_time: 0.0016 time: 0.1158 +04/16 16:31:46 - mmengine - INFO - Iter(train) [ 30050/160000] base_lr: 8.1988e-05 lr: 3.0313e-07 eta: 1 day, 11:59:29 time: 0.9975 data_time: 0.0045 memory: 8462 loss: 0.0197 decode.loss_ce: 0.0104 decode.acc_seg: 99.6412 aux.loss_ce: 0.0094 aux.acc_seg: 98.9674 +04/16 16:32:36 - mmengine - INFO - Iter(train) [ 30100/160000] base_lr: 8.1956e-05 lr: 3.0301e-07 eta: 1 day, 11:58:39 time: 0.9978 data_time: 0.0043 memory: 8462 loss: 0.0206 decode.loss_ce: 0.0113 decode.acc_seg: 99.5163 aux.loss_ce: 0.0093 aux.acc_seg: 99.0168 +04/16 16:33:26 - mmengine - INFO - Iter(train) [ 30150/160000] base_lr: 8.1925e-05 lr: 3.0289e-07 eta: 1 day, 11:57:50 time: 0.9981 data_time: 0.0046 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0100 decode.acc_seg: 99.6508 aux.loss_ce: 0.0088 aux.acc_seg: 98.9639 +04/16 16:34:16 - mmengine - INFO - Iter(train) [ 30200/160000] base_lr: 8.1893e-05 lr: 3.0278e-07 eta: 1 day, 11:57:01 time: 0.9997 data_time: 0.0043 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0087 decode.acc_seg: 99.6592 aux.loss_ce: 0.0080 aux.acc_seg: 99.2422 +04/16 16:35:06 - mmengine - INFO - Iter(train) [ 30250/160000] base_lr: 8.1862e-05 lr: 3.0266e-07 eta: 1 day, 11:56:11 time: 0.9984 data_time: 0.0042 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0094 decode.acc_seg: 99.6014 aux.loss_ce: 0.0085 aux.acc_seg: 99.0728 +04/16 16:35:56 - mmengine - INFO - Iter(train) [ 30300/160000] base_lr: 8.1830e-05 lr: 3.0254e-07 eta: 1 day, 11:55:21 time: 0.9986 data_time: 0.0043 memory: 8462 loss: 0.0171 decode.loss_ce: 0.0087 decode.acc_seg: 99.6727 aux.loss_ce: 0.0084 aux.acc_seg: 99.1022 +04/16 16:36:46 - mmengine - INFO - Iter(train) [ 30350/160000] base_lr: 8.1799e-05 lr: 3.0243e-07 eta: 1 day, 11:54:32 time: 0.9983 data_time: 0.0043 memory: 8462 loss: 0.0198 decode.loss_ce: 0.0105 decode.acc_seg: 99.5253 aux.loss_ce: 0.0093 aux.acc_seg: 99.0820 +04/16 16:37:36 - mmengine - INFO - Iter(train) [ 30400/160000] base_lr: 8.1767e-05 lr: 3.0231e-07 eta: 1 day, 11:53:43 time: 0.9994 data_time: 0.0046 memory: 8462 loss: 0.0249 decode.loss_ce: 0.0138 decode.acc_seg: 99.5796 aux.loss_ce: 0.0111 aux.acc_seg: 98.9058 +04/16 16:38:25 - mmengine - INFO - Iter(train) [ 30450/160000] base_lr: 8.1736e-05 lr: 3.0219e-07 eta: 1 day, 11:52:53 time: 0.9985 data_time: 0.0044 memory: 8462 loss: 0.0192 decode.loss_ce: 0.0100 decode.acc_seg: 99.4627 aux.loss_ce: 0.0093 aux.acc_seg: 98.7928 +04/16 16:39:15 - mmengine - INFO - Iter(train) [ 30500/160000] base_lr: 8.1704e-05 lr: 3.0208e-07 eta: 1 day, 11:52:04 time: 0.9983 data_time: 0.0045 memory: 8462 loss: 0.0206 decode.loss_ce: 0.0118 decode.acc_seg: 99.2430 aux.loss_ce: 0.0088 aux.acc_seg: 98.7139 +04/16 16:40:05 - mmengine - INFO - Iter(train) [ 30550/160000] base_lr: 8.1672e-05 lr: 3.0196e-07 eta: 1 day, 11:51:14 time: 0.9990 data_time: 0.0044 memory: 8462 loss: 0.0182 decode.loss_ce: 0.0096 decode.acc_seg: 99.6105 aux.loss_ce: 0.0086 aux.acc_seg: 99.0822 +04/16 16:40:55 - mmengine - INFO - Iter(train) [ 30600/160000] base_lr: 8.1641e-05 lr: 3.0184e-07 eta: 1 day, 11:50:25 time: 0.9981 data_time: 0.0043 memory: 8462 loss: 0.0204 decode.loss_ce: 0.0106 decode.acc_seg: 99.5234 aux.loss_ce: 0.0098 aux.acc_seg: 98.7244 +04/16 16:41:45 - mmengine - INFO - Iter(train) [ 30650/160000] base_lr: 8.1609e-05 lr: 3.0173e-07 eta: 1 day, 11:49:35 time: 0.9984 data_time: 0.0043 memory: 8462 loss: 0.0183 decode.loss_ce: 0.0092 decode.acc_seg: 99.6655 aux.loss_ce: 0.0091 aux.acc_seg: 99.1932 +04/16 16:42:35 - mmengine - INFO - Iter(train) [ 30700/160000] base_lr: 8.1578e-05 lr: 3.0161e-07 eta: 1 day, 11:48:46 time: 0.9983 data_time: 0.0043 memory: 8462 loss: 0.0209 decode.loss_ce: 0.0114 decode.acc_seg: 99.4728 aux.loss_ce: 0.0095 aux.acc_seg: 98.6988 +04/16 16:43:25 - mmengine - INFO - Iter(train) [ 30750/160000] base_lr: 8.1546e-05 lr: 3.0149e-07 eta: 1 day, 11:47:56 time: 0.9980 data_time: 0.0044 memory: 8462 loss: 0.0197 decode.loss_ce: 0.0109 decode.acc_seg: 99.5340 aux.loss_ce: 0.0089 aux.acc_seg: 98.9395 +04/16 16:44:15 - mmengine - INFO - Iter(train) [ 30800/160000] base_lr: 8.1515e-05 lr: 3.0138e-07 eta: 1 day, 11:47:06 time: 0.9982 data_time: 0.0048 memory: 8462 loss: 0.0222 decode.loss_ce: 0.0122 decode.acc_seg: 99.4019 aux.loss_ce: 0.0100 aux.acc_seg: 98.8768 +04/16 16:45:05 - mmengine - INFO - Iter(train) [ 30850/160000] base_lr: 8.1483e-05 lr: 3.0126e-07 eta: 1 day, 11:46:17 time: 0.9981 data_time: 0.0043 memory: 8462 loss: 0.0224 decode.loss_ce: 0.0121 decode.acc_seg: 99.5239 aux.loss_ce: 0.0103 aux.acc_seg: 98.8468 +04/16 16:45:55 - mmengine - INFO - Iter(train) [ 30900/160000] base_lr: 8.1452e-05 lr: 3.0114e-07 eta: 1 day, 11:45:27 time: 0.9981 data_time: 0.0044 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0097 decode.acc_seg: 99.6664 aux.loss_ce: 0.0081 aux.acc_seg: 99.2163 +04/16 16:46:45 - mmengine - INFO - Iter(train) [ 30950/160000] base_lr: 8.1420e-05 lr: 3.0103e-07 eta: 1 day, 11:44:38 time: 1.0002 data_time: 0.0045 memory: 8462 loss: 0.0226 decode.loss_ce: 0.0129 decode.acc_seg: 99.3189 aux.loss_ce: 0.0097 aux.acc_seg: 98.7026 +04/16 16:47:35 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 16:47:35 - mmengine - INFO - Iter(train) [ 31000/160000] base_lr: 8.1389e-05 lr: 3.0091e-07 eta: 1 day, 11:43:48 time: 0.9981 data_time: 0.0046 memory: 8462 loss: 0.0234 decode.loss_ce: 0.0122 decode.acc_seg: 99.5394 aux.loss_ce: 0.0112 aux.acc_seg: 98.7713 +04/16 16:48:25 - mmengine - INFO - Iter(train) [ 31050/160000] base_lr: 8.1357e-05 lr: 3.0079e-07 eta: 1 day, 11:42:59 time: 0.9982 data_time: 0.0048 memory: 8462 loss: 0.0222 decode.loss_ce: 0.0121 decode.acc_seg: 99.6630 aux.loss_ce: 0.0101 aux.acc_seg: 99.1295 +04/16 16:49:15 - mmengine - INFO - Iter(train) [ 31100/160000] base_lr: 8.1325e-05 lr: 3.0068e-07 eta: 1 day, 11:42:09 time: 0.9985 data_time: 0.0043 memory: 8462 loss: 0.0230 decode.loss_ce: 0.0118 decode.acc_seg: 99.3177 aux.loss_ce: 0.0112 aux.acc_seg: 98.3019 +04/16 16:50:05 - mmengine - INFO - Iter(train) [ 31150/160000] base_lr: 8.1294e-05 lr: 3.0056e-07 eta: 1 day, 11:41:20 time: 0.9979 data_time: 0.0043 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0091 decode.acc_seg: 99.6361 aux.loss_ce: 0.0078 aux.acc_seg: 99.0808 +04/16 16:50:55 - mmengine - INFO - Iter(train) [ 31200/160000] base_lr: 8.1262e-05 lr: 3.0044e-07 eta: 1 day, 11:40:30 time: 0.9997 data_time: 0.0042 memory: 8462 loss: 0.0207 decode.loss_ce: 0.0113 decode.acc_seg: 99.6029 aux.loss_ce: 0.0094 aux.acc_seg: 99.0639 +04/16 16:51:44 - mmengine - INFO - Iter(train) [ 31250/160000] base_lr: 8.1231e-05 lr: 3.0033e-07 eta: 1 day, 11:39:41 time: 0.9992 data_time: 0.0044 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0102 decode.acc_seg: 99.6359 aux.loss_ce: 0.0086 aux.acc_seg: 99.1867 +04/16 16:52:34 - mmengine - INFO - Iter(train) [ 31300/160000] base_lr: 8.1199e-05 lr: 3.0021e-07 eta: 1 day, 11:38:51 time: 0.9991 data_time: 0.0047 memory: 8462 loss: 0.0218 decode.loss_ce: 0.0116 decode.acc_seg: 99.6639 aux.loss_ce: 0.0101 aux.acc_seg: 99.0749 +04/16 16:53:24 - mmengine - INFO - Iter(train) [ 31350/160000] base_lr: 8.1168e-05 lr: 3.0009e-07 eta: 1 day, 11:38:02 time: 0.9986 data_time: 0.0045 memory: 8462 loss: 0.0200 decode.loss_ce: 0.0105 decode.acc_seg: 99.4705 aux.loss_ce: 0.0095 aux.acc_seg: 98.8827 +04/16 16:54:14 - mmengine - INFO - Iter(train) [ 31400/160000] base_lr: 8.1136e-05 lr: 2.9998e-07 eta: 1 day, 11:37:12 time: 0.9984 data_time: 0.0046 memory: 8462 loss: 0.0192 decode.loss_ce: 0.0098 decode.acc_seg: 99.3767 aux.loss_ce: 0.0094 aux.acc_seg: 98.6240 +04/16 16:55:04 - mmengine - INFO - Iter(train) [ 31450/160000] base_lr: 8.1105e-05 lr: 2.9986e-07 eta: 1 day, 11:36:23 time: 0.9981 data_time: 0.0041 memory: 8462 loss: 0.0206 decode.loss_ce: 0.0107 decode.acc_seg: 99.5136 aux.loss_ce: 0.0099 aux.acc_seg: 98.8297 +04/16 16:55:54 - mmengine - INFO - Iter(train) [ 31500/160000] base_lr: 8.1073e-05 lr: 2.9974e-07 eta: 1 day, 11:35:33 time: 0.9987 data_time: 0.0046 memory: 8462 loss: 0.0206 decode.loss_ce: 0.0111 decode.acc_seg: 99.4501 aux.loss_ce: 0.0095 aux.acc_seg: 98.7457 +04/16 16:56:44 - mmengine - INFO - Iter(train) [ 31550/160000] base_lr: 8.1042e-05 lr: 2.9963e-07 eta: 1 day, 11:34:44 time: 0.9987 data_time: 0.0046 memory: 8462 loss: 0.0237 decode.loss_ce: 0.0129 decode.acc_seg: 99.0498 aux.loss_ce: 0.0108 aux.acc_seg: 98.4348 +04/16 16:57:34 - mmengine - INFO - Iter(train) [ 31600/160000] base_lr: 8.1010e-05 lr: 2.9951e-07 eta: 1 day, 11:33:54 time: 0.9999 data_time: 0.0044 memory: 8462 loss: 0.0185 decode.loss_ce: 0.0095 decode.acc_seg: 99.6456 aux.loss_ce: 0.0090 aux.acc_seg: 99.1667 +04/16 16:58:24 - mmengine - INFO - Iter(train) [ 31650/160000] base_lr: 8.0978e-05 lr: 2.9939e-07 eta: 1 day, 11:33:05 time: 0.9995 data_time: 0.0042 memory: 8462 loss: 0.0258 decode.loss_ce: 0.0147 decode.acc_seg: 99.6164 aux.loss_ce: 0.0111 aux.acc_seg: 98.7766 +04/16 16:59:14 - mmengine - INFO - Iter(train) [ 31700/160000] base_lr: 8.0947e-05 lr: 2.9928e-07 eta: 1 day, 11:32:15 time: 0.9980 data_time: 0.0043 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0090 decode.acc_seg: 99.6387 aux.loss_ce: 0.0088 aux.acc_seg: 99.3845 +04/16 17:00:04 - mmengine - INFO - Iter(train) [ 31750/160000] base_lr: 8.0915e-05 lr: 2.9916e-07 eta: 1 day, 11:31:25 time: 0.9979 data_time: 0.0043 memory: 8462 loss: 0.0192 decode.loss_ce: 0.0101 decode.acc_seg: 99.6805 aux.loss_ce: 0.0091 aux.acc_seg: 99.3076 +04/16 17:00:54 - mmengine - INFO - Iter(train) [ 31800/160000] base_lr: 8.0884e-05 lr: 2.9904e-07 eta: 1 day, 11:30:36 time: 0.9976 data_time: 0.0051 memory: 8462 loss: 0.0220 decode.loss_ce: 0.0118 decode.acc_seg: 99.4101 aux.loss_ce: 0.0103 aux.acc_seg: 98.7715 +04/16 17:01:44 - mmengine - INFO - Iter(train) [ 31850/160000] base_lr: 8.0852e-05 lr: 2.9893e-07 eta: 1 day, 11:29:46 time: 0.9999 data_time: 0.0042 memory: 8462 loss: 0.0215 decode.loss_ce: 0.0116 decode.acc_seg: 99.4753 aux.loss_ce: 0.0099 aux.acc_seg: 98.6017 +04/16 17:02:34 - mmengine - INFO - Iter(train) [ 31900/160000] base_lr: 8.0821e-05 lr: 2.9881e-07 eta: 1 day, 11:28:57 time: 1.0000 data_time: 0.0044 memory: 8462 loss: 0.0203 decode.loss_ce: 0.0111 decode.acc_seg: 99.3425 aux.loss_ce: 0.0093 aux.acc_seg: 98.9721 +04/16 17:03:24 - mmengine - INFO - Iter(train) [ 31950/160000] base_lr: 8.0789e-05 lr: 2.9869e-07 eta: 1 day, 11:28:07 time: 0.9999 data_time: 0.0050 memory: 8462 loss: 0.0184 decode.loss_ce: 0.0099 decode.acc_seg: 99.6935 aux.loss_ce: 0.0085 aux.acc_seg: 99.3494 +04/16 17:04:14 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 17:04:14 - mmengine - INFO - Iter(train) [ 32000/160000] base_lr: 8.0758e-05 lr: 2.9858e-07 eta: 1 day, 11:27:18 time: 0.9979 data_time: 0.0043 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0099 decode.acc_seg: 99.3664 aux.loss_ce: 0.0090 aux.acc_seg: 98.5855 +04/16 17:05:04 - mmengine - INFO - Iter(train) [ 32050/160000] base_lr: 8.0726e-05 lr: 2.9846e-07 eta: 1 day, 11:26:28 time: 0.9994 data_time: 0.0044 memory: 8462 loss: 0.0208 decode.loss_ce: 0.0114 decode.acc_seg: 99.4728 aux.loss_ce: 0.0095 aux.acc_seg: 98.7848 +04/16 17:05:53 - mmengine - INFO - Iter(train) [ 32100/160000] base_lr: 8.0695e-05 lr: 2.9834e-07 eta: 1 day, 11:25:39 time: 0.9981 data_time: 0.0042 memory: 8462 loss: 0.0221 decode.loss_ce: 0.0118 decode.acc_seg: 99.6059 aux.loss_ce: 0.0103 aux.acc_seg: 99.2264 +04/16 17:06:43 - mmengine - INFO - Iter(train) [ 32150/160000] base_lr: 8.0663e-05 lr: 2.9823e-07 eta: 1 day, 11:24:49 time: 0.9984 data_time: 0.0042 memory: 8462 loss: 0.0225 decode.loss_ce: 0.0116 decode.acc_seg: 99.4173 aux.loss_ce: 0.0109 aux.acc_seg: 98.8316 +04/16 17:07:33 - mmengine - INFO - Iter(train) [ 32200/160000] base_lr: 8.0631e-05 lr: 2.9811e-07 eta: 1 day, 11:24:00 time: 0.9995 data_time: 0.0042 memory: 8462 loss: 0.0199 decode.loss_ce: 0.0104 decode.acc_seg: 99.5943 aux.loss_ce: 0.0095 aux.acc_seg: 99.0021 +04/16 17:08:23 - mmengine - INFO - Iter(train) [ 32250/160000] base_lr: 8.0600e-05 lr: 2.9799e-07 eta: 1 day, 11:23:10 time: 0.9984 data_time: 0.0047 memory: 8462 loss: 0.0210 decode.loss_ce: 0.0108 decode.acc_seg: 99.5726 aux.loss_ce: 0.0102 aux.acc_seg: 99.0511 +04/16 17:09:13 - mmengine - INFO - Iter(train) [ 32300/160000] base_lr: 8.0568e-05 lr: 2.9788e-07 eta: 1 day, 11:22:20 time: 0.9987 data_time: 0.0043 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0104 decode.acc_seg: 99.5050 aux.loss_ce: 0.0085 aux.acc_seg: 99.0969 +04/16 17:10:03 - mmengine - INFO - Iter(train) [ 32350/160000] base_lr: 8.0537e-05 lr: 2.9776e-07 eta: 1 day, 11:21:31 time: 0.9988 data_time: 0.0045 memory: 8462 loss: 0.0201 decode.loss_ce: 0.0103 decode.acc_seg: 99.6323 aux.loss_ce: 0.0098 aux.acc_seg: 98.8977 +04/16 17:10:53 - mmengine - INFO - Iter(train) [ 32400/160000] base_lr: 8.0505e-05 lr: 2.9764e-07 eta: 1 day, 11:20:41 time: 0.9981 data_time: 0.0043 memory: 8462 loss: 0.0215 decode.loss_ce: 0.0111 decode.acc_seg: 99.4240 aux.loss_ce: 0.0104 aux.acc_seg: 98.7015 +04/16 17:11:43 - mmengine - INFO - Iter(train) [ 32450/160000] base_lr: 8.0474e-05 lr: 2.9753e-07 eta: 1 day, 11:19:52 time: 0.9994 data_time: 0.0047 memory: 8462 loss: 0.0209 decode.loss_ce: 0.0111 decode.acc_seg: 99.4419 aux.loss_ce: 0.0097 aux.acc_seg: 98.7825 +04/16 17:12:33 - mmengine - INFO - Iter(train) [ 32500/160000] base_lr: 8.0442e-05 lr: 2.9741e-07 eta: 1 day, 11:19:02 time: 0.9996 data_time: 0.0049 memory: 8462 loss: 0.0194 decode.loss_ce: 0.0100 decode.acc_seg: 99.7042 aux.loss_ce: 0.0094 aux.acc_seg: 99.1493 +04/16 17:13:23 - mmengine - INFO - Iter(train) [ 32550/160000] base_lr: 8.0411e-05 lr: 2.9729e-07 eta: 1 day, 11:18:13 time: 0.9991 data_time: 0.0046 memory: 8462 loss: 0.0193 decode.loss_ce: 0.0103 decode.acc_seg: 99.6092 aux.loss_ce: 0.0090 aux.acc_seg: 98.9584 +04/16 17:14:13 - mmengine - INFO - Iter(train) [ 32600/160000] base_lr: 8.0379e-05 lr: 2.9718e-07 eta: 1 day, 11:17:23 time: 0.9984 data_time: 0.0047 memory: 8462 loss: 0.0211 decode.loss_ce: 0.0118 decode.acc_seg: 99.5871 aux.loss_ce: 0.0093 aux.acc_seg: 99.2870 +04/16 17:15:03 - mmengine - INFO - Iter(train) [ 32650/160000] base_lr: 8.0348e-05 lr: 2.9706e-07 eta: 1 day, 11:16:34 time: 0.9991 data_time: 0.0043 memory: 8462 loss: 0.0218 decode.loss_ce: 0.0113 decode.acc_seg: 99.6393 aux.loss_ce: 0.0105 aux.acc_seg: 98.9820 +04/16 17:15:53 - mmengine - INFO - Iter(train) [ 32700/160000] base_lr: 8.0316e-05 lr: 2.9694e-07 eta: 1 day, 11:15:44 time: 0.9999 data_time: 0.0046 memory: 8462 loss: 0.0196 decode.loss_ce: 0.0102 decode.acc_seg: 99.5453 aux.loss_ce: 0.0093 aux.acc_seg: 99.1310 +04/16 17:16:43 - mmengine - INFO - Iter(train) [ 32750/160000] base_lr: 8.0284e-05 lr: 2.9683e-07 eta: 1 day, 11:14:55 time: 0.9980 data_time: 0.0050 memory: 8462 loss: 0.0212 decode.loss_ce: 0.0108 decode.acc_seg: 99.5314 aux.loss_ce: 0.0104 aux.acc_seg: 98.5731 +04/16 17:17:33 - mmengine - INFO - Iter(train) [ 32800/160000] base_lr: 8.0253e-05 lr: 2.9671e-07 eta: 1 day, 11:14:05 time: 0.9977 data_time: 0.0046 memory: 8462 loss: 0.0187 decode.loss_ce: 0.0092 decode.acc_seg: 99.7011 aux.loss_ce: 0.0095 aux.acc_seg: 99.1590 +04/16 17:18:23 - mmengine - INFO - Iter(train) [ 32850/160000] base_lr: 8.0221e-05 lr: 2.9659e-07 eta: 1 day, 11:13:16 time: 0.9985 data_time: 0.0044 memory: 8462 loss: 0.0211 decode.loss_ce: 0.0116 decode.acc_seg: 99.6649 aux.loss_ce: 0.0095 aux.acc_seg: 99.2958 +04/16 17:19:13 - mmengine - INFO - Iter(train) [ 32900/160000] base_lr: 8.0190e-05 lr: 2.9648e-07 eta: 1 day, 11:12:26 time: 0.9994 data_time: 0.0044 memory: 8462 loss: 0.0191 decode.loss_ce: 0.0101 decode.acc_seg: 99.5682 aux.loss_ce: 0.0090 aux.acc_seg: 98.9614 +04/16 17:20:03 - mmengine - INFO - Iter(train) [ 32950/160000] base_lr: 8.0158e-05 lr: 2.9636e-07 eta: 1 day, 11:11:37 time: 0.9982 data_time: 0.0045 memory: 8462 loss: 0.0216 decode.loss_ce: 0.0110 decode.acc_seg: 99.2937 aux.loss_ce: 0.0106 aux.acc_seg: 98.3685 +04/16 17:20:53 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 17:20:53 - mmengine - INFO - Iter(train) [ 33000/160000] base_lr: 8.0127e-05 lr: 2.9624e-07 eta: 1 day, 11:10:47 time: 0.9979 data_time: 0.0043 memory: 8462 loss: 0.0197 decode.loss_ce: 0.0107 decode.acc_seg: 99.2453 aux.loss_ce: 0.0090 aux.acc_seg: 98.4140 +04/16 17:21:42 - mmengine - INFO - Iter(train) [ 33050/160000] base_lr: 8.0095e-05 lr: 2.9613e-07 eta: 1 day, 11:09:57 time: 0.9990 data_time: 0.0045 memory: 8462 loss: 0.0228 decode.loss_ce: 0.0124 decode.acc_seg: 99.5264 aux.loss_ce: 0.0104 aux.acc_seg: 99.0314 +04/16 17:22:32 - mmengine - INFO - Iter(train) [ 33100/160000] base_lr: 8.0064e-05 lr: 2.9601e-07 eta: 1 day, 11:09:08 time: 0.9981 data_time: 0.0047 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0097 decode.acc_seg: 99.4806 aux.loss_ce: 0.0090 aux.acc_seg: 98.8743 +04/16 17:23:22 - mmengine - INFO - Iter(train) [ 33150/160000] base_lr: 8.0032e-05 lr: 2.9589e-07 eta: 1 day, 11:08:18 time: 0.9993 data_time: 0.0046 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0091 decode.acc_seg: 99.6393 aux.loss_ce: 0.0085 aux.acc_seg: 99.2161 +04/16 17:24:12 - mmengine - INFO - Iter(train) [ 33200/160000] base_lr: 8.0001e-05 lr: 2.9578e-07 eta: 1 day, 11:07:29 time: 0.9987 data_time: 0.0045 memory: 8462 loss: 0.0184 decode.loss_ce: 0.0094 decode.acc_seg: 99.3786 aux.loss_ce: 0.0090 aux.acc_seg: 98.6816 +04/16 17:25:02 - mmengine - INFO - Iter(train) [ 33250/160000] base_lr: 7.9969e-05 lr: 2.9566e-07 eta: 1 day, 11:06:40 time: 0.9995 data_time: 0.0048 memory: 8462 loss: 0.0196 decode.loss_ce: 0.0095 decode.acc_seg: 99.5754 aux.loss_ce: 0.0101 aux.acc_seg: 99.0080 +04/16 17:25:52 - mmengine - INFO - Iter(train) [ 33300/160000] base_lr: 7.9937e-05 lr: 2.9555e-07 eta: 1 day, 11:05:50 time: 0.9994 data_time: 0.0048 memory: 8462 loss: 0.0191 decode.loss_ce: 0.0097 decode.acc_seg: 99.7211 aux.loss_ce: 0.0094 aux.acc_seg: 99.1596 +04/16 17:26:42 - mmengine - INFO - Iter(train) [ 33350/160000] base_lr: 7.9906e-05 lr: 2.9543e-07 eta: 1 day, 11:05:01 time: 0.9993 data_time: 0.0047 memory: 8462 loss: 0.0218 decode.loss_ce: 0.0113 decode.acc_seg: 99.5609 aux.loss_ce: 0.0105 aux.acc_seg: 98.9893 +04/16 17:27:32 - mmengine - INFO - Iter(train) [ 33400/160000] base_lr: 7.9874e-05 lr: 2.9531e-07 eta: 1 day, 11:04:11 time: 0.9983 data_time: 0.0044 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0084 decode.acc_seg: 99.8013 aux.loss_ce: 0.0083 aux.acc_seg: 99.4621 +04/16 17:28:22 - mmengine - INFO - Iter(train) [ 33450/160000] base_lr: 7.9843e-05 lr: 2.9520e-07 eta: 1 day, 11:03:21 time: 0.9999 data_time: 0.0047 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0098 decode.acc_seg: 99.7274 aux.loss_ce: 0.0091 aux.acc_seg: 99.3542 +04/16 17:29:12 - mmengine - INFO - Iter(train) [ 33500/160000] base_lr: 7.9811e-05 lr: 2.9508e-07 eta: 1 day, 11:02:32 time: 0.9996 data_time: 0.0049 memory: 8462 loss: 0.0234 decode.loss_ce: 0.0129 decode.acc_seg: 99.5190 aux.loss_ce: 0.0105 aux.acc_seg: 99.1327 +04/16 17:30:02 - mmengine - INFO - Iter(train) [ 33550/160000] base_lr: 7.9780e-05 lr: 2.9496e-07 eta: 1 day, 11:01:42 time: 0.9995 data_time: 0.0044 memory: 8462 loss: 0.0192 decode.loss_ce: 0.0100 decode.acc_seg: 99.5644 aux.loss_ce: 0.0092 aux.acc_seg: 98.8697 +04/16 17:30:52 - mmengine - INFO - Iter(train) [ 33600/160000] base_lr: 7.9748e-05 lr: 2.9485e-07 eta: 1 day, 11:00:53 time: 0.9981 data_time: 0.0047 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0101 decode.acc_seg: 99.7540 aux.loss_ce: 0.0089 aux.acc_seg: 99.4677 +04/16 17:31:42 - mmengine - INFO - Iter(train) [ 33650/160000] base_lr: 7.9717e-05 lr: 2.9473e-07 eta: 1 day, 11:00:03 time: 0.9982 data_time: 0.0046 memory: 8462 loss: 0.0186 decode.loss_ce: 0.0099 decode.acc_seg: 99.6511 aux.loss_ce: 0.0087 aux.acc_seg: 99.1922 +04/16 17:32:32 - mmengine - INFO - Iter(train) [ 33700/160000] base_lr: 7.9685e-05 lr: 2.9461e-07 eta: 1 day, 10:59:14 time: 1.0001 data_time: 0.0046 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0087 decode.acc_seg: 99.5108 aux.loss_ce: 0.0084 aux.acc_seg: 98.9992 +04/16 17:33:22 - mmengine - INFO - Iter(train) [ 33750/160000] base_lr: 7.9653e-05 lr: 2.9450e-07 eta: 1 day, 10:58:24 time: 0.9994 data_time: 0.0050 memory: 8462 loss: 0.0216 decode.loss_ce: 0.0115 decode.acc_seg: 99.4774 aux.loss_ce: 0.0101 aux.acc_seg: 99.0191 +04/16 17:34:12 - mmengine - INFO - Iter(train) [ 33800/160000] base_lr: 7.9622e-05 lr: 2.9438e-07 eta: 1 day, 10:57:35 time: 0.9994 data_time: 0.0048 memory: 8462 loss: 0.0214 decode.loss_ce: 0.0114 decode.acc_seg: 99.5604 aux.loss_ce: 0.0100 aux.acc_seg: 98.9998 +04/16 17:35:02 - mmengine - INFO - Iter(train) [ 33850/160000] base_lr: 7.9590e-05 lr: 2.9426e-07 eta: 1 day, 10:56:45 time: 0.9989 data_time: 0.0046 memory: 8462 loss: 0.0231 decode.loss_ce: 0.0126 decode.acc_seg: 99.5131 aux.loss_ce: 0.0105 aux.acc_seg: 99.0900 +04/16 17:35:52 - mmengine - INFO - Iter(train) [ 33900/160000] base_lr: 7.9559e-05 lr: 2.9415e-07 eta: 1 day, 10:55:56 time: 0.9991 data_time: 0.0044 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0096 decode.acc_seg: 99.6851 aux.loss_ce: 0.0083 aux.acc_seg: 99.1272 +04/16 17:36:42 - mmengine - INFO - Iter(train) [ 33950/160000] base_lr: 7.9527e-05 lr: 2.9403e-07 eta: 1 day, 10:55:06 time: 0.9993 data_time: 0.0043 memory: 8462 loss: 0.0215 decode.loss_ce: 0.0114 decode.acc_seg: 99.4322 aux.loss_ce: 0.0101 aux.acc_seg: 99.0152 +04/16 17:37:32 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 17:37:32 - mmengine - INFO - Iter(train) [ 34000/160000] base_lr: 7.9496e-05 lr: 2.9391e-07 eta: 1 day, 10:54:17 time: 0.9988 data_time: 0.0049 memory: 8462 loss: 0.0194 decode.loss_ce: 0.0102 decode.acc_seg: 99.2540 aux.loss_ce: 0.0092 aux.acc_seg: 98.8462 +04/16 17:38:22 - mmengine - INFO - Iter(train) [ 34050/160000] base_lr: 7.9464e-05 lr: 2.9380e-07 eta: 1 day, 10:53:27 time: 0.9985 data_time: 0.0044 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0091 decode.acc_seg: 99.6376 aux.loss_ce: 0.0088 aux.acc_seg: 99.0265 +04/16 17:39:12 - mmengine - INFO - Iter(train) [ 34100/160000] base_lr: 7.9433e-05 lr: 2.9368e-07 eta: 1 day, 10:52:38 time: 0.9986 data_time: 0.0045 memory: 8462 loss: 0.0171 decode.loss_ce: 0.0088 decode.acc_seg: 99.5995 aux.loss_ce: 0.0084 aux.acc_seg: 98.9571 +04/16 17:40:01 - mmengine - INFO - Iter(train) [ 34150/160000] base_lr: 7.9401e-05 lr: 2.9356e-07 eta: 1 day, 10:51:48 time: 0.9989 data_time: 0.0045 memory: 8462 loss: 0.0194 decode.loss_ce: 0.0101 decode.acc_seg: 99.6189 aux.loss_ce: 0.0092 aux.acc_seg: 99.4095 +04/16 17:40:51 - mmengine - INFO - Iter(train) [ 34200/160000] base_lr: 7.9370e-05 lr: 2.9345e-07 eta: 1 day, 10:50:58 time: 0.9987 data_time: 0.0044 memory: 8462 loss: 0.0184 decode.loss_ce: 0.0095 decode.acc_seg: 99.7444 aux.loss_ce: 0.0088 aux.acc_seg: 99.5531 +04/16 17:41:41 - mmengine - INFO - Iter(train) [ 34250/160000] base_lr: 7.9338e-05 lr: 2.9333e-07 eta: 1 day, 10:50:09 time: 0.9990 data_time: 0.0045 memory: 8462 loss: 0.0203 decode.loss_ce: 0.0104 decode.acc_seg: 99.7259 aux.loss_ce: 0.0099 aux.acc_seg: 99.3155 +04/16 17:42:31 - mmengine - INFO - Iter(train) [ 34300/160000] base_lr: 7.9306e-05 lr: 2.9321e-07 eta: 1 day, 10:49:19 time: 0.9994 data_time: 0.0048 memory: 8462 loss: 0.0184 decode.loss_ce: 0.0096 decode.acc_seg: 99.6750 aux.loss_ce: 0.0089 aux.acc_seg: 99.1764 +04/16 17:43:21 - mmengine - INFO - Iter(train) [ 34350/160000] base_lr: 7.9275e-05 lr: 2.9310e-07 eta: 1 day, 10:48:30 time: 0.9992 data_time: 0.0043 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0094 decode.acc_seg: 99.5104 aux.loss_ce: 0.0087 aux.acc_seg: 98.8253 +04/16 17:44:11 - mmengine - INFO - Iter(train) [ 34400/160000] base_lr: 7.9243e-05 lr: 2.9298e-07 eta: 1 day, 10:47:40 time: 0.9996 data_time: 0.0046 memory: 8462 loss: 0.0194 decode.loss_ce: 0.0098 decode.acc_seg: 99.6490 aux.loss_ce: 0.0096 aux.acc_seg: 99.0376 +04/16 17:45:01 - mmengine - INFO - Iter(train) [ 34450/160000] base_lr: 7.9212e-05 lr: 2.9286e-07 eta: 1 day, 10:46:51 time: 0.9994 data_time: 0.0044 memory: 8462 loss: 0.0246 decode.loss_ce: 0.0135 decode.acc_seg: 99.4993 aux.loss_ce: 0.0111 aux.acc_seg: 99.0532 +04/16 17:45:51 - mmengine - INFO - Iter(train) [ 34500/160000] base_lr: 7.9180e-05 lr: 2.9275e-07 eta: 1 day, 10:46:01 time: 0.9988 data_time: 0.0043 memory: 8462 loss: 0.0206 decode.loss_ce: 0.0106 decode.acc_seg: 99.4734 aux.loss_ce: 0.0100 aux.acc_seg: 98.9143 +04/16 17:46:41 - mmengine - INFO - Iter(train) [ 34550/160000] base_lr: 7.9149e-05 lr: 2.9263e-07 eta: 1 day, 10:45:12 time: 0.9996 data_time: 0.0044 memory: 8462 loss: 0.0199 decode.loss_ce: 0.0104 decode.acc_seg: 99.6119 aux.loss_ce: 0.0095 aux.acc_seg: 99.0353 +04/16 17:47:31 - mmengine - INFO - Iter(train) [ 34600/160000] base_lr: 7.9117e-05 lr: 2.9251e-07 eta: 1 day, 10:44:22 time: 0.9991 data_time: 0.0041 memory: 8462 loss: 0.0194 decode.loss_ce: 0.0103 decode.acc_seg: 99.6735 aux.loss_ce: 0.0091 aux.acc_seg: 99.1041 +04/16 17:48:21 - mmengine - INFO - Iter(train) [ 34650/160000] base_lr: 7.9086e-05 lr: 2.9240e-07 eta: 1 day, 10:43:32 time: 0.9985 data_time: 0.0043 memory: 8462 loss: 0.0182 decode.loss_ce: 0.0092 decode.acc_seg: 99.4553 aux.loss_ce: 0.0090 aux.acc_seg: 98.6092 +04/16 17:49:11 - mmengine - INFO - Iter(train) [ 34700/160000] base_lr: 7.9054e-05 lr: 2.9228e-07 eta: 1 day, 10:42:43 time: 0.9992 data_time: 0.0045 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0088 decode.acc_seg: 99.6115 aux.loss_ce: 0.0087 aux.acc_seg: 99.1495 +04/16 17:50:01 - mmengine - INFO - Iter(train) [ 34750/160000] base_lr: 7.9023e-05 lr: 2.9216e-07 eta: 1 day, 10:41:53 time: 0.9990 data_time: 0.0046 memory: 8462 loss: 0.0229 decode.loss_ce: 0.0131 decode.acc_seg: 99.5756 aux.loss_ce: 0.0098 aux.acc_seg: 99.0211 +04/16 17:50:51 - mmengine - INFO - Iter(train) [ 34800/160000] base_lr: 7.8991e-05 lr: 2.9205e-07 eta: 1 day, 10:41:04 time: 0.9985 data_time: 0.0049 memory: 8462 loss: 0.0218 decode.loss_ce: 0.0113 decode.acc_seg: 99.5152 aux.loss_ce: 0.0105 aux.acc_seg: 98.7673 +04/16 17:51:41 - mmengine - INFO - Iter(train) [ 34850/160000] base_lr: 7.8959e-05 lr: 2.9193e-07 eta: 1 day, 10:40:14 time: 0.9986 data_time: 0.0047 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0090 decode.acc_seg: 99.3679 aux.loss_ce: 0.0084 aux.acc_seg: 98.8329 +04/16 17:52:31 - mmengine - INFO - Iter(train) [ 34900/160000] base_lr: 7.8928e-05 lr: 2.9181e-07 eta: 1 day, 10:39:25 time: 0.9992 data_time: 0.0048 memory: 8462 loss: 0.0191 decode.loss_ce: 0.0099 decode.acc_seg: 99.6708 aux.loss_ce: 0.0092 aux.acc_seg: 99.1367 +04/16 17:53:21 - mmengine - INFO - Iter(train) [ 34950/160000] base_lr: 7.8896e-05 lr: 2.9170e-07 eta: 1 day, 10:38:35 time: 0.9986 data_time: 0.0044 memory: 8462 loss: 0.0219 decode.loss_ce: 0.0114 decode.acc_seg: 99.3448 aux.loss_ce: 0.0105 aux.acc_seg: 98.7076 +04/16 17:54:11 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 17:54:11 - mmengine - INFO - Iter(train) [ 35000/160000] base_lr: 7.8865e-05 lr: 2.9158e-07 eta: 1 day, 10:37:46 time: 0.9994 data_time: 0.0045 memory: 8462 loss: 0.0196 decode.loss_ce: 0.0101 decode.acc_seg: 99.6401 aux.loss_ce: 0.0095 aux.acc_seg: 99.0271 +04/16 17:55:01 - mmengine - INFO - Iter(train) [ 35050/160000] base_lr: 7.8833e-05 lr: 2.9146e-07 eta: 1 day, 10:36:56 time: 1.0000 data_time: 0.0047 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0093 decode.acc_seg: 99.6319 aux.loss_ce: 0.0085 aux.acc_seg: 99.1037 +04/16 17:55:51 - mmengine - INFO - Iter(train) [ 35100/160000] base_lr: 7.8802e-05 lr: 2.9135e-07 eta: 1 day, 10:36:07 time: 0.9993 data_time: 0.0044 memory: 8462 loss: 0.0224 decode.loss_ce: 0.0118 decode.acc_seg: 99.6220 aux.loss_ce: 0.0105 aux.acc_seg: 99.0681 +04/16 17:56:41 - mmengine - INFO - Iter(train) [ 35150/160000] base_lr: 7.8770e-05 lr: 2.9123e-07 eta: 1 day, 10:35:17 time: 1.0000 data_time: 0.0044 memory: 8462 loss: 0.0218 decode.loss_ce: 0.0114 decode.acc_seg: 99.3393 aux.loss_ce: 0.0105 aux.acc_seg: 98.2107 +04/16 17:57:31 - mmengine - INFO - Iter(train) [ 35200/160000] base_lr: 7.8739e-05 lr: 2.9111e-07 eta: 1 day, 10:34:28 time: 0.9992 data_time: 0.0045 memory: 8462 loss: 0.0204 decode.loss_ce: 0.0105 decode.acc_seg: 99.5390 aux.loss_ce: 0.0099 aux.acc_seg: 98.7232 +04/16 17:58:21 - mmengine - INFO - Iter(train) [ 35250/160000] base_lr: 7.8707e-05 lr: 2.9100e-07 eta: 1 day, 10:33:38 time: 0.9985 data_time: 0.0052 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0085 decode.acc_seg: 99.6075 aux.loss_ce: 0.0085 aux.acc_seg: 99.2310 +04/16 17:59:11 - mmengine - INFO - Iter(train) [ 35300/160000] base_lr: 7.8676e-05 lr: 2.9088e-07 eta: 1 day, 10:32:49 time: 0.9988 data_time: 0.0047 memory: 8462 loss: 0.0189 decode.loss_ce: 0.0102 decode.acc_seg: 99.6502 aux.loss_ce: 0.0088 aux.acc_seg: 99.2588 +04/16 18:00:01 - mmengine - INFO - Iter(train) [ 35350/160000] base_lr: 7.8644e-05 lr: 2.9076e-07 eta: 1 day, 10:31:59 time: 0.9991 data_time: 0.0042 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0089 decode.acc_seg: 99.6603 aux.loss_ce: 0.0083 aux.acc_seg: 99.3048 +04/16 18:00:50 - mmengine - INFO - Iter(train) [ 35400/160000] base_lr: 7.8612e-05 lr: 2.9065e-07 eta: 1 day, 10:31:09 time: 0.9988 data_time: 0.0046 memory: 8462 loss: 0.0187 decode.loss_ce: 0.0094 decode.acc_seg: 99.7528 aux.loss_ce: 0.0094 aux.acc_seg: 99.4188 +04/16 18:01:40 - mmengine - INFO - Iter(train) [ 35450/160000] base_lr: 7.8581e-05 lr: 2.9053e-07 eta: 1 day, 10:30:20 time: 0.9986 data_time: 0.0043 memory: 8462 loss: 0.0201 decode.loss_ce: 0.0105 decode.acc_seg: 99.6632 aux.loss_ce: 0.0096 aux.acc_seg: 99.0841 +04/16 18:02:30 - mmengine - INFO - Iter(train) [ 35500/160000] base_lr: 7.8549e-05 lr: 2.9041e-07 eta: 1 day, 10:29:30 time: 0.9994 data_time: 0.0048 memory: 8462 loss: 0.0184 decode.loss_ce: 0.0096 decode.acc_seg: 99.5234 aux.loss_ce: 0.0088 aux.acc_seg: 98.9374 +04/16 18:03:20 - mmengine - INFO - Iter(train) [ 35550/160000] base_lr: 7.8518e-05 lr: 2.9030e-07 eta: 1 day, 10:28:41 time: 0.9996 data_time: 0.0049 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0106 decode.acc_seg: 99.2283 aux.loss_ce: 0.0084 aux.acc_seg: 98.6685 +04/16 18:04:10 - mmengine - INFO - Iter(train) [ 35600/160000] base_lr: 7.8486e-05 lr: 2.9018e-07 eta: 1 day, 10:27:51 time: 0.9992 data_time: 0.0045 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0098 decode.acc_seg: 99.5480 aux.loss_ce: 0.0090 aux.acc_seg: 99.1564 +04/16 18:05:00 - mmengine - INFO - Iter(train) [ 35650/160000] base_lr: 7.8455e-05 lr: 2.9006e-07 eta: 1 day, 10:27:02 time: 0.9999 data_time: 0.0044 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0096 decode.acc_seg: 99.6662 aux.loss_ce: 0.0094 aux.acc_seg: 99.0801 +04/16 18:05:50 - mmengine - INFO - Iter(train) [ 35700/160000] base_lr: 7.8423e-05 lr: 2.8995e-07 eta: 1 day, 10:26:12 time: 0.9991 data_time: 0.0047 memory: 8462 loss: 0.0182 decode.loss_ce: 0.0095 decode.acc_seg: 99.6546 aux.loss_ce: 0.0087 aux.acc_seg: 99.3273 +04/16 18:06:40 - mmengine - INFO - Iter(train) [ 35750/160000] base_lr: 7.8392e-05 lr: 2.8983e-07 eta: 1 day, 10:25:23 time: 0.9987 data_time: 0.0049 memory: 8462 loss: 0.0200 decode.loss_ce: 0.0101 decode.acc_seg: 99.5106 aux.loss_ce: 0.0099 aux.acc_seg: 99.1920 +04/16 18:07:30 - mmengine - INFO - Iter(train) [ 35800/160000] base_lr: 7.8360e-05 lr: 2.8971e-07 eta: 1 day, 10:24:33 time: 0.9990 data_time: 0.0047 memory: 8462 loss: 0.0199 decode.loss_ce: 0.0105 decode.acc_seg: 99.5337 aux.loss_ce: 0.0093 aux.acc_seg: 99.0465 +04/16 18:08:20 - mmengine - INFO - Iter(train) [ 35850/160000] base_lr: 7.8329e-05 lr: 2.8960e-07 eta: 1 day, 10:23:44 time: 0.9997 data_time: 0.0044 memory: 8462 loss: 0.0196 decode.loss_ce: 0.0095 decode.acc_seg: 99.6407 aux.loss_ce: 0.0101 aux.acc_seg: 99.0826 +04/16 18:09:10 - mmengine - INFO - Iter(train) [ 35900/160000] base_lr: 7.8297e-05 lr: 2.8948e-07 eta: 1 day, 10:22:54 time: 0.9989 data_time: 0.0047 memory: 8462 loss: 0.0189 decode.loss_ce: 0.0097 decode.acc_seg: 99.5897 aux.loss_ce: 0.0092 aux.acc_seg: 99.1385 +04/16 18:10:00 - mmengine - INFO - Iter(train) [ 35950/160000] base_lr: 7.8265e-05 lr: 2.8936e-07 eta: 1 day, 10:22:05 time: 0.9987 data_time: 0.0048 memory: 8462 loss: 0.0202 decode.loss_ce: 0.0104 decode.acc_seg: 99.0692 aux.loss_ce: 0.0097 aux.acc_seg: 98.7633 +04/16 18:10:50 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 18:10:50 - mmengine - INFO - Iter(train) [ 36000/160000] base_lr: 7.8234e-05 lr: 2.8925e-07 eta: 1 day, 10:21:15 time: 0.9981 data_time: 0.0047 memory: 8462 loss: 0.0183 decode.loss_ce: 0.0094 decode.acc_seg: 99.6029 aux.loss_ce: 0.0090 aux.acc_seg: 98.8834 +04/16 18:11:40 - mmengine - INFO - Iter(train) [ 36050/160000] base_lr: 7.8202e-05 lr: 2.8913e-07 eta: 1 day, 10:20:26 time: 0.9997 data_time: 0.0044 memory: 8462 loss: 0.0183 decode.loss_ce: 0.0097 decode.acc_seg: 99.7293 aux.loss_ce: 0.0087 aux.acc_seg: 99.2447 +04/16 18:12:30 - mmengine - INFO - Iter(train) [ 36100/160000] base_lr: 7.8171e-05 lr: 2.8901e-07 eta: 1 day, 10:19:36 time: 0.9997 data_time: 0.0046 memory: 8462 loss: 0.0214 decode.loss_ce: 0.0109 decode.acc_seg: 99.2603 aux.loss_ce: 0.0105 aux.acc_seg: 98.5594 +04/16 18:13:20 - mmengine - INFO - Iter(train) [ 36150/160000] base_lr: 7.8139e-05 lr: 2.8890e-07 eta: 1 day, 10:18:47 time: 0.9995 data_time: 0.0047 memory: 8462 loss: 0.0192 decode.loss_ce: 0.0099 decode.acc_seg: 99.5678 aux.loss_ce: 0.0093 aux.acc_seg: 98.7810 +04/16 18:14:10 - mmengine - INFO - Iter(train) [ 36200/160000] base_lr: 7.8108e-05 lr: 2.8878e-07 eta: 1 day, 10:17:57 time: 0.9992 data_time: 0.0047 memory: 8462 loss: 0.0193 decode.loss_ce: 0.0100 decode.acc_seg: 99.6346 aux.loss_ce: 0.0093 aux.acc_seg: 99.3034 +04/16 18:15:00 - mmengine - INFO - Iter(train) [ 36250/160000] base_lr: 7.8076e-05 lr: 2.8866e-07 eta: 1 day, 10:17:07 time: 0.9997 data_time: 0.0042 memory: 8462 loss: 0.0195 decode.loss_ce: 0.0101 decode.acc_seg: 99.5901 aux.loss_ce: 0.0094 aux.acc_seg: 98.9717 +04/16 18:15:50 - mmengine - INFO - Iter(train) [ 36300/160000] base_lr: 7.8045e-05 lr: 2.8855e-07 eta: 1 day, 10:16:18 time: 0.9997 data_time: 0.0042 memory: 8462 loss: 0.0184 decode.loss_ce: 0.0092 decode.acc_seg: 99.5510 aux.loss_ce: 0.0093 aux.acc_seg: 98.6420 +04/16 18:16:40 - mmengine - INFO - Iter(train) [ 36350/160000] base_lr: 7.8013e-05 lr: 2.8843e-07 eta: 1 day, 10:15:28 time: 0.9988 data_time: 0.0045 memory: 8462 loss: 0.0192 decode.loss_ce: 0.0100 decode.acc_seg: 99.5678 aux.loss_ce: 0.0092 aux.acc_seg: 98.6696 +04/16 18:17:30 - mmengine - INFO - Iter(train) [ 36400/160000] base_lr: 7.7982e-05 lr: 2.8831e-07 eta: 1 day, 10:14:39 time: 0.9992 data_time: 0.0041 memory: 8462 loss: 0.0187 decode.loss_ce: 0.0093 decode.acc_seg: 99.6386 aux.loss_ce: 0.0093 aux.acc_seg: 99.1817 +04/16 18:18:20 - mmengine - INFO - Iter(train) [ 36450/160000] base_lr: 7.7950e-05 lr: 2.8820e-07 eta: 1 day, 10:13:49 time: 0.9990 data_time: 0.0049 memory: 8462 loss: 0.0202 decode.loss_ce: 0.0106 decode.acc_seg: 99.5274 aux.loss_ce: 0.0096 aux.acc_seg: 98.9008 +04/16 18:19:10 - mmengine - INFO - Iter(train) [ 36500/160000] base_lr: 7.7918e-05 lr: 2.8808e-07 eta: 1 day, 10:13:00 time: 0.9985 data_time: 0.0046 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0092 decode.acc_seg: 99.6563 aux.loss_ce: 0.0087 aux.acc_seg: 99.3046 +04/16 18:20:00 - mmengine - INFO - Iter(train) [ 36550/160000] base_lr: 7.7887e-05 lr: 2.8796e-07 eta: 1 day, 10:12:10 time: 0.9996 data_time: 0.0045 memory: 8462 loss: 0.0194 decode.loss_ce: 0.0103 decode.acc_seg: 99.3256 aux.loss_ce: 0.0092 aux.acc_seg: 98.7541 +04/16 18:20:50 - mmengine - INFO - Iter(train) [ 36600/160000] base_lr: 7.7855e-05 lr: 2.8785e-07 eta: 1 day, 10:11:21 time: 0.9989 data_time: 0.0047 memory: 8462 loss: 0.0191 decode.loss_ce: 0.0094 decode.acc_seg: 99.6321 aux.loss_ce: 0.0097 aux.acc_seg: 98.8707 +04/16 18:21:40 - mmengine - INFO - Iter(train) [ 36650/160000] base_lr: 7.7824e-05 lr: 2.8773e-07 eta: 1 day, 10:10:31 time: 0.9993 data_time: 0.0045 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0090 decode.acc_seg: 99.4877 aux.loss_ce: 0.0087 aux.acc_seg: 98.6252 +04/16 18:22:30 - mmengine - INFO - Iter(train) [ 36700/160000] base_lr: 7.7792e-05 lr: 2.8761e-07 eta: 1 day, 10:09:41 time: 0.9996 data_time: 0.0045 memory: 8462 loss: 0.0180 decode.loss_ce: 0.0094 decode.acc_seg: 99.7061 aux.loss_ce: 0.0087 aux.acc_seg: 99.2981 +04/16 18:23:20 - mmengine - INFO - Iter(train) [ 36750/160000] base_lr: 7.7761e-05 lr: 2.8750e-07 eta: 1 day, 10:08:52 time: 0.9982 data_time: 0.0046 memory: 8462 loss: 0.0200 decode.loss_ce: 0.0104 decode.acc_seg: 99.5789 aux.loss_ce: 0.0096 aux.acc_seg: 99.2010 +04/16 18:24:10 - mmengine - INFO - Iter(train) [ 36800/160000] base_lr: 7.7729e-05 lr: 2.8738e-07 eta: 1 day, 10:08:02 time: 0.9985 data_time: 0.0044 memory: 8462 loss: 0.0193 decode.loss_ce: 0.0096 decode.acc_seg: 99.6603 aux.loss_ce: 0.0098 aux.acc_seg: 99.0469 +04/16 18:25:00 - mmengine - INFO - Iter(train) [ 36850/160000] base_lr: 7.7698e-05 lr: 2.8726e-07 eta: 1 day, 10:07:13 time: 0.9994 data_time: 0.0050 memory: 8462 loss: 0.0201 decode.loss_ce: 0.0107 decode.acc_seg: 99.7566 aux.loss_ce: 0.0094 aux.acc_seg: 99.3483 +04/16 18:25:50 - mmengine - INFO - Iter(train) [ 36900/160000] base_lr: 7.7666e-05 lr: 2.8715e-07 eta: 1 day, 10:06:23 time: 1.0014 data_time: 0.0045 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0094 decode.acc_seg: 99.6845 aux.loss_ce: 0.0087 aux.acc_seg: 99.1907 +04/16 18:26:40 - mmengine - INFO - Iter(train) [ 36950/160000] base_lr: 7.7635e-05 lr: 2.8703e-07 eta: 1 day, 10:05:34 time: 1.0000 data_time: 0.0046 memory: 8462 loss: 0.0217 decode.loss_ce: 0.0114 decode.acc_seg: 99.5628 aux.loss_ce: 0.0103 aux.acc_seg: 98.9550 +04/16 18:27:30 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 18:27:30 - mmengine - INFO - Iter(train) [ 37000/160000] base_lr: 7.7603e-05 lr: 2.8691e-07 eta: 1 day, 10:04:44 time: 1.0006 data_time: 0.0045 memory: 8462 loss: 0.0191 decode.loss_ce: 0.0097 decode.acc_seg: 99.5579 aux.loss_ce: 0.0095 aux.acc_seg: 98.8516 +04/16 18:28:20 - mmengine - INFO - Iter(train) [ 37050/160000] base_lr: 7.7571e-05 lr: 2.8680e-07 eta: 1 day, 10:03:55 time: 1.0008 data_time: 0.0046 memory: 8462 loss: 0.0171 decode.loss_ce: 0.0089 decode.acc_seg: 99.7751 aux.loss_ce: 0.0082 aux.acc_seg: 99.3525 +04/16 18:29:10 - mmengine - INFO - Iter(train) [ 37100/160000] base_lr: 7.7540e-05 lr: 2.8668e-07 eta: 1 day, 10:03:05 time: 1.0001 data_time: 0.0045 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0088 decode.acc_seg: 99.5516 aux.loss_ce: 0.0078 aux.acc_seg: 98.9372 +04/16 18:29:59 - mmengine - INFO - Iter(train) [ 37150/160000] base_lr: 7.7508e-05 lr: 2.8656e-07 eta: 1 day, 10:02:16 time: 0.9991 data_time: 0.0046 memory: 8462 loss: 0.0183 decode.loss_ce: 0.0096 decode.acc_seg: 99.6342 aux.loss_ce: 0.0087 aux.acc_seg: 99.2096 +04/16 18:30:49 - mmengine - INFO - Iter(train) [ 37200/160000] base_lr: 7.7477e-05 lr: 2.8645e-07 eta: 1 day, 10:01:26 time: 0.9997 data_time: 0.0048 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0093 decode.acc_seg: 99.1800 aux.loss_ce: 0.0082 aux.acc_seg: 98.7444 +04/16 18:31:39 - mmengine - INFO - Iter(train) [ 37250/160000] base_lr: 7.7445e-05 lr: 2.8633e-07 eta: 1 day, 10:00:37 time: 0.9992 data_time: 0.0048 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0090 decode.acc_seg: 99.4818 aux.loss_ce: 0.0086 aux.acc_seg: 99.1272 +04/16 18:32:29 - mmengine - INFO - Iter(train) [ 37300/160000] base_lr: 7.7414e-05 lr: 2.8621e-07 eta: 1 day, 9:59:47 time: 0.9982 data_time: 0.0048 memory: 8462 loss: 0.0189 decode.loss_ce: 0.0099 decode.acc_seg: 99.5996 aux.loss_ce: 0.0090 aux.acc_seg: 98.9307 +04/16 18:33:19 - mmengine - INFO - Iter(train) [ 37350/160000] base_lr: 7.7382e-05 lr: 2.8610e-07 eta: 1 day, 9:58:57 time: 0.9999 data_time: 0.0042 memory: 8462 loss: 0.0197 decode.loss_ce: 0.0102 decode.acc_seg: 99.6471 aux.loss_ce: 0.0096 aux.acc_seg: 99.0953 +04/16 18:34:09 - mmengine - INFO - Iter(train) [ 37400/160000] base_lr: 7.7351e-05 lr: 2.8598e-07 eta: 1 day, 9:58:08 time: 0.9992 data_time: 0.0046 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0089 decode.acc_seg: 99.7519 aux.loss_ce: 0.0086 aux.acc_seg: 99.5729 +04/16 18:34:59 - mmengine - INFO - Iter(train) [ 37450/160000] base_lr: 7.7319e-05 lr: 2.8586e-07 eta: 1 day, 9:57:18 time: 0.9985 data_time: 0.0045 memory: 8462 loss: 0.0205 decode.loss_ce: 0.0102 decode.acc_seg: 99.4196 aux.loss_ce: 0.0103 aux.acc_seg: 98.3147 +04/16 18:35:49 - mmengine - INFO - Iter(train) [ 37500/160000] base_lr: 7.7288e-05 lr: 2.8575e-07 eta: 1 day, 9:56:29 time: 0.9999 data_time: 0.0049 memory: 8462 loss: 0.0219 decode.loss_ce: 0.0113 decode.acc_seg: 99.5279 aux.loss_ce: 0.0106 aux.acc_seg: 98.9407 +04/16 18:36:39 - mmengine - INFO - Iter(train) [ 37550/160000] base_lr: 7.7256e-05 lr: 2.8563e-07 eta: 1 day, 9:55:39 time: 1.0001 data_time: 0.0045 memory: 8462 loss: 0.0196 decode.loss_ce: 0.0099 decode.acc_seg: 99.6906 aux.loss_ce: 0.0097 aux.acc_seg: 99.2290 +04/16 18:37:29 - mmengine - INFO - Iter(train) [ 37600/160000] base_lr: 7.7224e-05 lr: 2.8551e-07 eta: 1 day, 9:54:50 time: 0.9987 data_time: 0.0043 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0093 decode.acc_seg: 99.5823 aux.loss_ce: 0.0081 aux.acc_seg: 99.2067 +04/16 18:38:19 - mmengine - INFO - Iter(train) [ 37650/160000] base_lr: 7.7193e-05 lr: 2.8540e-07 eta: 1 day, 9:54:00 time: 0.9978 data_time: 0.0046 memory: 8462 loss: 0.0171 decode.loss_ce: 0.0089 decode.acc_seg: 99.7307 aux.loss_ce: 0.0082 aux.acc_seg: 99.3969 +04/16 18:39:09 - mmengine - INFO - Iter(train) [ 37700/160000] base_lr: 7.7161e-05 lr: 2.8528e-07 eta: 1 day, 9:53:11 time: 0.9992 data_time: 0.0047 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0096 decode.acc_seg: 99.4446 aux.loss_ce: 0.0081 aux.acc_seg: 98.9332 +04/16 18:39:59 - mmengine - INFO - Iter(train) [ 37750/160000] base_lr: 7.7130e-05 lr: 2.8516e-07 eta: 1 day, 9:52:21 time: 0.9991 data_time: 0.0045 memory: 8462 loss: 0.0199 decode.loss_ce: 0.0102 decode.acc_seg: 99.5785 aux.loss_ce: 0.0097 aux.acc_seg: 99.0084 +04/16 18:40:49 - mmengine - INFO - Iter(train) [ 37800/160000] base_lr: 7.7098e-05 lr: 2.8505e-07 eta: 1 day, 9:51:31 time: 0.9975 data_time: 0.0041 memory: 8462 loss: 0.0180 decode.loss_ce: 0.0088 decode.acc_seg: 99.5834 aux.loss_ce: 0.0092 aux.acc_seg: 99.1297 +04/16 18:41:39 - mmengine - INFO - Iter(train) [ 37850/160000] base_lr: 7.7067e-05 lr: 2.8493e-07 eta: 1 day, 9:50:42 time: 0.9996 data_time: 0.0044 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0088 decode.acc_seg: 99.6132 aux.loss_ce: 0.0082 aux.acc_seg: 99.1386 +04/16 18:42:29 - mmengine - INFO - Iter(train) [ 37900/160000] base_lr: 7.7035e-05 lr: 2.8481e-07 eta: 1 day, 9:49:52 time: 0.9993 data_time: 0.0049 memory: 8462 loss: 0.0199 decode.loss_ce: 0.0105 decode.acc_seg: 99.5394 aux.loss_ce: 0.0094 aux.acc_seg: 99.1041 +04/16 18:43:19 - mmengine - INFO - Iter(train) [ 37950/160000] base_lr: 7.7004e-05 lr: 2.8470e-07 eta: 1 day, 9:49:03 time: 1.0004 data_time: 0.0044 memory: 8462 loss: 0.0214 decode.loss_ce: 0.0117 decode.acc_seg: 99.0931 aux.loss_ce: 0.0097 aux.acc_seg: 98.3505 +04/16 18:44:09 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 18:44:09 - mmengine - INFO - Iter(train) [ 38000/160000] base_lr: 7.6972e-05 lr: 2.8458e-07 eta: 1 day, 9:48:13 time: 0.9998 data_time: 0.0047 memory: 8462 loss: 0.0186 decode.loss_ce: 0.0093 decode.acc_seg: 99.6929 aux.loss_ce: 0.0094 aux.acc_seg: 99.2748 +04/16 18:44:59 - mmengine - INFO - Iter(train) [ 38050/160000] base_lr: 7.6941e-05 lr: 2.8447e-07 eta: 1 day, 9:47:24 time: 1.0002 data_time: 0.0043 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0094 decode.acc_seg: 99.6050 aux.loss_ce: 0.0096 aux.acc_seg: 99.0541 +04/16 18:45:49 - mmengine - INFO - Iter(train) [ 38100/160000] base_lr: 7.6909e-05 lr: 2.8435e-07 eta: 1 day, 9:46:34 time: 0.9999 data_time: 0.0047 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0088 decode.acc_seg: 99.5081 aux.loss_ce: 0.0088 aux.acc_seg: 98.7522 +04/16 18:46:39 - mmengine - INFO - Iter(train) [ 38150/160000] base_lr: 7.6877e-05 lr: 2.8423e-07 eta: 1 day, 9:45:45 time: 0.9997 data_time: 0.0045 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0078 decode.acc_seg: 99.7622 aux.loss_ce: 0.0075 aux.acc_seg: 99.4211 +04/16 18:47:29 - mmengine - INFO - Iter(train) [ 38200/160000] base_lr: 7.6846e-05 lr: 2.8412e-07 eta: 1 day, 9:44:55 time: 0.9987 data_time: 0.0043 memory: 8462 loss: 0.0186 decode.loss_ce: 0.0096 decode.acc_seg: 99.6744 aux.loss_ce: 0.0089 aux.acc_seg: 99.0860 +04/16 18:48:19 - mmengine - INFO - Iter(train) [ 38250/160000] base_lr: 7.6814e-05 lr: 2.8400e-07 eta: 1 day, 9:44:06 time: 1.0001 data_time: 0.0050 memory: 8462 loss: 0.0171 decode.loss_ce: 0.0088 decode.acc_seg: 99.6277 aux.loss_ce: 0.0083 aux.acc_seg: 99.1913 +04/16 18:49:09 - mmengine - INFO - Iter(train) [ 38300/160000] base_lr: 7.6783e-05 lr: 2.8388e-07 eta: 1 day, 9:43:16 time: 1.0000 data_time: 0.0043 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0099 decode.acc_seg: 99.6052 aux.loss_ce: 0.0090 aux.acc_seg: 99.2760 +04/16 18:49:59 - mmengine - INFO - Iter(train) [ 38350/160000] base_lr: 7.6751e-05 lr: 2.8377e-07 eta: 1 day, 9:42:26 time: 0.9981 data_time: 0.0045 memory: 8462 loss: 0.0214 decode.loss_ce: 0.0111 decode.acc_seg: 99.5485 aux.loss_ce: 0.0103 aux.acc_seg: 98.7944 +04/16 18:50:49 - mmengine - INFO - Iter(train) [ 38400/160000] base_lr: 7.6720e-05 lr: 2.8365e-07 eta: 1 day, 9:41:37 time: 0.9983 data_time: 0.0045 memory: 8462 loss: 0.0195 decode.loss_ce: 0.0105 decode.acc_seg: 99.6445 aux.loss_ce: 0.0091 aux.acc_seg: 99.1837 +04/16 18:51:39 - mmengine - INFO - Iter(train) [ 38450/160000] base_lr: 7.6688e-05 lr: 2.8353e-07 eta: 1 day, 9:40:47 time: 0.9997 data_time: 0.0044 memory: 8462 loss: 0.0185 decode.loss_ce: 0.0093 decode.acc_seg: 99.5972 aux.loss_ce: 0.0093 aux.acc_seg: 98.8462 +04/16 18:52:29 - mmengine - INFO - Iter(train) [ 38500/160000] base_lr: 7.6657e-05 lr: 2.8342e-07 eta: 1 day, 9:39:58 time: 0.9984 data_time: 0.0044 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0088 decode.acc_seg: 99.6521 aux.loss_ce: 0.0084 aux.acc_seg: 99.0269 +04/16 18:53:19 - mmengine - INFO - Iter(train) [ 38550/160000] base_lr: 7.6625e-05 lr: 2.8330e-07 eta: 1 day, 9:39:08 time: 0.9995 data_time: 0.0050 memory: 8462 loss: 0.0165 decode.loss_ce: 0.0081 decode.acc_seg: 99.6111 aux.loss_ce: 0.0084 aux.acc_seg: 99.1764 +04/16 18:54:09 - mmengine - INFO - Iter(train) [ 38600/160000] base_lr: 7.6594e-05 lr: 2.8318e-07 eta: 1 day, 9:38:18 time: 0.9992 data_time: 0.0044 memory: 8462 loss: 0.0209 decode.loss_ce: 0.0109 decode.acc_seg: 99.6038 aux.loss_ce: 0.0100 aux.acc_seg: 98.9075 +04/16 18:54:59 - mmengine - INFO - Iter(train) [ 38650/160000] base_lr: 7.6562e-05 lr: 2.8307e-07 eta: 1 day, 9:37:29 time: 1.0002 data_time: 0.0046 memory: 8462 loss: 0.0171 decode.loss_ce: 0.0090 decode.acc_seg: 99.7551 aux.loss_ce: 0.0082 aux.acc_seg: 99.3549 +04/16 18:55:49 - mmengine - INFO - Iter(train) [ 38700/160000] base_lr: 7.6530e-05 lr: 2.8295e-07 eta: 1 day, 9:36:39 time: 0.9990 data_time: 0.0046 memory: 8462 loss: 0.0202 decode.loss_ce: 0.0108 decode.acc_seg: 99.6368 aux.loss_ce: 0.0094 aux.acc_seg: 99.2714 +04/16 18:56:39 - mmengine - INFO - Iter(train) [ 38750/160000] base_lr: 7.6499e-05 lr: 2.8283e-07 eta: 1 day, 9:35:50 time: 0.9998 data_time: 0.0045 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0092 decode.acc_seg: 99.7028 aux.loss_ce: 0.0086 aux.acc_seg: 99.0713 +04/16 18:57:29 - mmengine - INFO - Iter(train) [ 38800/160000] base_lr: 7.6467e-05 lr: 2.8272e-07 eta: 1 day, 9:35:00 time: 0.9993 data_time: 0.0047 memory: 8462 loss: 0.0212 decode.loss_ce: 0.0122 decode.acc_seg: 99.4604 aux.loss_ce: 0.0091 aux.acc_seg: 98.9594 +04/16 18:58:19 - mmengine - INFO - Iter(train) [ 38850/160000] base_lr: 7.6436e-05 lr: 2.8260e-07 eta: 1 day, 9:34:11 time: 1.0001 data_time: 0.0050 memory: 8462 loss: 0.0189 decode.loss_ce: 0.0095 decode.acc_seg: 99.6334 aux.loss_ce: 0.0094 aux.acc_seg: 98.6723 +04/16 18:59:09 - mmengine - INFO - Iter(train) [ 38900/160000] base_lr: 7.6404e-05 lr: 2.8248e-07 eta: 1 day, 9:33:21 time: 0.9996 data_time: 0.0043 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0098 decode.acc_seg: 99.7036 aux.loss_ce: 0.0090 aux.acc_seg: 99.3160 +04/16 18:59:59 - mmengine - INFO - Iter(train) [ 38950/160000] base_lr: 7.6373e-05 lr: 2.8237e-07 eta: 1 day, 9:32:32 time: 0.9992 data_time: 0.0042 memory: 8462 loss: 0.0186 decode.loss_ce: 0.0102 decode.acc_seg: 99.7040 aux.loss_ce: 0.0084 aux.acc_seg: 99.3305 +04/16 19:00:49 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 19:00:49 - mmengine - INFO - Iter(train) [ 39000/160000] base_lr: 7.6341e-05 lr: 2.8225e-07 eta: 1 day, 9:31:42 time: 0.9997 data_time: 0.0048 memory: 8462 loss: 0.0201 decode.loss_ce: 0.0103 decode.acc_seg: 99.6683 aux.loss_ce: 0.0098 aux.acc_seg: 99.2292 +04/16 19:01:39 - mmengine - INFO - Iter(train) [ 39050/160000] base_lr: 7.6310e-05 lr: 2.8213e-07 eta: 1 day, 9:30:53 time: 0.9997 data_time: 0.0048 memory: 8462 loss: 0.0189 decode.loss_ce: 0.0098 decode.acc_seg: 99.3977 aux.loss_ce: 0.0091 aux.acc_seg: 98.8224 +04/16 19:02:29 - mmengine - INFO - Iter(train) [ 39100/160000] base_lr: 7.6278e-05 lr: 2.8202e-07 eta: 1 day, 9:30:03 time: 1.0014 data_time: 0.0046 memory: 8462 loss: 0.0193 decode.loss_ce: 0.0099 decode.acc_seg: 99.6540 aux.loss_ce: 0.0094 aux.acc_seg: 99.2640 +04/16 19:03:19 - mmengine - INFO - Iter(train) [ 39150/160000] base_lr: 7.6247e-05 lr: 2.8190e-07 eta: 1 day, 9:29:14 time: 0.9989 data_time: 0.0046 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0091 decode.acc_seg: 99.6185 aux.loss_ce: 0.0087 aux.acc_seg: 99.2218 +04/16 19:04:09 - mmengine - INFO - Iter(train) [ 39200/160000] base_lr: 7.6215e-05 lr: 2.8178e-07 eta: 1 day, 9:28:24 time: 0.9998 data_time: 0.0045 memory: 8462 loss: 0.0213 decode.loss_ce: 0.0108 decode.acc_seg: 99.2207 aux.loss_ce: 0.0105 aux.acc_seg: 98.3152 +04/16 19:04:59 - mmengine - INFO - Iter(train) [ 39250/160000] base_lr: 7.6183e-05 lr: 2.8167e-07 eta: 1 day, 9:27:34 time: 0.9994 data_time: 0.0043 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0093 decode.acc_seg: 99.7082 aux.loss_ce: 0.0088 aux.acc_seg: 99.0751 +04/16 19:05:49 - mmengine - INFO - Iter(train) [ 39300/160000] base_lr: 7.6152e-05 lr: 2.8155e-07 eta: 1 day, 9:26:45 time: 0.9989 data_time: 0.0045 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0092 decode.acc_seg: 99.4886 aux.loss_ce: 0.0083 aux.acc_seg: 99.0915 +04/16 19:06:39 - mmengine - INFO - Iter(train) [ 39350/160000] base_lr: 7.6120e-05 lr: 2.8143e-07 eta: 1 day, 9:25:55 time: 1.0006 data_time: 0.0042 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0091 decode.acc_seg: 99.7728 aux.loss_ce: 0.0085 aux.acc_seg: 99.3750 +04/16 19:07:29 - mmengine - INFO - Iter(train) [ 39400/160000] base_lr: 7.6089e-05 lr: 2.8132e-07 eta: 1 day, 9:25:06 time: 1.0001 data_time: 0.0042 memory: 8462 loss: 0.0180 decode.loss_ce: 0.0090 decode.acc_seg: 99.6607 aux.loss_ce: 0.0090 aux.acc_seg: 99.2296 +04/16 19:08:19 - mmengine - INFO - Iter(train) [ 39450/160000] base_lr: 7.6057e-05 lr: 2.8120e-07 eta: 1 day, 9:24:16 time: 1.0009 data_time: 0.0044 memory: 8462 loss: 0.0202 decode.loss_ce: 0.0110 decode.acc_seg: 99.5613 aux.loss_ce: 0.0092 aux.acc_seg: 99.1129 +04/16 19:09:09 - mmengine - INFO - Iter(train) [ 39500/160000] base_lr: 7.6026e-05 lr: 2.8108e-07 eta: 1 day, 9:23:27 time: 1.0001 data_time: 0.0047 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0086 decode.acc_seg: 99.5474 aux.loss_ce: 0.0085 aux.acc_seg: 98.8457 +04/16 19:09:59 - mmengine - INFO - Iter(train) [ 39550/160000] base_lr: 7.5994e-05 lr: 2.8097e-07 eta: 1 day, 9:22:37 time: 1.0000 data_time: 0.0047 memory: 8462 loss: 0.0207 decode.loss_ce: 0.0106 decode.acc_seg: 99.4545 aux.loss_ce: 0.0101 aux.acc_seg: 98.9027 +04/16 19:10:49 - mmengine - INFO - Iter(train) [ 39600/160000] base_lr: 7.5963e-05 lr: 2.8085e-07 eta: 1 day, 9:21:48 time: 1.0012 data_time: 0.0046 memory: 8462 loss: 0.0199 decode.loss_ce: 0.0102 decode.acc_seg: 99.7526 aux.loss_ce: 0.0096 aux.acc_seg: 99.2126 +04/16 19:11:39 - mmengine - INFO - Iter(train) [ 39650/160000] base_lr: 7.5931e-05 lr: 2.8073e-07 eta: 1 day, 9:20:58 time: 1.0005 data_time: 0.0045 memory: 8462 loss: 0.0189 decode.loss_ce: 0.0095 decode.acc_seg: 99.6218 aux.loss_ce: 0.0093 aux.acc_seg: 98.9985 +04/16 19:12:29 - mmengine - INFO - Iter(train) [ 39700/160000] base_lr: 7.5900e-05 lr: 2.8062e-07 eta: 1 day, 9:20:09 time: 1.0000 data_time: 0.0043 memory: 8462 loss: 0.0195 decode.loss_ce: 0.0102 decode.acc_seg: 99.7480 aux.loss_ce: 0.0093 aux.acc_seg: 99.3753 +04/16 19:13:19 - mmengine - INFO - Iter(train) [ 39750/160000] base_lr: 7.5868e-05 lr: 2.8050e-07 eta: 1 day, 9:19:19 time: 0.9997 data_time: 0.0044 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0094 decode.acc_seg: 99.5581 aux.loss_ce: 0.0084 aux.acc_seg: 98.9313 +04/16 19:14:09 - mmengine - INFO - Iter(train) [ 39800/160000] base_lr: 7.5836e-05 lr: 2.8038e-07 eta: 1 day, 9:18:30 time: 1.0000 data_time: 0.0045 memory: 8462 loss: 0.0231 decode.loss_ce: 0.0119 decode.acc_seg: 99.5714 aux.loss_ce: 0.0112 aux.acc_seg: 98.8041 +04/16 19:14:59 - mmengine - INFO - Iter(train) [ 39850/160000] base_lr: 7.5805e-05 lr: 2.8027e-07 eta: 1 day, 9:17:40 time: 0.9999 data_time: 0.0050 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0082 decode.acc_seg: 99.5243 aux.loss_ce: 0.0079 aux.acc_seg: 99.1131 +04/16 19:15:49 - mmengine - INFO - Iter(train) [ 39900/160000] base_lr: 7.5773e-05 lr: 2.8015e-07 eta: 1 day, 9:16:51 time: 0.9988 data_time: 0.0042 memory: 8462 loss: 0.0231 decode.loss_ce: 0.0121 decode.acc_seg: 99.7026 aux.loss_ce: 0.0110 aux.acc_seg: 99.2476 +04/16 19:16:39 - mmengine - INFO - Iter(train) [ 39950/160000] base_lr: 7.5742e-05 lr: 2.8003e-07 eta: 1 day, 9:16:01 time: 1.0016 data_time: 0.0045 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0083 decode.acc_seg: 99.5977 aux.loss_ce: 0.0078 aux.acc_seg: 99.3048 +04/16 19:17:29 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 19:17:29 - mmengine - INFO - Iter(train) [ 40000/160000] base_lr: 7.5710e-05 lr: 2.7992e-07 eta: 1 day, 9:15:12 time: 0.9995 data_time: 0.0045 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0081 decode.acc_seg: 99.6803 aux.loss_ce: 0.0080 aux.acc_seg: 99.0583 +04/16 19:17:29 - mmengine - INFO - Saving checkpoint at 40000 iterations +04/16 19:17:38 - mmengine - INFO - Iter(val) [ 50/355] eta: 0:00:35 time: 0.1155 data_time: 0.0013 memory: 4004 +04/16 19:17:44 - mmengine - INFO - Iter(val) [100/355] eta: 0:00:29 time: 0.1152 data_time: 0.0013 memory: 4004 +04/16 19:17:50 - mmengine - INFO - Iter(val) [150/355] eta: 0:00:23 time: 0.1157 data_time: 0.0014 memory: 4004 +04/16 19:17:56 - mmengine - INFO - Iter(val) [200/355] eta: 0:00:17 time: 0.1158 data_time: 0.0015 memory: 4004 +04/16 19:18:02 - mmengine - INFO - Iter(val) [250/355] eta: 0:00:12 time: 0.1160 data_time: 0.0015 memory: 4004 +04/16 19:18:07 - mmengine - INFO - Iter(val) [300/355] eta: 0:00:06 time: 0.1157 data_time: 0.0014 memory: 4004 +04/16 19:18:13 - mmengine - INFO - Iter(val) [350/355] eta: 0:00:00 time: 0.1158 data_time: 0.0014 memory: 4004 +04/16 19:18:14 - mmengine - INFO - per class results: +04/16 19:18:14 - mmengine - INFO - ++------------+-------+-------+ +| Class | IoU | Acc | ++------------+-------+-------+ +| background | 99.19 | 99.6 | +| contrast | 82.27 | 90.12 | ++------------+-------+-------+ +04/16 19:18:14 - mmengine - INFO - Iter(val) [355/355] aAcc: 99.2200 mIoU: 90.7300 mAcc: 94.8600 data_time: 0.0016 time: 0.1159 +04/16 19:19:04 - mmengine - INFO - Iter(train) [ 40050/160000] base_lr: 7.5679e-05 lr: 2.7980e-07 eta: 1 day, 9:14:23 time: 0.9998 data_time: 0.0045 memory: 8462 loss: 0.0185 decode.loss_ce: 0.0095 decode.acc_seg: 99.6161 aux.loss_ce: 0.0090 aux.acc_seg: 99.1032 +04/16 19:19:54 - mmengine - INFO - Iter(train) [ 40100/160000] base_lr: 7.5647e-05 lr: 2.7968e-07 eta: 1 day, 9:13:34 time: 1.0007 data_time: 0.0046 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0091 decode.acc_seg: 99.5743 aux.loss_ce: 0.0096 aux.acc_seg: 98.7871 +04/16 19:20:44 - mmengine - INFO - Iter(train) [ 40150/160000] base_lr: 7.5616e-05 lr: 2.7957e-07 eta: 1 day, 9:12:44 time: 1.0006 data_time: 0.0048 memory: 8462 loss: 0.0203 decode.loss_ce: 0.0101 decode.acc_seg: 99.5077 aux.loss_ce: 0.0102 aux.acc_seg: 98.5815 +04/16 19:21:34 - mmengine - INFO - Iter(train) [ 40200/160000] base_lr: 7.5584e-05 lr: 2.7945e-07 eta: 1 day, 9:11:55 time: 0.9997 data_time: 0.0049 memory: 8462 loss: 0.0204 decode.loss_ce: 0.0107 decode.acc_seg: 99.3263 aux.loss_ce: 0.0098 aux.acc_seg: 98.9107 +04/16 19:22:24 - mmengine - INFO - Iter(train) [ 40250/160000] base_lr: 7.5553e-05 lr: 2.7933e-07 eta: 1 day, 9:11:05 time: 0.9995 data_time: 0.0048 memory: 8462 loss: 0.0229 decode.loss_ce: 0.0130 decode.acc_seg: 99.6286 aux.loss_ce: 0.0099 aux.acc_seg: 99.2659 +04/16 19:23:14 - mmengine - INFO - Iter(train) [ 40300/160000] base_lr: 7.5521e-05 lr: 2.7922e-07 eta: 1 day, 9:10:16 time: 1.0003 data_time: 0.0049 memory: 8462 loss: 0.0182 decode.loss_ce: 0.0095 decode.acc_seg: 99.6624 aux.loss_ce: 0.0088 aux.acc_seg: 99.3376 +04/16 19:24:04 - mmengine - INFO - Iter(train) [ 40350/160000] base_lr: 7.5489e-05 lr: 2.7910e-07 eta: 1 day, 9:09:26 time: 0.9995 data_time: 0.0044 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0068 decode.acc_seg: 99.7240 aux.loss_ce: 0.0072 aux.acc_seg: 99.3477 +04/16 19:24:54 - mmengine - INFO - Iter(train) [ 40400/160000] base_lr: 7.5458e-05 lr: 2.7898e-07 eta: 1 day, 9:08:36 time: 0.9993 data_time: 0.0046 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0086 decode.acc_seg: 99.6208 aux.loss_ce: 0.0081 aux.acc_seg: 98.8434 +04/16 19:25:44 - mmengine - INFO - Iter(train) [ 40450/160000] base_lr: 7.5426e-05 lr: 2.7887e-07 eta: 1 day, 9:07:47 time: 1.0007 data_time: 0.0046 memory: 8462 loss: 0.0195 decode.loss_ce: 0.0100 decode.acc_seg: 99.5022 aux.loss_ce: 0.0094 aux.acc_seg: 98.7232 +04/16 19:26:34 - mmengine - INFO - Iter(train) [ 40500/160000] base_lr: 7.5395e-05 lr: 2.7875e-07 eta: 1 day, 9:06:57 time: 0.9991 data_time: 0.0042 memory: 8462 loss: 0.0192 decode.loss_ce: 0.0102 decode.acc_seg: 99.5878 aux.loss_ce: 0.0090 aux.acc_seg: 98.8897 +04/16 19:27:24 - mmengine - INFO - Iter(train) [ 40550/160000] base_lr: 7.5363e-05 lr: 2.7863e-07 eta: 1 day, 9:06:08 time: 1.0002 data_time: 0.0045 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0103 decode.acc_seg: 99.4846 aux.loss_ce: 0.0085 aux.acc_seg: 99.0162 +04/16 19:28:14 - mmengine - INFO - Iter(train) [ 40600/160000] base_lr: 7.5332e-05 lr: 2.7852e-07 eta: 1 day, 9:05:18 time: 1.0003 data_time: 0.0043 memory: 8462 loss: 0.0186 decode.loss_ce: 0.0093 decode.acc_seg: 99.5790 aux.loss_ce: 0.0093 aux.acc_seg: 99.0683 +04/16 19:29:04 - mmengine - INFO - Iter(train) [ 40650/160000] base_lr: 7.5300e-05 lr: 2.7840e-07 eta: 1 day, 9:04:29 time: 1.0003 data_time: 0.0044 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0089 decode.acc_seg: 99.6830 aux.loss_ce: 0.0091 aux.acc_seg: 99.3746 +04/16 19:29:54 - mmengine - INFO - Iter(train) [ 40700/160000] base_lr: 7.5269e-05 lr: 2.7828e-07 eta: 1 day, 9:03:39 time: 1.0007 data_time: 0.0047 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0093 decode.acc_seg: 99.6933 aux.loss_ce: 0.0088 aux.acc_seg: 99.0902 +04/16 19:30:44 - mmengine - INFO - Iter(train) [ 40750/160000] base_lr: 7.5237e-05 lr: 2.7817e-07 eta: 1 day, 9:02:50 time: 1.0000 data_time: 0.0044 memory: 8462 loss: 0.0180 decode.loss_ce: 0.0095 decode.acc_seg: 99.6046 aux.loss_ce: 0.0085 aux.acc_seg: 99.1165 +04/16 19:31:34 - mmengine - INFO - Iter(train) [ 40800/160000] base_lr: 7.5206e-05 lr: 2.7805e-07 eta: 1 day, 9:02:00 time: 1.0013 data_time: 0.0043 memory: 8462 loss: 0.0169 decode.loss_ce: 0.0086 decode.acc_seg: 99.6269 aux.loss_ce: 0.0084 aux.acc_seg: 99.3904 +04/16 19:32:24 - mmengine - INFO - Iter(train) [ 40850/160000] base_lr: 7.5174e-05 lr: 2.7793e-07 eta: 1 day, 9:01:11 time: 1.0006 data_time: 0.0045 memory: 8462 loss: 0.0187 decode.loss_ce: 0.0098 decode.acc_seg: 99.5447 aux.loss_ce: 0.0089 aux.acc_seg: 99.0856 +04/16 19:33:14 - mmengine - INFO - Iter(train) [ 40900/160000] base_lr: 7.5142e-05 lr: 2.7782e-07 eta: 1 day, 9:00:21 time: 0.9992 data_time: 0.0043 memory: 8462 loss: 0.0183 decode.loss_ce: 0.0093 decode.acc_seg: 99.3580 aux.loss_ce: 0.0090 aux.acc_seg: 98.1644 +04/16 19:34:04 - mmengine - INFO - Iter(train) [ 40950/160000] base_lr: 7.5111e-05 lr: 2.7770e-07 eta: 1 day, 8:59:31 time: 0.9990 data_time: 0.0043 memory: 8462 loss: 0.0211 decode.loss_ce: 0.0113 decode.acc_seg: 99.5575 aux.loss_ce: 0.0098 aux.acc_seg: 98.8930 +04/16 19:34:54 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 19:34:54 - mmengine - INFO - Iter(train) [ 41000/160000] base_lr: 7.5079e-05 lr: 2.7758e-07 eta: 1 day, 8:58:42 time: 1.0010 data_time: 0.0042 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0083 decode.acc_seg: 99.5953 aux.loss_ce: 0.0077 aux.acc_seg: 99.0036 +04/16 19:35:44 - mmengine - INFO - Iter(train) [ 41050/160000] base_lr: 7.5048e-05 lr: 2.7747e-07 eta: 1 day, 8:57:52 time: 0.9994 data_time: 0.0044 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0088 decode.acc_seg: 99.6519 aux.loss_ce: 0.0084 aux.acc_seg: 99.2237 +04/16 19:36:34 - mmengine - INFO - Iter(train) [ 41100/160000] base_lr: 7.5016e-05 lr: 2.7735e-07 eta: 1 day, 8:57:02 time: 0.9982 data_time: 0.0046 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0081 decode.acc_seg: 99.7375 aux.loss_ce: 0.0076 aux.acc_seg: 99.3769 +04/16 19:37:24 - mmengine - INFO - Iter(train) [ 41150/160000] base_lr: 7.4985e-05 lr: 2.7723e-07 eta: 1 day, 8:56:13 time: 0.9989 data_time: 0.0048 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0091 decode.acc_seg: 99.6120 aux.loss_ce: 0.0087 aux.acc_seg: 99.0131 +04/16 19:38:14 - mmengine - INFO - Iter(train) [ 41200/160000] base_lr: 7.4953e-05 lr: 2.7712e-07 eta: 1 day, 8:55:23 time: 1.0006 data_time: 0.0050 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0083 decode.acc_seg: 99.6492 aux.loss_ce: 0.0083 aux.acc_seg: 99.1137 +04/16 19:39:04 - mmengine - INFO - Iter(train) [ 41250/160000] base_lr: 7.4922e-05 lr: 2.7700e-07 eta: 1 day, 8:54:34 time: 1.0003 data_time: 0.0045 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0090 decode.acc_seg: 99.5592 aux.loss_ce: 0.0086 aux.acc_seg: 99.1472 +04/16 19:39:54 - mmengine - INFO - Iter(train) [ 41300/160000] base_lr: 7.4890e-05 lr: 2.7688e-07 eta: 1 day, 8:53:44 time: 0.9998 data_time: 0.0046 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0090 decode.acc_seg: 99.6645 aux.loss_ce: 0.0087 aux.acc_seg: 99.1854 +04/16 19:40:44 - mmengine - INFO - Iter(train) [ 41350/160000] base_lr: 7.4859e-05 lr: 2.7677e-07 eta: 1 day, 8:52:55 time: 1.0001 data_time: 0.0047 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0093 decode.acc_seg: 99.7248 aux.loss_ce: 0.0088 aux.acc_seg: 99.3046 +04/16 19:41:34 - mmengine - INFO - Iter(train) [ 41400/160000] base_lr: 7.4827e-05 lr: 2.7665e-07 eta: 1 day, 8:52:05 time: 0.9995 data_time: 0.0052 memory: 8462 loss: 0.0186 decode.loss_ce: 0.0090 decode.acc_seg: 99.7309 aux.loss_ce: 0.0096 aux.acc_seg: 99.1282 +04/16 19:42:24 - mmengine - INFO - Iter(train) [ 41450/160000] base_lr: 7.4795e-05 lr: 2.7653e-07 eta: 1 day, 8:51:15 time: 0.9994 data_time: 0.0045 memory: 8462 loss: 0.0192 decode.loss_ce: 0.0093 decode.acc_seg: 99.6325 aux.loss_ce: 0.0099 aux.acc_seg: 98.8464 +04/16 19:43:14 - mmengine - INFO - Iter(train) [ 41500/160000] base_lr: 7.4764e-05 lr: 2.7642e-07 eta: 1 day, 8:50:26 time: 0.9989 data_time: 0.0049 memory: 8462 loss: 0.0216 decode.loss_ce: 0.0108 decode.acc_seg: 99.5132 aux.loss_ce: 0.0109 aux.acc_seg: 98.7885 +04/16 19:44:04 - mmengine - INFO - Iter(train) [ 41550/160000] base_lr: 7.4732e-05 lr: 2.7630e-07 eta: 1 day, 8:49:36 time: 0.9990 data_time: 0.0044 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0098 decode.acc_seg: 99.6649 aux.loss_ce: 0.0090 aux.acc_seg: 99.2594 +04/16 19:44:54 - mmengine - INFO - Iter(train) [ 41600/160000] base_lr: 7.4701e-05 lr: 2.7618e-07 eta: 1 day, 8:48:47 time: 0.9991 data_time: 0.0044 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0084 decode.acc_seg: 99.7141 aux.loss_ce: 0.0091 aux.acc_seg: 99.4532 +04/16 19:45:44 - mmengine - INFO - Iter(train) [ 41650/160000] base_lr: 7.4669e-05 lr: 2.7607e-07 eta: 1 day, 8:47:57 time: 0.9995 data_time: 0.0045 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0089 decode.acc_seg: 99.7446 aux.loss_ce: 0.0089 aux.acc_seg: 99.2710 +04/16 19:46:34 - mmengine - INFO - Iter(train) [ 41700/160000] base_lr: 7.4638e-05 lr: 2.7595e-07 eta: 1 day, 8:47:07 time: 0.9989 data_time: 0.0042 memory: 8462 loss: 0.0195 decode.loss_ce: 0.0099 decode.acc_seg: 99.6471 aux.loss_ce: 0.0096 aux.acc_seg: 99.0694 +04/16 19:47:24 - mmengine - INFO - Iter(train) [ 41750/160000] base_lr: 7.4606e-05 lr: 2.7583e-07 eta: 1 day, 8:46:18 time: 0.9997 data_time: 0.0048 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0079 decode.acc_seg: 99.7646 aux.loss_ce: 0.0086 aux.acc_seg: 99.1426 +04/16 19:48:14 - mmengine - INFO - Iter(train) [ 41800/160000] base_lr: 7.4575e-05 lr: 2.7572e-07 eta: 1 day, 8:45:28 time: 0.9994 data_time: 0.0046 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0089 decode.acc_seg: 99.6597 aux.loss_ce: 0.0089 aux.acc_seg: 98.9962 +04/16 19:49:04 - mmengine - INFO - Iter(train) [ 41850/160000] base_lr: 7.4543e-05 lr: 2.7560e-07 eta: 1 day, 8:44:39 time: 0.9997 data_time: 0.0051 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0093 decode.acc_seg: 99.6857 aux.loss_ce: 0.0088 aux.acc_seg: 99.4478 +04/16 19:49:54 - mmengine - INFO - Iter(train) [ 41900/160000] base_lr: 7.4512e-05 lr: 2.7548e-07 eta: 1 day, 8:43:49 time: 0.9992 data_time: 0.0046 memory: 8462 loss: 0.0146 decode.loss_ce: 0.0074 decode.acc_seg: 99.6201 aux.loss_ce: 0.0072 aux.acc_seg: 99.2416 +04/16 19:50:44 - mmengine - INFO - Iter(train) [ 41950/160000] base_lr: 7.4480e-05 lr: 2.7537e-07 eta: 1 day, 8:42:59 time: 0.9993 data_time: 0.0045 memory: 8462 loss: 0.0208 decode.loss_ce: 0.0104 decode.acc_seg: 99.4757 aux.loss_ce: 0.0104 aux.acc_seg: 98.8832 +04/16 19:51:34 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 19:51:34 - mmengine - INFO - Iter(train) [ 42000/160000] base_lr: 7.4448e-05 lr: 2.7525e-07 eta: 1 day, 8:42:10 time: 0.9996 data_time: 0.0045 memory: 8462 loss: 0.0199 decode.loss_ce: 0.0102 decode.acc_seg: 99.6668 aux.loss_ce: 0.0097 aux.acc_seg: 99.1030 +04/16 19:52:24 - mmengine - INFO - Iter(train) [ 42050/160000] base_lr: 7.4417e-05 lr: 2.7513e-07 eta: 1 day, 8:41:20 time: 1.0008 data_time: 0.0042 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0097 decode.acc_seg: 99.6855 aux.loss_ce: 0.0093 aux.acc_seg: 98.9084 +04/16 19:53:14 - mmengine - INFO - Iter(train) [ 42100/160000] base_lr: 7.4385e-05 lr: 2.7502e-07 eta: 1 day, 8:40:30 time: 0.9993 data_time: 0.0045 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0090 decode.acc_seg: 99.6357 aux.loss_ce: 0.0087 aux.acc_seg: 99.1886 +04/16 19:54:04 - mmengine - INFO - Iter(train) [ 42150/160000] base_lr: 7.4354e-05 lr: 2.7490e-07 eta: 1 day, 8:39:41 time: 0.9988 data_time: 0.0043 memory: 8462 loss: 0.0192 decode.loss_ce: 0.0093 decode.acc_seg: 99.5745 aux.loss_ce: 0.0099 aux.acc_seg: 98.7017 +04/16 19:54:54 - mmengine - INFO - Iter(train) [ 42200/160000] base_lr: 7.4322e-05 lr: 2.7478e-07 eta: 1 day, 8:38:51 time: 1.0003 data_time: 0.0045 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0090 decode.acc_seg: 99.6887 aux.loss_ce: 0.0083 aux.acc_seg: 99.2498 +04/16 19:55:44 - mmengine - INFO - Iter(train) [ 42250/160000] base_lr: 7.4291e-05 lr: 2.7467e-07 eta: 1 day, 8:38:02 time: 0.9978 data_time: 0.0046 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0093 decode.acc_seg: 99.5541 aux.loss_ce: 0.0086 aux.acc_seg: 98.9265 +04/16 19:56:34 - mmengine - INFO - Iter(train) [ 42300/160000] base_lr: 7.4259e-05 lr: 2.7455e-07 eta: 1 day, 8:37:12 time: 1.0007 data_time: 0.0045 memory: 8462 loss: 0.0186 decode.loss_ce: 0.0096 decode.acc_seg: 99.6393 aux.loss_ce: 0.0089 aux.acc_seg: 99.1846 +04/16 19:57:24 - mmengine - INFO - Iter(train) [ 42350/160000] base_lr: 7.4228e-05 lr: 2.7443e-07 eta: 1 day, 8:36:22 time: 0.9991 data_time: 0.0044 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0091 decode.acc_seg: 99.6151 aux.loss_ce: 0.0097 aux.acc_seg: 98.7541 +04/16 19:58:14 - mmengine - INFO - Iter(train) [ 42400/160000] base_lr: 7.4196e-05 lr: 2.7432e-07 eta: 1 day, 8:35:33 time: 1.0001 data_time: 0.0049 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0081 decode.acc_seg: 99.6111 aux.loss_ce: 0.0080 aux.acc_seg: 99.0931 +04/16 19:59:04 - mmengine - INFO - Iter(train) [ 42450/160000] base_lr: 7.4165e-05 lr: 2.7420e-07 eta: 1 day, 8:34:43 time: 1.0012 data_time: 0.0051 memory: 8462 loss: 0.0212 decode.loss_ce: 0.0105 decode.acc_seg: 99.4593 aux.loss_ce: 0.0107 aux.acc_seg: 98.3274 +04/16 19:59:54 - mmengine - INFO - Iter(train) [ 42500/160000] base_lr: 7.4133e-05 lr: 2.7408e-07 eta: 1 day, 8:33:54 time: 1.0000 data_time: 0.0048 memory: 8462 loss: 0.0189 decode.loss_ce: 0.0096 decode.acc_seg: 99.6241 aux.loss_ce: 0.0093 aux.acc_seg: 98.9977 +04/16 20:00:44 - mmengine - INFO - Iter(train) [ 42550/160000] base_lr: 7.4101e-05 lr: 2.7397e-07 eta: 1 day, 8:33:04 time: 1.0005 data_time: 0.0042 memory: 8462 loss: 0.0169 decode.loss_ce: 0.0088 decode.acc_seg: 99.5930 aux.loss_ce: 0.0081 aux.acc_seg: 98.8770 +04/16 20:01:34 - mmengine - INFO - Iter(train) [ 42600/160000] base_lr: 7.4070e-05 lr: 2.7385e-07 eta: 1 day, 8:32:15 time: 1.0010 data_time: 0.0045 memory: 8462 loss: 0.0182 decode.loss_ce: 0.0095 decode.acc_seg: 99.4852 aux.loss_ce: 0.0087 aux.acc_seg: 98.8741 +04/16 20:02:24 - mmengine - INFO - Iter(train) [ 42650/160000] base_lr: 7.4038e-05 lr: 2.7373e-07 eta: 1 day, 8:31:25 time: 1.0007 data_time: 0.0044 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0076 decode.acc_seg: 99.5325 aux.loss_ce: 0.0078 aux.acc_seg: 99.0631 +04/16 20:03:14 - mmengine - INFO - Iter(train) [ 42700/160000] base_lr: 7.4007e-05 lr: 2.7362e-07 eta: 1 day, 8:30:36 time: 1.0007 data_time: 0.0043 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0088 decode.acc_seg: 99.5106 aux.loss_ce: 0.0088 aux.acc_seg: 99.1570 +04/16 20:04:04 - mmengine - INFO - Iter(train) [ 42750/160000] base_lr: 7.3975e-05 lr: 2.7350e-07 eta: 1 day, 8:29:46 time: 0.9996 data_time: 0.0045 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0087 decode.acc_seg: 99.6370 aux.loss_ce: 0.0089 aux.acc_seg: 99.0051 +04/16 20:04:54 - mmengine - INFO - Iter(train) [ 42800/160000] base_lr: 7.3944e-05 lr: 2.7339e-07 eta: 1 day, 8:28:57 time: 1.0002 data_time: 0.0043 memory: 8462 loss: 0.0187 decode.loss_ce: 0.0096 decode.acc_seg: 99.5687 aux.loss_ce: 0.0090 aux.acc_seg: 98.9897 +04/16 20:05:44 - mmengine - INFO - Iter(train) [ 42850/160000] base_lr: 7.3912e-05 lr: 2.7327e-07 eta: 1 day, 8:28:07 time: 1.0002 data_time: 0.0042 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0088 decode.acc_seg: 99.6162 aux.loss_ce: 0.0088 aux.acc_seg: 99.2292 +04/16 20:06:34 - mmengine - INFO - Iter(train) [ 42900/160000] base_lr: 7.3881e-05 lr: 2.7315e-07 eta: 1 day, 8:27:17 time: 1.0003 data_time: 0.0046 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0071 decode.acc_seg: 99.5518 aux.loss_ce: 0.0079 aux.acc_seg: 99.0183 +04/16 20:07:24 - mmengine - INFO - Iter(train) [ 42950/160000] base_lr: 7.3849e-05 lr: 2.7304e-07 eta: 1 day, 8:26:28 time: 1.0000 data_time: 0.0045 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0086 decode.acc_seg: 99.6401 aux.loss_ce: 0.0089 aux.acc_seg: 99.1362 +04/16 20:08:14 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 20:08:14 - mmengine - INFO - Iter(train) [ 43000/160000] base_lr: 7.3818e-05 lr: 2.7292e-07 eta: 1 day, 8:25:38 time: 1.0019 data_time: 0.0042 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0089 decode.acc_seg: 99.7192 aux.loss_ce: 0.0088 aux.acc_seg: 99.4164 +04/16 20:09:04 - mmengine - INFO - Iter(train) [ 43050/160000] base_lr: 7.3786e-05 lr: 2.7280e-07 eta: 1 day, 8:24:49 time: 1.0001 data_time: 0.0044 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0090 decode.acc_seg: 99.5909 aux.loss_ce: 0.0090 aux.acc_seg: 99.0608 +04/16 20:09:54 - mmengine - INFO - Iter(train) [ 43100/160000] base_lr: 7.3754e-05 lr: 2.7269e-07 eta: 1 day, 8:23:59 time: 1.0013 data_time: 0.0042 memory: 8462 loss: 0.0197 decode.loss_ce: 0.0109 decode.acc_seg: 99.6798 aux.loss_ce: 0.0089 aux.acc_seg: 99.3006 +04/16 20:10:44 - mmengine - INFO - Iter(train) [ 43150/160000] base_lr: 7.3723e-05 lr: 2.7257e-07 eta: 1 day, 8:23:10 time: 1.0012 data_time: 0.0044 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0087 decode.acc_seg: 99.5703 aux.loss_ce: 0.0087 aux.acc_seg: 99.1564 +04/16 20:11:34 - mmengine - INFO - Iter(train) [ 43200/160000] base_lr: 7.3691e-05 lr: 2.7245e-07 eta: 1 day, 8:22:20 time: 0.9997 data_time: 0.0041 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0075 decode.acc_seg: 99.6157 aux.loss_ce: 0.0074 aux.acc_seg: 99.0807 +04/16 20:12:24 - mmengine - INFO - Iter(train) [ 43250/160000] base_lr: 7.3660e-05 lr: 2.7234e-07 eta: 1 day, 8:21:31 time: 1.0002 data_time: 0.0045 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0091 decode.acc_seg: 99.7007 aux.loss_ce: 0.0084 aux.acc_seg: 99.2489 +04/16 20:13:14 - mmengine - INFO - Iter(train) [ 43300/160000] base_lr: 7.3628e-05 lr: 2.7222e-07 eta: 1 day, 8:20:41 time: 0.9998 data_time: 0.0045 memory: 8462 loss: 0.0189 decode.loss_ce: 0.0094 decode.acc_seg: 99.6492 aux.loss_ce: 0.0095 aux.acc_seg: 99.1577 +04/16 20:14:04 - mmengine - INFO - Iter(train) [ 43350/160000] base_lr: 7.3597e-05 lr: 2.7210e-07 eta: 1 day, 8:19:51 time: 0.9996 data_time: 0.0044 memory: 8462 loss: 0.0187 decode.loss_ce: 0.0101 decode.acc_seg: 99.6384 aux.loss_ce: 0.0086 aux.acc_seg: 99.2035 +04/16 20:14:54 - mmengine - INFO - Iter(train) [ 43400/160000] base_lr: 7.3565e-05 lr: 2.7199e-07 eta: 1 day, 8:19:02 time: 0.9990 data_time: 0.0044 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0088 decode.acc_seg: 99.6355 aux.loss_ce: 0.0084 aux.acc_seg: 99.0963 +04/16 20:15:44 - mmengine - INFO - Iter(train) [ 43450/160000] base_lr: 7.3534e-05 lr: 2.7187e-07 eta: 1 day, 8:18:12 time: 1.0000 data_time: 0.0044 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0082 decode.acc_seg: 99.6965 aux.loss_ce: 0.0086 aux.acc_seg: 98.9571 +04/16 20:16:34 - mmengine - INFO - Iter(train) [ 43500/160000] base_lr: 7.3502e-05 lr: 2.7175e-07 eta: 1 day, 8:17:22 time: 1.0007 data_time: 0.0042 memory: 8462 loss: 0.0196 decode.loss_ce: 0.0099 decode.acc_seg: 99.6550 aux.loss_ce: 0.0096 aux.acc_seg: 99.1711 +04/16 20:17:24 - mmengine - INFO - Iter(train) [ 43550/160000] base_lr: 7.3470e-05 lr: 2.7164e-07 eta: 1 day, 8:16:33 time: 0.9997 data_time: 0.0045 memory: 8462 loss: 0.0195 decode.loss_ce: 0.0102 decode.acc_seg: 99.7112 aux.loss_ce: 0.0094 aux.acc_seg: 99.1861 +04/16 20:18:14 - mmengine - INFO - Iter(train) [ 43600/160000] base_lr: 7.3439e-05 lr: 2.7152e-07 eta: 1 day, 8:15:43 time: 0.9999 data_time: 0.0045 memory: 8462 loss: 0.0194 decode.loss_ce: 0.0096 decode.acc_seg: 99.5728 aux.loss_ce: 0.0098 aux.acc_seg: 98.7864 +04/16 20:19:04 - mmengine - INFO - Iter(train) [ 43650/160000] base_lr: 7.3407e-05 lr: 2.7140e-07 eta: 1 day, 8:14:54 time: 0.9997 data_time: 0.0046 memory: 8462 loss: 0.0164 decode.loss_ce: 0.0081 decode.acc_seg: 99.5665 aux.loss_ce: 0.0083 aux.acc_seg: 99.0601 +04/16 20:19:54 - mmengine - INFO - Iter(train) [ 43700/160000] base_lr: 7.3376e-05 lr: 2.7129e-07 eta: 1 day, 8:14:04 time: 0.9995 data_time: 0.0045 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0084 decode.acc_seg: 99.6069 aux.loss_ce: 0.0091 aux.acc_seg: 99.0421 +04/16 20:20:44 - mmengine - INFO - Iter(train) [ 43750/160000] base_lr: 7.3344e-05 lr: 2.7117e-07 eta: 1 day, 8:13:14 time: 1.0006 data_time: 0.0049 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0088 decode.acc_seg: 99.4873 aux.loss_ce: 0.0087 aux.acc_seg: 98.5909 +04/16 20:21:34 - mmengine - INFO - Iter(train) [ 43800/160000] base_lr: 7.3313e-05 lr: 2.7105e-07 eta: 1 day, 8:12:25 time: 1.0010 data_time: 0.0056 memory: 8462 loss: 0.0164 decode.loss_ce: 0.0079 decode.acc_seg: 99.7057 aux.loss_ce: 0.0085 aux.acc_seg: 99.2668 +04/16 20:22:24 - mmengine - INFO - Iter(train) [ 43850/160000] base_lr: 7.3281e-05 lr: 2.7094e-07 eta: 1 day, 8:11:35 time: 0.9998 data_time: 0.0046 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0090 decode.acc_seg: 99.7219 aux.loss_ce: 0.0086 aux.acc_seg: 99.3162 +04/16 20:23:14 - mmengine - INFO - Iter(train) [ 43900/160000] base_lr: 7.3250e-05 lr: 2.7082e-07 eta: 1 day, 8:10:46 time: 1.0003 data_time: 0.0047 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0078 decode.acc_seg: 99.7856 aux.loss_ce: 0.0076 aux.acc_seg: 99.4522 +04/16 20:24:04 - mmengine - INFO - Iter(train) [ 43950/160000] base_lr: 7.3218e-05 lr: 2.7070e-07 eta: 1 day, 8:09:56 time: 1.0004 data_time: 0.0050 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0087 decode.acc_seg: 99.7627 aux.loss_ce: 0.0081 aux.acc_seg: 99.3587 +04/16 20:24:54 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 20:24:54 - mmengine - INFO - Iter(train) [ 44000/160000] base_lr: 7.3187e-05 lr: 2.7059e-07 eta: 1 day, 8:09:07 time: 0.9995 data_time: 0.0042 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0086 decode.acc_seg: 99.6723 aux.loss_ce: 0.0090 aux.acc_seg: 99.0702 +04/16 20:25:44 - mmengine - INFO - Iter(train) [ 44050/160000] base_lr: 7.3155e-05 lr: 2.7047e-07 eta: 1 day, 8:08:17 time: 1.0002 data_time: 0.0045 memory: 8462 loss: 0.0169 decode.loss_ce: 0.0086 decode.acc_seg: 99.6407 aux.loss_ce: 0.0083 aux.acc_seg: 99.0522 +04/16 20:26:34 - mmengine - INFO - Iter(train) [ 44100/160000] base_lr: 7.3123e-05 lr: 2.7035e-07 eta: 1 day, 8:07:28 time: 1.0014 data_time: 0.0051 memory: 8462 loss: 0.0191 decode.loss_ce: 0.0093 decode.acc_seg: 99.6677 aux.loss_ce: 0.0098 aux.acc_seg: 99.2271 +04/16 20:27:24 - mmengine - INFO - Iter(train) [ 44150/160000] base_lr: 7.3092e-05 lr: 2.7024e-07 eta: 1 day, 8:06:38 time: 1.0008 data_time: 0.0044 memory: 8462 loss: 0.0169 decode.loss_ce: 0.0084 decode.acc_seg: 99.6956 aux.loss_ce: 0.0085 aux.acc_seg: 99.1930 +04/16 20:28:14 - mmengine - INFO - Iter(train) [ 44200/160000] base_lr: 7.3060e-05 lr: 2.7012e-07 eta: 1 day, 8:05:48 time: 1.0001 data_time: 0.0045 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0085 decode.acc_seg: 99.7025 aux.loss_ce: 0.0092 aux.acc_seg: 99.0425 +04/16 20:29:04 - mmengine - INFO - Iter(train) [ 44250/160000] base_lr: 7.3029e-05 lr: 2.7000e-07 eta: 1 day, 8:04:59 time: 0.9991 data_time: 0.0045 memory: 8462 loss: 0.0195 decode.loss_ce: 0.0098 decode.acc_seg: 99.6687 aux.loss_ce: 0.0096 aux.acc_seg: 99.1934 +04/16 20:29:54 - mmengine - INFO - Iter(train) [ 44300/160000] base_lr: 7.2997e-05 lr: 2.6989e-07 eta: 1 day, 8:04:09 time: 1.0002 data_time: 0.0046 memory: 8462 loss: 0.0198 decode.loss_ce: 0.0101 decode.acc_seg: 99.6441 aux.loss_ce: 0.0097 aux.acc_seg: 99.1291 +04/16 20:30:44 - mmengine - INFO - Iter(train) [ 44350/160000] base_lr: 7.2966e-05 lr: 2.6977e-07 eta: 1 day, 8:03:20 time: 0.9993 data_time: 0.0052 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0081 decode.acc_seg: 99.7873 aux.loss_ce: 0.0080 aux.acc_seg: 99.3408 +04/16 20:31:34 - mmengine - INFO - Iter(train) [ 44400/160000] base_lr: 7.2934e-05 lr: 2.6965e-07 eta: 1 day, 8:02:30 time: 1.0005 data_time: 0.0049 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0087 decode.acc_seg: 99.6256 aux.loss_ce: 0.0085 aux.acc_seg: 99.0400 +04/16 20:32:24 - mmengine - INFO - Iter(train) [ 44450/160000] base_lr: 7.2903e-05 lr: 2.6954e-07 eta: 1 day, 8:01:40 time: 0.9996 data_time: 0.0047 memory: 8462 loss: 0.0184 decode.loss_ce: 0.0094 decode.acc_seg: 99.6099 aux.loss_ce: 0.0091 aux.acc_seg: 99.1119 +04/16 20:33:14 - mmengine - INFO - Iter(train) [ 44500/160000] base_lr: 7.2871e-05 lr: 2.6942e-07 eta: 1 day, 8:00:51 time: 1.0001 data_time: 0.0046 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0090 decode.acc_seg: 99.4818 aux.loss_ce: 0.0082 aux.acc_seg: 98.8588 +04/16 20:34:04 - mmengine - INFO - Iter(train) [ 44550/160000] base_lr: 7.2840e-05 lr: 2.6930e-07 eta: 1 day, 8:00:01 time: 0.9996 data_time: 0.0045 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0092 decode.acc_seg: 99.5657 aux.loss_ce: 0.0084 aux.acc_seg: 99.1383 +04/16 20:34:54 - mmengine - INFO - Iter(train) [ 44600/160000] base_lr: 7.2808e-05 lr: 2.6919e-07 eta: 1 day, 7:59:12 time: 0.9999 data_time: 0.0046 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0084 decode.acc_seg: 99.5745 aux.loss_ce: 0.0089 aux.acc_seg: 98.9307 +04/16 20:35:44 - mmengine - INFO - Iter(train) [ 44650/160000] base_lr: 7.2776e-05 lr: 2.6907e-07 eta: 1 day, 7:58:22 time: 0.9989 data_time: 0.0045 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0076 decode.acc_seg: 99.6227 aux.loss_ce: 0.0081 aux.acc_seg: 98.8358 +04/16 20:36:34 - mmengine - INFO - Iter(train) [ 44700/160000] base_lr: 7.2745e-05 lr: 2.6895e-07 eta: 1 day, 7:57:32 time: 0.9996 data_time: 0.0043 memory: 8462 loss: 0.0191 decode.loss_ce: 0.0105 decode.acc_seg: 99.0187 aux.loss_ce: 0.0085 aux.acc_seg: 98.4308 +04/16 20:37:24 - mmengine - INFO - Iter(train) [ 44750/160000] base_lr: 7.2713e-05 lr: 2.6884e-07 eta: 1 day, 7:56:43 time: 1.0005 data_time: 0.0045 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0088 decode.acc_seg: 99.5255 aux.loss_ce: 0.0082 aux.acc_seg: 98.7513 +04/16 20:38:14 - mmengine - INFO - Iter(train) [ 44800/160000] base_lr: 7.2682e-05 lr: 2.6872e-07 eta: 1 day, 7:55:53 time: 1.0002 data_time: 0.0050 memory: 8462 loss: 0.0185 decode.loss_ce: 0.0096 decode.acc_seg: 99.6208 aux.loss_ce: 0.0089 aux.acc_seg: 99.0129 +04/16 20:39:04 - mmengine - INFO - Iter(train) [ 44850/160000] base_lr: 7.2650e-05 lr: 2.6860e-07 eta: 1 day, 7:55:04 time: 1.0005 data_time: 0.0048 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0086 decode.acc_seg: 99.7356 aux.loss_ce: 0.0095 aux.acc_seg: 99.1257 +04/16 20:39:54 - mmengine - INFO - Iter(train) [ 44900/160000] base_lr: 7.2619e-05 lr: 2.6849e-07 eta: 1 day, 7:54:14 time: 0.9996 data_time: 0.0043 memory: 8462 loss: 0.0208 decode.loss_ce: 0.0106 decode.acc_seg: 99.5712 aux.loss_ce: 0.0103 aux.acc_seg: 99.1388 +04/16 20:40:44 - mmengine - INFO - Iter(train) [ 44950/160000] base_lr: 7.2587e-05 lr: 2.6837e-07 eta: 1 day, 7:53:25 time: 1.0008 data_time: 0.0045 memory: 8462 loss: 0.0197 decode.loss_ce: 0.0100 decode.acc_seg: 99.7240 aux.loss_ce: 0.0097 aux.acc_seg: 99.2456 +04/16 20:41:34 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 20:41:34 - mmengine - INFO - Iter(train) [ 45000/160000] base_lr: 7.2556e-05 lr: 2.6825e-07 eta: 1 day, 7:52:35 time: 1.0009 data_time: 0.0043 memory: 8462 loss: 0.0185 decode.loss_ce: 0.0097 decode.acc_seg: 99.5369 aux.loss_ce: 0.0088 aux.acc_seg: 99.2157 +04/16 20:42:24 - mmengine - INFO - Iter(train) [ 45050/160000] base_lr: 7.2524e-05 lr: 2.6814e-07 eta: 1 day, 7:51:45 time: 0.9995 data_time: 0.0046 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0084 decode.acc_seg: 99.7570 aux.loss_ce: 0.0083 aux.acc_seg: 99.4604 +04/16 20:43:14 - mmengine - INFO - Iter(train) [ 45100/160000] base_lr: 7.2493e-05 lr: 2.6802e-07 eta: 1 day, 7:50:56 time: 1.0006 data_time: 0.0047 memory: 8462 loss: 0.0162 decode.loss_ce: 0.0082 decode.acc_seg: 99.6527 aux.loss_ce: 0.0080 aux.acc_seg: 99.4343 +04/16 20:44:04 - mmengine - INFO - Iter(train) [ 45150/160000] base_lr: 7.2461e-05 lr: 2.6790e-07 eta: 1 day, 7:50:06 time: 0.9982 data_time: 0.0044 memory: 8462 loss: 0.0151 decode.loss_ce: 0.0076 decode.acc_seg: 99.6601 aux.loss_ce: 0.0075 aux.acc_seg: 99.0047 +04/16 20:44:55 - mmengine - INFO - Iter(train) [ 45200/160000] base_lr: 7.2429e-05 lr: 2.6779e-07 eta: 1 day, 7:49:17 time: 1.0013 data_time: 0.0045 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0080 decode.acc_seg: 99.7080 aux.loss_ce: 0.0080 aux.acc_seg: 99.2273 +04/16 20:45:45 - mmengine - INFO - Iter(train) [ 45250/160000] base_lr: 7.2398e-05 lr: 2.6767e-07 eta: 1 day, 7:48:27 time: 0.9992 data_time: 0.0048 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0080 decode.acc_seg: 99.6346 aux.loss_ce: 0.0087 aux.acc_seg: 99.0280 +04/16 20:46:35 - mmengine - INFO - Iter(train) [ 45300/160000] base_lr: 7.2366e-05 lr: 2.6755e-07 eta: 1 day, 7:47:37 time: 1.0023 data_time: 0.0042 memory: 8462 loss: 0.0180 decode.loss_ce: 0.0094 decode.acc_seg: 99.6607 aux.loss_ce: 0.0085 aux.acc_seg: 98.9693 +04/16 20:47:25 - mmengine - INFO - Iter(train) [ 45350/160000] base_lr: 7.2335e-05 lr: 2.6744e-07 eta: 1 day, 7:46:48 time: 1.0015 data_time: 0.0047 memory: 8462 loss: 0.0185 decode.loss_ce: 0.0095 decode.acc_seg: 99.6508 aux.loss_ce: 0.0090 aux.acc_seg: 98.9176 +04/16 20:48:15 - mmengine - INFO - Iter(train) [ 45400/160000] base_lr: 7.2303e-05 lr: 2.6732e-07 eta: 1 day, 7:45:58 time: 1.0004 data_time: 0.0049 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0089 decode.acc_seg: 99.5935 aux.loss_ce: 0.0090 aux.acc_seg: 99.2535 +04/16 20:49:05 - mmengine - INFO - Iter(train) [ 45450/160000] base_lr: 7.2272e-05 lr: 2.6720e-07 eta: 1 day, 7:45:09 time: 1.0003 data_time: 0.0046 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0087 decode.acc_seg: 99.7452 aux.loss_ce: 0.0086 aux.acc_seg: 99.3311 +04/16 20:49:55 - mmengine - INFO - Iter(train) [ 45500/160000] base_lr: 7.2240e-05 lr: 2.6709e-07 eta: 1 day, 7:44:19 time: 1.0011 data_time: 0.0048 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0081 decode.acc_seg: 99.5834 aux.loss_ce: 0.0087 aux.acc_seg: 98.7307 +04/16 20:50:45 - mmengine - INFO - Iter(train) [ 45550/160000] base_lr: 7.2209e-05 lr: 2.6697e-07 eta: 1 day, 7:43:29 time: 1.0000 data_time: 0.0047 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0088 decode.acc_seg: 99.4583 aux.loss_ce: 0.0088 aux.acc_seg: 98.8039 +04/16 20:51:35 - mmengine - INFO - Iter(train) [ 45600/160000] base_lr: 7.2177e-05 lr: 2.6685e-07 eta: 1 day, 7:42:40 time: 1.0012 data_time: 0.0048 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0081 decode.acc_seg: 99.5573 aux.loss_ce: 0.0079 aux.acc_seg: 98.9632 +04/16 20:52:25 - mmengine - INFO - Iter(train) [ 45650/160000] base_lr: 7.2146e-05 lr: 2.6674e-07 eta: 1 day, 7:41:50 time: 0.9990 data_time: 0.0044 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0070 decode.acc_seg: 99.6689 aux.loss_ce: 0.0070 aux.acc_seg: 99.1955 +04/16 20:53:15 - mmengine - INFO - Iter(train) [ 45700/160000] base_lr: 7.2114e-05 lr: 2.6662e-07 eta: 1 day, 7:41:00 time: 1.0007 data_time: 0.0044 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0091 decode.acc_seg: 99.7379 aux.loss_ce: 0.0099 aux.acc_seg: 99.1184 +04/16 20:54:05 - mmengine - INFO - Iter(train) [ 45750/160000] base_lr: 7.2082e-05 lr: 2.6650e-07 eta: 1 day, 7:40:11 time: 0.9990 data_time: 0.0048 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0076 decode.acc_seg: 99.7192 aux.loss_ce: 0.0073 aux.acc_seg: 99.5207 +04/16 20:54:55 - mmengine - INFO - Iter(train) [ 45800/160000] base_lr: 7.2051e-05 lr: 2.6639e-07 eta: 1 day, 7:39:21 time: 0.9999 data_time: 0.0042 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0076 decode.acc_seg: 99.6199 aux.loss_ce: 0.0076 aux.acc_seg: 99.3338 +04/16 20:55:45 - mmengine - INFO - Iter(train) [ 45850/160000] base_lr: 7.2019e-05 lr: 2.6627e-07 eta: 1 day, 7:38:32 time: 0.9996 data_time: 0.0047 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0083 decode.acc_seg: 99.6424 aux.loss_ce: 0.0089 aux.acc_seg: 98.9332 +04/16 20:56:35 - mmengine - INFO - Iter(train) [ 45900/160000] base_lr: 7.1988e-05 lr: 2.6615e-07 eta: 1 day, 7:37:42 time: 1.0008 data_time: 0.0050 memory: 8462 loss: 0.0171 decode.loss_ce: 0.0084 decode.acc_seg: 99.5726 aux.loss_ce: 0.0087 aux.acc_seg: 98.9002 +04/16 20:57:25 - mmengine - INFO - Iter(train) [ 45950/160000] base_lr: 7.1956e-05 lr: 2.6604e-07 eta: 1 day, 7:36:52 time: 1.0001 data_time: 0.0049 memory: 8462 loss: 0.0163 decode.loss_ce: 0.0081 decode.acc_seg: 99.4663 aux.loss_ce: 0.0082 aux.acc_seg: 98.8310 +04/16 20:58:15 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 20:58:15 - mmengine - INFO - Iter(train) [ 46000/160000] base_lr: 7.1925e-05 lr: 2.6592e-07 eta: 1 day, 7:36:03 time: 0.9997 data_time: 0.0047 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0082 decode.acc_seg: 99.7248 aux.loss_ce: 0.0086 aux.acc_seg: 99.3486 +04/16 20:59:05 - mmengine - INFO - Iter(train) [ 46050/160000] base_lr: 7.1893e-05 lr: 2.6580e-07 eta: 1 day, 7:35:13 time: 0.9997 data_time: 0.0044 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0087 decode.acc_seg: 99.6130 aux.loss_ce: 0.0080 aux.acc_seg: 99.0934 +04/16 20:59:55 - mmengine - INFO - Iter(train) [ 46100/160000] base_lr: 7.1862e-05 lr: 2.6569e-07 eta: 1 day, 7:34:23 time: 0.9987 data_time: 0.0047 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0082 decode.acc_seg: 99.7351 aux.loss_ce: 0.0085 aux.acc_seg: 99.1802 +04/16 21:00:45 - mmengine - INFO - Iter(train) [ 46150/160000] base_lr: 7.1830e-05 lr: 2.6557e-07 eta: 1 day, 7:33:34 time: 0.9996 data_time: 0.0055 memory: 8462 loss: 0.0162 decode.loss_ce: 0.0083 decode.acc_seg: 99.5489 aux.loss_ce: 0.0079 aux.acc_seg: 99.2378 +04/16 21:01:35 - mmengine - INFO - Iter(train) [ 46200/160000] base_lr: 7.1799e-05 lr: 2.6545e-07 eta: 1 day, 7:32:44 time: 1.0006 data_time: 0.0049 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0090 decode.acc_seg: 99.5787 aux.loss_ce: 0.0087 aux.acc_seg: 99.0063 +04/16 21:02:25 - mmengine - INFO - Iter(train) [ 46250/160000] base_lr: 7.1767e-05 lr: 2.6534e-07 eta: 1 day, 7:31:54 time: 0.9997 data_time: 0.0045 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0077 decode.acc_seg: 99.5375 aux.loss_ce: 0.0079 aux.acc_seg: 99.3679 +04/16 21:03:15 - mmengine - INFO - Iter(train) [ 46300/160000] base_lr: 7.1735e-05 lr: 2.6522e-07 eta: 1 day, 7:31:05 time: 1.0023 data_time: 0.0043 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0095 decode.acc_seg: 99.7063 aux.loss_ce: 0.0096 aux.acc_seg: 99.1468 +04/16 21:04:05 - mmengine - INFO - Iter(train) [ 46350/160000] base_lr: 7.1704e-05 lr: 2.6510e-07 eta: 1 day, 7:30:15 time: 0.9996 data_time: 0.0045 memory: 8462 loss: 0.0183 decode.loss_ce: 0.0096 decode.acc_seg: 99.3879 aux.loss_ce: 0.0087 aux.acc_seg: 98.9149 +04/16 21:04:55 - mmengine - INFO - Iter(train) [ 46400/160000] base_lr: 7.1672e-05 lr: 2.6499e-07 eta: 1 day, 7:29:25 time: 1.0006 data_time: 0.0050 memory: 8462 loss: 0.0180 decode.loss_ce: 0.0085 decode.acc_seg: 99.7248 aux.loss_ce: 0.0095 aux.acc_seg: 98.8390 +04/16 21:05:45 - mmengine - INFO - Iter(train) [ 46450/160000] base_lr: 7.1641e-05 lr: 2.6487e-07 eta: 1 day, 7:28:36 time: 1.0003 data_time: 0.0048 memory: 8462 loss: 0.0142 decode.loss_ce: 0.0070 decode.acc_seg: 99.7429 aux.loss_ce: 0.0072 aux.acc_seg: 99.1627 +04/16 21:06:35 - mmengine - INFO - Iter(train) [ 46500/160000] base_lr: 7.1609e-05 lr: 2.6475e-07 eta: 1 day, 7:27:46 time: 0.9993 data_time: 0.0044 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0094 decode.acc_seg: 99.6775 aux.loss_ce: 0.0087 aux.acc_seg: 99.2115 +04/16 21:07:25 - mmengine - INFO - Iter(train) [ 46550/160000] base_lr: 7.1578e-05 lr: 2.6464e-07 eta: 1 day, 7:26:57 time: 1.0005 data_time: 0.0048 memory: 8462 loss: 0.0185 decode.loss_ce: 0.0094 decode.acc_seg: 99.7189 aux.loss_ce: 0.0091 aux.acc_seg: 99.2790 +04/16 21:08:15 - mmengine - INFO - Iter(train) [ 46600/160000] base_lr: 7.1546e-05 lr: 2.6452e-07 eta: 1 day, 7:26:07 time: 0.9988 data_time: 0.0049 memory: 8462 loss: 0.0184 decode.loss_ce: 0.0088 decode.acc_seg: 99.6393 aux.loss_ce: 0.0096 aux.acc_seg: 98.8394 +04/16 21:09:05 - mmengine - INFO - Iter(train) [ 46650/160000] base_lr: 7.1515e-05 lr: 2.6440e-07 eta: 1 day, 7:25:17 time: 0.9993 data_time: 0.0048 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0077 decode.acc_seg: 99.6086 aux.loss_ce: 0.0079 aux.acc_seg: 99.1339 +04/16 21:09:55 - mmengine - INFO - Iter(train) [ 46700/160000] base_lr: 7.1483e-05 lr: 2.6429e-07 eta: 1 day, 7:24:27 time: 0.9990 data_time: 0.0049 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0098 decode.acc_seg: 99.6712 aux.loss_ce: 0.0091 aux.acc_seg: 98.9929 +04/16 21:10:45 - mmengine - INFO - Iter(train) [ 46750/160000] base_lr: 7.1452e-05 lr: 2.6417e-07 eta: 1 day, 7:23:38 time: 1.0000 data_time: 0.0046 memory: 8462 loss: 0.0171 decode.loss_ce: 0.0085 decode.acc_seg: 99.5754 aux.loss_ce: 0.0086 aux.acc_seg: 99.1449 +04/16 21:11:35 - mmengine - INFO - Iter(train) [ 46800/160000] base_lr: 7.1420e-05 lr: 2.6405e-07 eta: 1 day, 7:22:48 time: 0.9990 data_time: 0.0046 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0090 decode.acc_seg: 99.5483 aux.loss_ce: 0.0089 aux.acc_seg: 98.8108 +04/16 21:12:25 - mmengine - INFO - Iter(train) [ 46850/160000] base_lr: 7.1388e-05 lr: 2.6394e-07 eta: 1 day, 7:21:58 time: 1.0011 data_time: 0.0049 memory: 8462 loss: 0.0163 decode.loss_ce: 0.0081 decode.acc_seg: 99.5256 aux.loss_ce: 0.0081 aux.acc_seg: 98.9433 +04/16 21:13:15 - mmengine - INFO - Iter(train) [ 46900/160000] base_lr: 7.1357e-05 lr: 2.6382e-07 eta: 1 day, 7:21:09 time: 1.0004 data_time: 0.0045 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0082 decode.acc_seg: 99.6595 aux.loss_ce: 0.0084 aux.acc_seg: 99.3092 +04/16 21:14:05 - mmengine - INFO - Iter(train) [ 46950/160000] base_lr: 7.1325e-05 lr: 2.6370e-07 eta: 1 day, 7:20:19 time: 1.0002 data_time: 0.0043 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0072 decode.acc_seg: 99.7730 aux.loss_ce: 0.0072 aux.acc_seg: 99.4516 +04/16 21:14:55 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 21:14:55 - mmengine - INFO - Iter(train) [ 47000/160000] base_lr: 7.1294e-05 lr: 2.6359e-07 eta: 1 day, 7:19:29 time: 0.9986 data_time: 0.0044 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0083 decode.acc_seg: 99.6246 aux.loss_ce: 0.0085 aux.acc_seg: 99.0425 +04/16 21:15:45 - mmengine - INFO - Iter(train) [ 47050/160000] base_lr: 7.1262e-05 lr: 2.6347e-07 eta: 1 day, 7:18:40 time: 1.0002 data_time: 0.0044 memory: 8462 loss: 0.0186 decode.loss_ce: 0.0095 decode.acc_seg: 99.6002 aux.loss_ce: 0.0091 aux.acc_seg: 99.0852 +04/16 21:16:35 - mmengine - INFO - Iter(train) [ 47100/160000] base_lr: 7.1231e-05 lr: 2.6335e-07 eta: 1 day, 7:17:50 time: 1.0004 data_time: 0.0042 memory: 8462 loss: 0.0180 decode.loss_ce: 0.0090 decode.acc_seg: 99.5827 aux.loss_ce: 0.0089 aux.acc_seg: 98.9855 +04/16 21:17:25 - mmengine - INFO - Iter(train) [ 47150/160000] base_lr: 7.1199e-05 lr: 2.6324e-07 eta: 1 day, 7:17:01 time: 0.9993 data_time: 0.0052 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0085 decode.acc_seg: 99.4734 aux.loss_ce: 0.0090 aux.acc_seg: 98.6307 +04/16 21:18:15 - mmengine - INFO - Iter(train) [ 47200/160000] base_lr: 7.1168e-05 lr: 2.6312e-07 eta: 1 day, 7:16:11 time: 1.0009 data_time: 0.0048 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0085 decode.acc_seg: 99.5968 aux.loss_ce: 0.0085 aux.acc_seg: 99.2611 +04/16 21:19:05 - mmengine - INFO - Iter(train) [ 47250/160000] base_lr: 7.1136e-05 lr: 2.6300e-07 eta: 1 day, 7:15:21 time: 0.9998 data_time: 0.0042 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0080 decode.acc_seg: 99.7351 aux.loss_ce: 0.0081 aux.acc_seg: 99.2352 +04/16 21:19:55 - mmengine - INFO - Iter(train) [ 47300/160000] base_lr: 7.1105e-05 lr: 2.6289e-07 eta: 1 day, 7:14:32 time: 1.0009 data_time: 0.0047 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0090 decode.acc_seg: 99.4915 aux.loss_ce: 0.0089 aux.acc_seg: 98.8310 +04/16 21:20:45 - mmengine - INFO - Iter(train) [ 47350/160000] base_lr: 7.1073e-05 lr: 2.6277e-07 eta: 1 day, 7:13:42 time: 0.9995 data_time: 0.0046 memory: 8462 loss: 0.0163 decode.loss_ce: 0.0083 decode.acc_seg: 99.5287 aux.loss_ce: 0.0080 aux.acc_seg: 98.7986 +04/16 21:21:35 - mmengine - INFO - Iter(train) [ 47400/160000] base_lr: 7.1041e-05 lr: 2.6265e-07 eta: 1 day, 7:12:52 time: 0.9991 data_time: 0.0046 memory: 8462 loss: 0.0163 decode.loss_ce: 0.0078 decode.acc_seg: 99.7190 aux.loss_ce: 0.0085 aux.acc_seg: 99.3147 +04/16 21:22:25 - mmengine - INFO - Iter(train) [ 47450/160000] base_lr: 7.1010e-05 lr: 2.6254e-07 eta: 1 day, 7:12:03 time: 0.9995 data_time: 0.0044 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0079 decode.acc_seg: 99.6927 aux.loss_ce: 0.0079 aux.acc_seg: 99.1320 +04/16 21:23:15 - mmengine - INFO - Iter(train) [ 47500/160000] base_lr: 7.0978e-05 lr: 2.6242e-07 eta: 1 day, 7:11:13 time: 1.0008 data_time: 0.0046 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0077 decode.acc_seg: 99.8337 aux.loss_ce: 0.0083 aux.acc_seg: 99.4114 +04/16 21:24:05 - mmengine - INFO - Iter(train) [ 47550/160000] base_lr: 7.0947e-05 lr: 2.6231e-07 eta: 1 day, 7:10:24 time: 0.9997 data_time: 0.0046 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0077 decode.acc_seg: 99.6178 aux.loss_ce: 0.0080 aux.acc_seg: 98.8905 +04/16 21:24:55 - mmengine - INFO - Iter(train) [ 47600/160000] base_lr: 7.0915e-05 lr: 2.6219e-07 eta: 1 day, 7:09:34 time: 0.9997 data_time: 0.0045 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0071 decode.acc_seg: 99.7498 aux.loss_ce: 0.0076 aux.acc_seg: 99.4558 +04/16 21:25:45 - mmengine - INFO - Iter(train) [ 47650/160000] base_lr: 7.0884e-05 lr: 2.6207e-07 eta: 1 day, 7:08:44 time: 0.9998 data_time: 0.0046 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0084 decode.acc_seg: 99.6456 aux.loss_ce: 0.0084 aux.acc_seg: 98.7534 +04/16 21:26:35 - mmengine - INFO - Iter(train) [ 47700/160000] base_lr: 7.0852e-05 lr: 2.6196e-07 eta: 1 day, 7:07:54 time: 0.9998 data_time: 0.0044 memory: 8462 loss: 0.0159 decode.loss_ce: 0.0079 decode.acc_seg: 99.7168 aux.loss_ce: 0.0080 aux.acc_seg: 99.2168 +04/16 21:27:25 - mmengine - INFO - Iter(train) [ 47750/160000] base_lr: 7.0821e-05 lr: 2.6184e-07 eta: 1 day, 7:07:05 time: 0.9984 data_time: 0.0043 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0076 decode.acc_seg: 99.7892 aux.loss_ce: 0.0080 aux.acc_seg: 99.4446 +04/16 21:28:15 - mmengine - INFO - Iter(train) [ 47800/160000] base_lr: 7.0789e-05 lr: 2.6172e-07 eta: 1 day, 7:06:15 time: 0.9998 data_time: 0.0049 memory: 8462 loss: 0.0159 decode.loss_ce: 0.0080 decode.acc_seg: 99.7206 aux.loss_ce: 0.0079 aux.acc_seg: 99.1978 +04/16 21:29:05 - mmengine - INFO - Iter(train) [ 47850/160000] base_lr: 7.0758e-05 lr: 2.6161e-07 eta: 1 day, 7:05:25 time: 1.0007 data_time: 0.0047 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0093 decode.acc_seg: 99.6248 aux.loss_ce: 0.0097 aux.acc_seg: 98.7993 +04/16 21:29:55 - mmengine - INFO - Iter(train) [ 47900/160000] base_lr: 7.0726e-05 lr: 2.6149e-07 eta: 1 day, 7:04:36 time: 0.9998 data_time: 0.0044 memory: 8462 loss: 0.0151 decode.loss_ce: 0.0072 decode.acc_seg: 99.6380 aux.loss_ce: 0.0078 aux.acc_seg: 98.9882 +04/16 21:30:45 - mmengine - INFO - Iter(train) [ 47950/160000] base_lr: 7.0694e-05 lr: 2.6137e-07 eta: 1 day, 7:03:46 time: 0.9992 data_time: 0.0047 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0076 decode.acc_seg: 99.6914 aux.loss_ce: 0.0078 aux.acc_seg: 99.1798 +04/16 21:31:35 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 21:31:35 - mmengine - INFO - Iter(train) [ 48000/160000] base_lr: 7.0663e-05 lr: 2.6126e-07 eta: 1 day, 7:02:56 time: 0.9998 data_time: 0.0044 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0098 decode.acc_seg: 99.6243 aux.loss_ce: 0.0091 aux.acc_seg: 99.0707 +04/16 21:32:25 - mmengine - INFO - Iter(train) [ 48050/160000] base_lr: 7.0631e-05 lr: 2.6114e-07 eta: 1 day, 7:02:07 time: 0.9993 data_time: 0.0046 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0092 decode.acc_seg: 99.6897 aux.loss_ce: 0.0089 aux.acc_seg: 99.1552 +04/16 21:33:15 - mmengine - INFO - Iter(train) [ 48100/160000] base_lr: 7.0600e-05 lr: 2.6102e-07 eta: 1 day, 7:01:17 time: 0.9995 data_time: 0.0045 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0084 decode.acc_seg: 99.6603 aux.loss_ce: 0.0084 aux.acc_seg: 99.1549 +04/16 21:34:05 - mmengine - INFO - Iter(train) [ 48150/160000] base_lr: 7.0568e-05 lr: 2.6091e-07 eta: 1 day, 7:00:27 time: 0.9995 data_time: 0.0047 memory: 8462 loss: 0.0203 decode.loss_ce: 0.0103 decode.acc_seg: 99.4989 aux.loss_ce: 0.0101 aux.acc_seg: 98.7778 +04/16 21:34:55 - mmengine - INFO - Iter(train) [ 48200/160000] base_lr: 7.0537e-05 lr: 2.6079e-07 eta: 1 day, 6:59:37 time: 0.9996 data_time: 0.0046 memory: 8462 loss: 0.0189 decode.loss_ce: 0.0098 decode.acc_seg: 99.7543 aux.loss_ce: 0.0091 aux.acc_seg: 99.4373 +04/16 21:35:45 - mmengine - INFO - Iter(train) [ 48250/160000] base_lr: 7.0505e-05 lr: 2.6067e-07 eta: 1 day, 6:58:48 time: 0.9992 data_time: 0.0051 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0081 decode.acc_seg: 99.6651 aux.loss_ce: 0.0086 aux.acc_seg: 99.2311 +04/16 21:36:35 - mmengine - INFO - Iter(train) [ 48300/160000] base_lr: 7.0474e-05 lr: 2.6056e-07 eta: 1 day, 6:57:58 time: 0.9994 data_time: 0.0043 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0076 decode.acc_seg: 99.6485 aux.loss_ce: 0.0078 aux.acc_seg: 99.0545 +04/16 21:37:25 - mmengine - INFO - Iter(train) [ 48350/160000] base_lr: 7.0442e-05 lr: 2.6044e-07 eta: 1 day, 6:57:08 time: 1.0002 data_time: 0.0043 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0079 decode.acc_seg: 99.7498 aux.loss_ce: 0.0078 aux.acc_seg: 99.1947 +04/16 21:38:15 - mmengine - INFO - Iter(train) [ 48400/160000] base_lr: 7.0411e-05 lr: 2.6032e-07 eta: 1 day, 6:56:19 time: 1.0002 data_time: 0.0047 memory: 8462 loss: 0.0184 decode.loss_ce: 0.0091 decode.acc_seg: 99.5787 aux.loss_ce: 0.0094 aux.acc_seg: 98.7988 +04/16 21:39:05 - mmengine - INFO - Iter(train) [ 48450/160000] base_lr: 7.0379e-05 lr: 2.6021e-07 eta: 1 day, 6:55:29 time: 1.0026 data_time: 0.0048 memory: 8462 loss: 0.0159 decode.loss_ce: 0.0080 decode.acc_seg: 99.7229 aux.loss_ce: 0.0078 aux.acc_seg: 99.3111 +04/16 21:39:55 - mmengine - INFO - Iter(train) [ 48500/160000] base_lr: 7.0347e-05 lr: 2.6009e-07 eta: 1 day, 6:54:39 time: 0.9990 data_time: 0.0045 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0083 decode.acc_seg: 99.8358 aux.loss_ce: 0.0083 aux.acc_seg: 99.4041 +04/16 21:40:45 - mmengine - INFO - Iter(train) [ 48550/160000] base_lr: 7.0316e-05 lr: 2.5997e-07 eta: 1 day, 6:53:50 time: 0.9994 data_time: 0.0044 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0082 decode.acc_seg: 99.7225 aux.loss_ce: 0.0090 aux.acc_seg: 99.2270 +04/16 21:41:35 - mmengine - INFO - Iter(train) [ 48600/160000] base_lr: 7.0284e-05 lr: 2.5986e-07 eta: 1 day, 6:53:00 time: 1.0002 data_time: 0.0049 memory: 8462 loss: 0.0192 decode.loss_ce: 0.0098 decode.acc_seg: 99.4205 aux.loss_ce: 0.0094 aux.acc_seg: 98.5340 +04/16 21:42:25 - mmengine - INFO - Iter(train) [ 48650/160000] base_lr: 7.0253e-05 lr: 2.5974e-07 eta: 1 day, 6:52:10 time: 0.9992 data_time: 0.0046 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0082 decode.acc_seg: 99.6180 aux.loss_ce: 0.0078 aux.acc_seg: 98.8667 +04/16 21:43:15 - mmengine - INFO - Iter(train) [ 48700/160000] base_lr: 7.0221e-05 lr: 2.5962e-07 eta: 1 day, 6:51:21 time: 1.0000 data_time: 0.0047 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0077 decode.acc_seg: 99.6328 aux.loss_ce: 0.0085 aux.acc_seg: 98.6403 +04/16 21:44:05 - mmengine - INFO - Iter(train) [ 48750/160000] base_lr: 7.0190e-05 lr: 2.5951e-07 eta: 1 day, 6:50:31 time: 1.0004 data_time: 0.0046 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0080 decode.acc_seg: 99.7383 aux.loss_ce: 0.0081 aux.acc_seg: 99.1625 +04/16 21:44:55 - mmengine - INFO - Iter(train) [ 48800/160000] base_lr: 7.0158e-05 lr: 2.5939e-07 eta: 1 day, 6:49:41 time: 0.9988 data_time: 0.0044 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0079 decode.acc_seg: 99.7572 aux.loss_ce: 0.0074 aux.acc_seg: 99.3153 +04/16 21:45:45 - mmengine - INFO - Iter(train) [ 48850/160000] base_lr: 7.0127e-05 lr: 2.5927e-07 eta: 1 day, 6:48:51 time: 0.9990 data_time: 0.0046 memory: 8462 loss: 0.0134 decode.loss_ce: 0.0067 decode.acc_seg: 99.6578 aux.loss_ce: 0.0066 aux.acc_seg: 99.3183 +04/16 21:46:35 - mmengine - INFO - Iter(train) [ 48900/160000] base_lr: 7.0095e-05 lr: 2.5916e-07 eta: 1 day, 6:48:02 time: 1.0000 data_time: 0.0044 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0080 decode.acc_seg: 99.6305 aux.loss_ce: 0.0080 aux.acc_seg: 98.9502 +04/16 21:47:25 - mmengine - INFO - Iter(train) [ 48950/160000] base_lr: 7.0064e-05 lr: 2.5904e-07 eta: 1 day, 6:47:12 time: 0.9987 data_time: 0.0046 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0088 decode.acc_seg: 99.4066 aux.loss_ce: 0.0089 aux.acc_seg: 98.9624 +04/16 21:48:15 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 21:48:15 - mmengine - INFO - Iter(train) [ 49000/160000] base_lr: 7.0032e-05 lr: 2.5892e-07 eta: 1 day, 6:46:22 time: 0.9999 data_time: 0.0049 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0089 decode.acc_seg: 99.5052 aux.loss_ce: 0.0092 aux.acc_seg: 99.0047 +04/16 21:49:05 - mmengine - INFO - Iter(train) [ 49050/160000] base_lr: 7.0000e-05 lr: 2.5881e-07 eta: 1 day, 6:45:33 time: 0.9999 data_time: 0.0045 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0089 decode.acc_seg: 99.7383 aux.loss_ce: 0.0088 aux.acc_seg: 99.0404 +04/16 21:49:55 - mmengine - INFO - Iter(train) [ 49100/160000] base_lr: 6.9969e-05 lr: 2.5869e-07 eta: 1 day, 6:44:43 time: 0.9999 data_time: 0.0045 memory: 8462 loss: 0.0197 decode.loss_ce: 0.0095 decode.acc_seg: 99.6180 aux.loss_ce: 0.0102 aux.acc_seg: 98.9094 +04/16 21:50:45 - mmengine - INFO - Iter(train) [ 49150/160000] base_lr: 6.9937e-05 lr: 2.5857e-07 eta: 1 day, 6:43:53 time: 1.0000 data_time: 0.0046 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0090 decode.acc_seg: 99.7953 aux.loss_ce: 0.0087 aux.acc_seg: 99.5382 +04/16 21:51:35 - mmengine - INFO - Iter(train) [ 49200/160000] base_lr: 6.9906e-05 lr: 2.5846e-07 eta: 1 day, 6:43:04 time: 0.9994 data_time: 0.0044 memory: 8462 loss: 0.0162 decode.loss_ce: 0.0080 decode.acc_seg: 99.7887 aux.loss_ce: 0.0082 aux.acc_seg: 99.2756 +04/16 21:52:25 - mmengine - INFO - Iter(train) [ 49250/160000] base_lr: 6.9874e-05 lr: 2.5834e-07 eta: 1 day, 6:42:14 time: 0.9990 data_time: 0.0049 memory: 8462 loss: 0.0129 decode.loss_ce: 0.0063 decode.acc_seg: 99.7440 aux.loss_ce: 0.0067 aux.acc_seg: 99.3977 +04/16 21:53:15 - mmengine - INFO - Iter(train) [ 49300/160000] base_lr: 6.9843e-05 lr: 2.5822e-07 eta: 1 day, 6:41:24 time: 0.9993 data_time: 0.0045 memory: 8462 loss: 0.0196 decode.loss_ce: 0.0095 decode.acc_seg: 99.5224 aux.loss_ce: 0.0101 aux.acc_seg: 98.4156 +04/16 21:54:05 - mmengine - INFO - Iter(train) [ 49350/160000] base_lr: 6.9811e-05 lr: 2.5811e-07 eta: 1 day, 6:40:34 time: 0.9999 data_time: 0.0044 memory: 8462 loss: 0.0162 decode.loss_ce: 0.0080 decode.acc_seg: 99.7000 aux.loss_ce: 0.0082 aux.acc_seg: 99.0486 +04/16 21:54:55 - mmengine - INFO - Iter(train) [ 49400/160000] base_lr: 6.9780e-05 lr: 2.5799e-07 eta: 1 day, 6:39:45 time: 0.9992 data_time: 0.0046 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0080 decode.acc_seg: 99.7208 aux.loss_ce: 0.0081 aux.acc_seg: 99.2958 +04/16 21:55:45 - mmengine - INFO - Iter(train) [ 49450/160000] base_lr: 6.9748e-05 lr: 2.5787e-07 eta: 1 day, 6:38:55 time: 0.9991 data_time: 0.0043 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0080 decode.acc_seg: 99.6529 aux.loss_ce: 0.0081 aux.acc_seg: 98.9460 +04/16 21:56:35 - mmengine - INFO - Iter(train) [ 49500/160000] base_lr: 6.9717e-05 lr: 2.5776e-07 eta: 1 day, 6:38:05 time: 0.9994 data_time: 0.0045 memory: 8462 loss: 0.0185 decode.loss_ce: 0.0097 decode.acc_seg: 99.6525 aux.loss_ce: 0.0088 aux.acc_seg: 99.2466 +04/16 21:57:25 - mmengine - INFO - Iter(train) [ 49550/160000] base_lr: 6.9685e-05 lr: 2.5764e-07 eta: 1 day, 6:37:15 time: 0.9997 data_time: 0.0051 memory: 8462 loss: 0.0159 decode.loss_ce: 0.0079 decode.acc_seg: 99.6832 aux.loss_ce: 0.0080 aux.acc_seg: 98.9904 +04/16 21:58:15 - mmengine - INFO - Iter(train) [ 49600/160000] base_lr: 6.9653e-05 lr: 2.5752e-07 eta: 1 day, 6:36:26 time: 1.0000 data_time: 0.0047 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0087 decode.acc_seg: 99.5953 aux.loss_ce: 0.0088 aux.acc_seg: 98.9237 +04/16 21:59:05 - mmengine - INFO - Iter(train) [ 49650/160000] base_lr: 6.9622e-05 lr: 2.5741e-07 eta: 1 day, 6:35:36 time: 1.0004 data_time: 0.0047 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0087 decode.acc_seg: 99.5466 aux.loss_ce: 0.0086 aux.acc_seg: 99.1205 +04/16 21:59:54 - mmengine - INFO - Iter(train) [ 49700/160000] base_lr: 6.9590e-05 lr: 2.5729e-07 eta: 1 day, 6:34:46 time: 0.9998 data_time: 0.0045 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0082 decode.acc_seg: 99.5956 aux.loss_ce: 0.0083 aux.acc_seg: 99.2729 +04/16 22:00:44 - mmengine - INFO - Iter(train) [ 49750/160000] base_lr: 6.9559e-05 lr: 2.5717e-07 eta: 1 day, 6:33:57 time: 0.9998 data_time: 0.0055 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0079 decode.acc_seg: 99.6555 aux.loss_ce: 0.0086 aux.acc_seg: 99.3425 +04/16 22:01:35 - mmengine - INFO - Iter(train) [ 49800/160000] base_lr: 6.9527e-05 lr: 2.5706e-07 eta: 1 day, 6:33:07 time: 0.9996 data_time: 0.0045 memory: 8462 loss: 0.0142 decode.loss_ce: 0.0070 decode.acc_seg: 99.5544 aux.loss_ce: 0.0072 aux.acc_seg: 98.7452 +04/16 22:02:24 - mmengine - INFO - Iter(train) [ 49850/160000] base_lr: 6.9496e-05 lr: 2.5694e-07 eta: 1 day, 6:32:17 time: 0.9992 data_time: 0.0046 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0079 decode.acc_seg: 99.6910 aux.loss_ce: 0.0078 aux.acc_seg: 99.1886 +04/16 22:03:14 - mmengine - INFO - Iter(train) [ 49900/160000] base_lr: 6.9464e-05 lr: 2.5682e-07 eta: 1 day, 6:31:27 time: 0.9999 data_time: 0.0045 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0081 decode.acc_seg: 99.6103 aux.loss_ce: 0.0085 aux.acc_seg: 99.0919 +04/16 22:04:04 - mmengine - INFO - Iter(train) [ 49950/160000] base_lr: 6.9433e-05 lr: 2.5671e-07 eta: 1 day, 6:30:38 time: 0.9997 data_time: 0.0048 memory: 8462 loss: 0.0189 decode.loss_ce: 0.0094 decode.acc_seg: 99.4986 aux.loss_ce: 0.0095 aux.acc_seg: 98.8159 +04/16 22:04:54 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 22:04:54 - mmengine - INFO - Iter(train) [ 50000/160000] base_lr: 6.9401e-05 lr: 2.5659e-07 eta: 1 day, 6:29:48 time: 0.9995 data_time: 0.0047 memory: 8462 loss: 0.0165 decode.loss_ce: 0.0079 decode.acc_seg: 99.7543 aux.loss_ce: 0.0085 aux.acc_seg: 99.2180 +04/16 22:04:54 - mmengine - INFO - Saving checkpoint at 50000 iterations +04/16 22:05:04 - mmengine - INFO - Iter(val) [ 50/355] eta: 0:00:35 time: 0.1157 data_time: 0.0016 memory: 4004 +04/16 22:05:10 - mmengine - INFO - Iter(val) [100/355] eta: 0:00:29 time: 0.1155 data_time: 0.0014 memory: 4004 +04/16 22:05:16 - mmengine - INFO - Iter(val) [150/355] eta: 0:00:23 time: 0.1161 data_time: 0.0017 memory: 4004 +04/16 22:05:22 - mmengine - INFO - Iter(val) [200/355] eta: 0:00:17 time: 0.1161 data_time: 0.0017 memory: 4004 +04/16 22:05:28 - mmengine - INFO - Iter(val) [250/355] eta: 0:00:12 time: 0.1158 data_time: 0.0015 memory: 4004 +04/16 22:05:33 - mmengine - INFO - Iter(val) [300/355] eta: 0:00:06 time: 0.1158 data_time: 0.0014 memory: 4004 +04/16 22:05:39 - mmengine - INFO - Iter(val) [350/355] eta: 0:00:00 time: 0.1155 data_time: 0.0012 memory: 4004 +04/16 22:05:40 - mmengine - INFO - per class results: +04/16 22:05:40 - mmengine - INFO - ++------------+-------+-------+ +| Class | IoU | Acc | ++------------+-------+-------+ +| background | 99.2 | 99.59 | +| contrast | 82.51 | 90.52 | ++------------+-------+-------+ +04/16 22:05:40 - mmengine - INFO - Iter(val) [355/355] aAcc: 99.2300 mIoU: 90.8500 mAcc: 95.0600 data_time: 0.0017 time: 0.1160 +04/16 22:06:30 - mmengine - INFO - Iter(train) [ 50050/160000] base_lr: 6.9370e-05 lr: 2.5647e-07 eta: 1 day, 6:28:59 time: 0.9975 data_time: 0.0047 memory: 8462 loss: 0.0141 decode.loss_ce: 0.0071 decode.acc_seg: 99.7061 aux.loss_ce: 0.0070 aux.acc_seg: 99.3778 +04/16 22:07:20 - mmengine - INFO - Iter(train) [ 50100/160000] base_lr: 6.9338e-05 lr: 2.5636e-07 eta: 1 day, 6:28:09 time: 0.9990 data_time: 0.0045 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0084 decode.acc_seg: 99.7219 aux.loss_ce: 0.0086 aux.acc_seg: 99.2453 +04/16 22:08:10 - mmengine - INFO - Iter(train) [ 50150/160000] base_lr: 6.9306e-05 lr: 2.5624e-07 eta: 1 day, 6:27:19 time: 0.9989 data_time: 0.0042 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0081 decode.acc_seg: 99.6168 aux.loss_ce: 0.0081 aux.acc_seg: 98.7686 +04/16 22:09:00 - mmengine - INFO - Iter(train) [ 50200/160000] base_lr: 6.9275e-05 lr: 2.5612e-07 eta: 1 day, 6:26:30 time: 0.9978 data_time: 0.0044 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0084 decode.acc_seg: 99.6302 aux.loss_ce: 0.0082 aux.acc_seg: 99.0292 +04/16 22:09:50 - mmengine - INFO - Iter(train) [ 50250/160000] base_lr: 6.9243e-05 lr: 2.5601e-07 eta: 1 day, 6:25:40 time: 1.0004 data_time: 0.0045 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0078 decode.acc_seg: 99.5680 aux.loss_ce: 0.0090 aux.acc_seg: 98.6553 +04/16 22:10:40 - mmengine - INFO - Iter(train) [ 50300/160000] base_lr: 6.9212e-05 lr: 2.5589e-07 eta: 1 day, 6:24:50 time: 0.9994 data_time: 0.0046 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0085 decode.acc_seg: 99.6376 aux.loss_ce: 0.0094 aux.acc_seg: 98.9412 +04/16 22:11:30 - mmengine - INFO - Iter(train) [ 50350/160000] base_lr: 6.9180e-05 lr: 2.5577e-07 eta: 1 day, 6:24:00 time: 0.9992 data_time: 0.0046 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0087 decode.acc_seg: 99.6986 aux.loss_ce: 0.0092 aux.acc_seg: 99.2458 +04/16 22:12:20 - mmengine - INFO - Iter(train) [ 50400/160000] base_lr: 6.9149e-05 lr: 2.5566e-07 eta: 1 day, 6:23:11 time: 1.0010 data_time: 0.0047 memory: 8462 loss: 0.0165 decode.loss_ce: 0.0080 decode.acc_seg: 99.7988 aux.loss_ce: 0.0085 aux.acc_seg: 99.4024 +04/16 22:13:10 - mmengine - INFO - Iter(train) [ 50450/160000] base_lr: 6.9117e-05 lr: 2.5554e-07 eta: 1 day, 6:22:21 time: 1.0004 data_time: 0.0043 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0073 decode.acc_seg: 99.6393 aux.loss_ce: 0.0076 aux.acc_seg: 99.0263 +04/16 22:14:00 - mmengine - INFO - Iter(train) [ 50500/160000] base_lr: 6.9086e-05 lr: 2.5542e-07 eta: 1 day, 6:21:31 time: 0.9992 data_time: 0.0047 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0085 decode.acc_seg: 99.5884 aux.loss_ce: 0.0089 aux.acc_seg: 99.0612 +04/16 22:14:50 - mmengine - INFO - Iter(train) [ 50550/160000] base_lr: 6.9054e-05 lr: 2.5531e-07 eta: 1 day, 6:20:41 time: 0.9989 data_time: 0.0045 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0076 decode.acc_seg: 99.7112 aux.loss_ce: 0.0079 aux.acc_seg: 99.3021 +04/16 22:15:40 - mmengine - INFO - Iter(train) [ 50600/160000] base_lr: 6.9023e-05 lr: 2.5519e-07 eta: 1 day, 6:19:51 time: 1.0007 data_time: 0.0049 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0074 decode.acc_seg: 99.5321 aux.loss_ce: 0.0076 aux.acc_seg: 98.6328 +04/16 22:16:30 - mmengine - INFO - Iter(train) [ 50650/160000] base_lr: 6.8991e-05 lr: 2.5507e-07 eta: 1 day, 6:19:02 time: 0.9999 data_time: 0.0045 memory: 8462 loss: 0.0163 decode.loss_ce: 0.0081 decode.acc_seg: 99.5033 aux.loss_ce: 0.0083 aux.acc_seg: 98.3282 +04/16 22:17:20 - mmengine - INFO - Iter(train) [ 50700/160000] base_lr: 6.8959e-05 lr: 2.5496e-07 eta: 1 day, 6:18:12 time: 0.9985 data_time: 0.0047 memory: 8462 loss: 0.0146 decode.loss_ce: 0.0070 decode.acc_seg: 99.7309 aux.loss_ce: 0.0077 aux.acc_seg: 99.2088 +04/16 22:18:09 - mmengine - INFO - Iter(train) [ 50750/160000] base_lr: 6.8928e-05 lr: 2.5484e-07 eta: 1 day, 6:17:22 time: 0.9993 data_time: 0.0050 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0079 decode.acc_seg: 99.6696 aux.loss_ce: 0.0078 aux.acc_seg: 99.0030 +04/16 22:18:59 - mmengine - INFO - Iter(train) [ 50800/160000] base_lr: 6.8896e-05 lr: 2.5472e-07 eta: 1 day, 6:16:32 time: 0.9997 data_time: 0.0050 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0075 decode.acc_seg: 99.7080 aux.loss_ce: 0.0080 aux.acc_seg: 98.9895 +04/16 22:19:49 - mmengine - INFO - Iter(train) [ 50850/160000] base_lr: 6.8865e-05 lr: 2.5461e-07 eta: 1 day, 6:15:43 time: 0.9996 data_time: 0.0047 memory: 8462 loss: 0.0162 decode.loss_ce: 0.0080 decode.acc_seg: 99.5073 aux.loss_ce: 0.0082 aux.acc_seg: 98.5756 +04/16 22:20:39 - mmengine - INFO - Iter(train) [ 50900/160000] base_lr: 6.8833e-05 lr: 2.5449e-07 eta: 1 day, 6:14:53 time: 0.9996 data_time: 0.0054 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0076 decode.acc_seg: 99.6412 aux.loss_ce: 0.0078 aux.acc_seg: 99.0788 +04/16 22:21:29 - mmengine - INFO - Iter(train) [ 50950/160000] base_lr: 6.8802e-05 lr: 2.5437e-07 eta: 1 day, 6:14:03 time: 1.0000 data_time: 0.0050 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0084 decode.acc_seg: 99.7515 aux.loss_ce: 0.0093 aux.acc_seg: 99.1920 +04/16 22:22:19 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 22:22:19 - mmengine - INFO - Iter(train) [ 51000/160000] base_lr: 6.8770e-05 lr: 2.5426e-07 eta: 1 day, 6:13:13 time: 0.9996 data_time: 0.0052 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0075 decode.acc_seg: 99.5951 aux.loss_ce: 0.0086 aux.acc_seg: 99.1821 +04/16 22:23:09 - mmengine - INFO - Iter(train) [ 51050/160000] base_lr: 6.8739e-05 lr: 2.5414e-07 eta: 1 day, 6:12:23 time: 0.9980 data_time: 0.0049 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0082 decode.acc_seg: 99.5388 aux.loss_ce: 0.0087 aux.acc_seg: 98.9073 +04/16 22:23:59 - mmengine - INFO - Iter(train) [ 51100/160000] base_lr: 6.8707e-05 lr: 2.5402e-07 eta: 1 day, 6:11:34 time: 0.9982 data_time: 0.0046 memory: 8462 loss: 0.0180 decode.loss_ce: 0.0088 decode.acc_seg: 99.5554 aux.loss_ce: 0.0092 aux.acc_seg: 98.8064 +04/16 22:24:49 - mmengine - INFO - Iter(train) [ 51150/160000] base_lr: 6.8676e-05 lr: 2.5391e-07 eta: 1 day, 6:10:44 time: 1.0000 data_time: 0.0050 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0089 decode.acc_seg: 99.7726 aux.loss_ce: 0.0091 aux.acc_seg: 99.3719 +04/16 22:25:39 - mmengine - INFO - Iter(train) [ 51200/160000] base_lr: 6.8644e-05 lr: 2.5379e-07 eta: 1 day, 6:09:54 time: 0.9995 data_time: 0.0053 memory: 8462 loss: 0.0171 decode.loss_ce: 0.0079 decode.acc_seg: 99.7620 aux.loss_ce: 0.0092 aux.acc_seg: 99.0164 +04/16 22:26:29 - mmengine - INFO - Iter(train) [ 51250/160000] base_lr: 6.8612e-05 lr: 2.5367e-07 eta: 1 day, 6:09:04 time: 0.9986 data_time: 0.0043 memory: 8462 loss: 0.0191 decode.loss_ce: 0.0099 decode.acc_seg: 99.6565 aux.loss_ce: 0.0092 aux.acc_seg: 99.2025 +04/16 22:27:19 - mmengine - INFO - Iter(train) [ 51300/160000] base_lr: 6.8581e-05 lr: 2.5356e-07 eta: 1 day, 6:08:14 time: 0.9980 data_time: 0.0045 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0085 decode.acc_seg: 99.6807 aux.loss_ce: 0.0088 aux.acc_seg: 99.3999 +04/16 22:28:09 - mmengine - INFO - Iter(train) [ 51350/160000] base_lr: 6.8549e-05 lr: 2.5344e-07 eta: 1 day, 6:07:25 time: 0.9983 data_time: 0.0048 memory: 8462 loss: 0.0183 decode.loss_ce: 0.0091 decode.acc_seg: 99.6429 aux.loss_ce: 0.0092 aux.acc_seg: 99.0431 +04/16 22:28:59 - mmengine - INFO - Iter(train) [ 51400/160000] base_lr: 6.8518e-05 lr: 2.5332e-07 eta: 1 day, 6:06:35 time: 0.9991 data_time: 0.0046 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0087 decode.acc_seg: 99.4207 aux.loss_ce: 0.0090 aux.acc_seg: 98.5525 +04/16 22:29:49 - mmengine - INFO - Iter(train) [ 51450/160000] base_lr: 6.8486e-05 lr: 2.5321e-07 eta: 1 day, 6:05:45 time: 0.9985 data_time: 0.0044 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0081 decode.acc_seg: 99.6510 aux.loss_ce: 0.0079 aux.acc_seg: 99.2872 +04/16 22:30:39 - mmengine - INFO - Iter(train) [ 51500/160000] base_lr: 6.8455e-05 lr: 2.5309e-07 eta: 1 day, 6:04:55 time: 0.9987 data_time: 0.0052 memory: 8462 loss: 0.0165 decode.loss_ce: 0.0085 decode.acc_seg: 99.4526 aux.loss_ce: 0.0080 aux.acc_seg: 98.9122 +04/16 22:31:29 - mmengine - INFO - Iter(train) [ 51550/160000] base_lr: 6.8423e-05 lr: 2.5297e-07 eta: 1 day, 6:04:06 time: 0.9998 data_time: 0.0051 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0074 decode.acc_seg: 99.8579 aux.loss_ce: 0.0083 aux.acc_seg: 99.5220 +04/16 22:32:19 - mmengine - INFO - Iter(train) [ 51600/160000] base_lr: 6.8392e-05 lr: 2.5286e-07 eta: 1 day, 6:03:16 time: 1.0012 data_time: 0.0047 memory: 8462 loss: 0.0159 decode.loss_ce: 0.0077 decode.acc_seg: 99.5605 aux.loss_ce: 0.0082 aux.acc_seg: 98.6015 +04/16 22:33:09 - mmengine - INFO - Iter(train) [ 51650/160000] base_lr: 6.8360e-05 lr: 2.5274e-07 eta: 1 day, 6:02:26 time: 0.9984 data_time: 0.0046 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0078 decode.acc_seg: 99.7519 aux.loss_ce: 0.0079 aux.acc_seg: 99.1133 +04/16 22:33:59 - mmengine - INFO - Iter(train) [ 51700/160000] base_lr: 6.8329e-05 lr: 2.5262e-07 eta: 1 day, 6:01:36 time: 0.9997 data_time: 0.0048 memory: 8462 loss: 0.0191 decode.loss_ce: 0.0092 decode.acc_seg: 99.6235 aux.loss_ce: 0.0099 aux.acc_seg: 98.6456 +04/16 22:34:49 - mmengine - INFO - Iter(train) [ 51750/160000] base_lr: 6.8297e-05 lr: 2.5251e-07 eta: 1 day, 6:00:46 time: 0.9982 data_time: 0.0049 memory: 8462 loss: 0.0142 decode.loss_ce: 0.0069 decode.acc_seg: 99.4482 aux.loss_ce: 0.0073 aux.acc_seg: 99.1529 +04/16 22:35:39 - mmengine - INFO - Iter(train) [ 51800/160000] base_lr: 6.8265e-05 lr: 2.5239e-07 eta: 1 day, 5:59:57 time: 0.9980 data_time: 0.0046 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0078 decode.acc_seg: 99.7250 aux.loss_ce: 0.0077 aux.acc_seg: 99.1871 +04/16 22:36:29 - mmengine - INFO - Iter(train) [ 51850/160000] base_lr: 6.8234e-05 lr: 2.5227e-07 eta: 1 day, 5:59:07 time: 0.9995 data_time: 0.0047 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0085 decode.acc_seg: 99.4257 aux.loss_ce: 0.0082 aux.acc_seg: 98.5600 +04/16 22:37:19 - mmengine - INFO - Iter(train) [ 51900/160000] base_lr: 6.8202e-05 lr: 2.5216e-07 eta: 1 day, 5:58:17 time: 0.9998 data_time: 0.0044 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0083 decode.acc_seg: 99.7498 aux.loss_ce: 0.0085 aux.acc_seg: 99.4606 +04/16 22:38:09 - mmengine - INFO - Iter(train) [ 51950/160000] base_lr: 6.8171e-05 lr: 2.5204e-07 eta: 1 day, 5:57:27 time: 0.9980 data_time: 0.0044 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0073 decode.acc_seg: 99.6624 aux.loss_ce: 0.0077 aux.acc_seg: 99.2783 +04/16 22:38:59 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 22:38:59 - mmengine - INFO - Iter(train) [ 52000/160000] base_lr: 6.8139e-05 lr: 2.5192e-07 eta: 1 day, 5:56:37 time: 0.9982 data_time: 0.0048 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0087 decode.acc_seg: 99.7227 aux.loss_ce: 0.0087 aux.acc_seg: 99.4970 +04/16 22:39:49 - mmengine - INFO - Iter(train) [ 52050/160000] base_lr: 6.8108e-05 lr: 2.5181e-07 eta: 1 day, 5:55:47 time: 0.9975 data_time: 0.0048 memory: 8462 loss: 0.0189 decode.loss_ce: 0.0101 decode.acc_seg: 99.6786 aux.loss_ce: 0.0088 aux.acc_seg: 99.0822 +04/16 22:40:38 - mmengine - INFO - Iter(train) [ 52100/160000] base_lr: 6.8076e-05 lr: 2.5169e-07 eta: 1 day, 5:54:57 time: 0.9991 data_time: 0.0046 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0082 decode.acc_seg: 99.5260 aux.loss_ce: 0.0085 aux.acc_seg: 98.5950 +04/16 22:41:28 - mmengine - INFO - Iter(train) [ 52150/160000] base_lr: 6.8045e-05 lr: 2.5157e-07 eta: 1 day, 5:54:08 time: 0.9993 data_time: 0.0044 memory: 8462 loss: 0.0145 decode.loss_ce: 0.0071 decode.acc_seg: 99.7528 aux.loss_ce: 0.0074 aux.acc_seg: 99.5098 +04/16 22:42:18 - mmengine - INFO - Iter(train) [ 52200/160000] base_lr: 6.8013e-05 lr: 2.5146e-07 eta: 1 day, 5:53:18 time: 0.9987 data_time: 0.0047 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0077 decode.acc_seg: 99.6742 aux.loss_ce: 0.0073 aux.acc_seg: 99.1848 +04/16 22:43:08 - mmengine - INFO - Iter(train) [ 52250/160000] base_lr: 6.7982e-05 lr: 2.5134e-07 eta: 1 day, 5:52:28 time: 0.9990 data_time: 0.0046 memory: 8462 loss: 0.0198 decode.loss_ce: 0.0098 decode.acc_seg: 99.3906 aux.loss_ce: 0.0100 aux.acc_seg: 98.8962 +04/16 22:43:58 - mmengine - INFO - Iter(train) [ 52300/160000] base_lr: 6.7950e-05 lr: 2.5122e-07 eta: 1 day, 5:51:38 time: 0.9986 data_time: 0.0046 memory: 8462 loss: 0.0141 decode.loss_ce: 0.0067 decode.acc_seg: 99.7282 aux.loss_ce: 0.0074 aux.acc_seg: 99.0484 +04/16 22:44:48 - mmengine - INFO - Iter(train) [ 52350/160000] base_lr: 6.7918e-05 lr: 2.5111e-07 eta: 1 day, 5:50:48 time: 0.9993 data_time: 0.0046 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0085 decode.acc_seg: 99.6134 aux.loss_ce: 0.0094 aux.acc_seg: 99.1953 +04/16 22:45:38 - mmengine - INFO - Iter(train) [ 52400/160000] base_lr: 6.7887e-05 lr: 2.5099e-07 eta: 1 day, 5:49:58 time: 0.9989 data_time: 0.0045 memory: 8462 loss: 0.0165 decode.loss_ce: 0.0084 decode.acc_seg: 99.6756 aux.loss_ce: 0.0081 aux.acc_seg: 99.2476 +04/16 22:46:28 - mmengine - INFO - Iter(train) [ 52450/160000] base_lr: 6.7855e-05 lr: 2.5088e-07 eta: 1 day, 5:49:08 time: 0.9982 data_time: 0.0045 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0082 decode.acc_seg: 99.6368 aux.loss_ce: 0.0084 aux.acc_seg: 99.0398 +04/16 22:47:18 - mmengine - INFO - Iter(train) [ 52500/160000] base_lr: 6.7824e-05 lr: 2.5076e-07 eta: 1 day, 5:48:18 time: 0.9989 data_time: 0.0045 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0082 decode.acc_seg: 99.6225 aux.loss_ce: 0.0078 aux.acc_seg: 98.9750 +04/16 22:48:08 - mmengine - INFO - Iter(train) [ 52550/160000] base_lr: 6.7792e-05 lr: 2.5064e-07 eta: 1 day, 5:47:29 time: 0.9984 data_time: 0.0045 memory: 8462 loss: 0.0171 decode.loss_ce: 0.0084 decode.acc_seg: 99.6349 aux.loss_ce: 0.0088 aux.acc_seg: 99.2867 +04/16 22:48:58 - mmengine - INFO - Iter(train) [ 52600/160000] base_lr: 6.7761e-05 lr: 2.5053e-07 eta: 1 day, 5:46:39 time: 0.9983 data_time: 0.0043 memory: 8462 loss: 0.0185 decode.loss_ce: 0.0092 decode.acc_seg: 99.5438 aux.loss_ce: 0.0093 aux.acc_seg: 98.9717 +04/16 22:49:48 - mmengine - INFO - Iter(train) [ 52650/160000] base_lr: 6.7729e-05 lr: 2.5041e-07 eta: 1 day, 5:45:49 time: 0.9980 data_time: 0.0050 memory: 8462 loss: 0.0134 decode.loss_ce: 0.0065 decode.acc_seg: 99.7768 aux.loss_ce: 0.0068 aux.acc_seg: 99.2828 +04/16 22:50:38 - mmengine - INFO - Iter(train) [ 52700/160000] base_lr: 6.7698e-05 lr: 2.5029e-07 eta: 1 day, 5:44:59 time: 0.9978 data_time: 0.0046 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0093 decode.acc_seg: 99.6967 aux.loss_ce: 0.0086 aux.acc_seg: 99.1804 +04/16 22:51:27 - mmengine - INFO - Iter(train) [ 52750/160000] base_lr: 6.7666e-05 lr: 2.5018e-07 eta: 1 day, 5:44:09 time: 0.9980 data_time: 0.0046 memory: 8462 loss: 0.0182 decode.loss_ce: 0.0089 decode.acc_seg: 99.5892 aux.loss_ce: 0.0093 aux.acc_seg: 98.9882 +04/16 22:52:17 - mmengine - INFO - Iter(train) [ 52800/160000] base_lr: 6.7634e-05 lr: 2.5006e-07 eta: 1 day, 5:43:19 time: 0.9984 data_time: 0.0046 memory: 8462 loss: 0.0162 decode.loss_ce: 0.0080 decode.acc_seg: 99.5016 aux.loss_ce: 0.0082 aux.acc_seg: 98.7511 +04/16 22:53:07 - mmengine - INFO - Iter(train) [ 52850/160000] base_lr: 6.7603e-05 lr: 2.4994e-07 eta: 1 day, 5:42:29 time: 0.9977 data_time: 0.0051 memory: 8462 loss: 0.0171 decode.loss_ce: 0.0083 decode.acc_seg: 99.7194 aux.loss_ce: 0.0088 aux.acc_seg: 99.2336 +04/16 22:53:57 - mmengine - INFO - Iter(train) [ 52900/160000] base_lr: 6.7571e-05 lr: 2.4983e-07 eta: 1 day, 5:41:39 time: 0.9979 data_time: 0.0048 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0076 decode.acc_seg: 99.6332 aux.loss_ce: 0.0084 aux.acc_seg: 99.3027 +04/16 22:54:47 - mmengine - INFO - Iter(train) [ 52950/160000] base_lr: 6.7540e-05 lr: 2.4971e-07 eta: 1 day, 5:40:49 time: 0.9977 data_time: 0.0052 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0075 decode.acc_seg: 99.6027 aux.loss_ce: 0.0082 aux.acc_seg: 98.8695 +04/16 22:55:37 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 22:55:37 - mmengine - INFO - Iter(train) [ 53000/160000] base_lr: 6.7508e-05 lr: 2.4959e-07 eta: 1 day, 5:40:00 time: 0.9995 data_time: 0.0050 memory: 8462 loss: 0.0183 decode.loss_ce: 0.0098 decode.acc_seg: 99.7593 aux.loss_ce: 0.0085 aux.acc_seg: 99.2371 +04/16 22:56:27 - mmengine - INFO - Iter(train) [ 53050/160000] base_lr: 6.7477e-05 lr: 2.4948e-07 eta: 1 day, 5:39:10 time: 0.9956 data_time: 0.0045 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0089 decode.acc_seg: 99.7759 aux.loss_ce: 0.0088 aux.acc_seg: 99.2552 +04/16 22:57:17 - mmengine - INFO - Iter(train) [ 53100/160000] base_lr: 6.7445e-05 lr: 2.4936e-07 eta: 1 day, 5:38:20 time: 0.9984 data_time: 0.0046 memory: 8462 loss: 0.0162 decode.loss_ce: 0.0078 decode.acc_seg: 99.7066 aux.loss_ce: 0.0085 aux.acc_seg: 99.1474 +04/16 22:58:07 - mmengine - INFO - Iter(train) [ 53150/160000] base_lr: 6.7414e-05 lr: 2.4924e-07 eta: 1 day, 5:37:30 time: 0.9986 data_time: 0.0045 memory: 8462 loss: 0.0196 decode.loss_ce: 0.0092 decode.acc_seg: 99.6326 aux.loss_ce: 0.0103 aux.acc_seg: 98.9431 +04/16 22:58:57 - mmengine - INFO - Iter(train) [ 53200/160000] base_lr: 6.7382e-05 lr: 2.4913e-07 eta: 1 day, 5:36:40 time: 0.9965 data_time: 0.0045 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0079 decode.acc_seg: 99.7290 aux.loss_ce: 0.0081 aux.acc_seg: 99.2523 +04/16 22:59:46 - mmengine - INFO - Iter(train) [ 53250/160000] base_lr: 6.7351e-05 lr: 2.4901e-07 eta: 1 day, 5:35:50 time: 0.9971 data_time: 0.0044 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0085 decode.acc_seg: 99.4743 aux.loss_ce: 0.0088 aux.acc_seg: 98.8598 +04/16 23:00:36 - mmengine - INFO - Iter(train) [ 53300/160000] base_lr: 6.7319e-05 lr: 2.4889e-07 eta: 1 day, 5:35:00 time: 0.9971 data_time: 0.0046 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0081 decode.acc_seg: 99.7292 aux.loss_ce: 0.0085 aux.acc_seg: 99.1329 +04/16 23:01:26 - mmengine - INFO - Iter(train) [ 53350/160000] base_lr: 6.7287e-05 lr: 2.4878e-07 eta: 1 day, 5:34:10 time: 0.9985 data_time: 0.0055 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0085 decode.acc_seg: 99.3555 aux.loss_ce: 0.0089 aux.acc_seg: 98.5073 +04/16 23:02:16 - mmengine - INFO - Iter(train) [ 53400/160000] base_lr: 6.7256e-05 lr: 2.4866e-07 eta: 1 day, 5:33:20 time: 0.9964 data_time: 0.0047 memory: 8462 loss: 0.0169 decode.loss_ce: 0.0080 decode.acc_seg: 99.5564 aux.loss_ce: 0.0089 aux.acc_seg: 98.7888 +04/16 23:03:06 - mmengine - INFO - Iter(train) [ 53450/160000] base_lr: 6.7224e-05 lr: 2.4854e-07 eta: 1 day, 5:32:30 time: 0.9976 data_time: 0.0048 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0072 decode.acc_seg: 99.7770 aux.loss_ce: 0.0077 aux.acc_seg: 99.4011 +04/16 23:03:56 - mmengine - INFO - Iter(train) [ 53500/160000] base_lr: 6.7193e-05 lr: 2.4843e-07 eta: 1 day, 5:31:40 time: 0.9953 data_time: 0.0045 memory: 8462 loss: 0.0162 decode.loss_ce: 0.0082 decode.acc_seg: 99.6235 aux.loss_ce: 0.0080 aux.acc_seg: 99.2918 +04/16 23:04:46 - mmengine - INFO - Iter(train) [ 53550/160000] base_lr: 6.7161e-05 lr: 2.4831e-07 eta: 1 day, 5:30:50 time: 0.9969 data_time: 0.0051 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0082 decode.acc_seg: 99.6235 aux.loss_ce: 0.0089 aux.acc_seg: 99.0492 +04/16 23:05:35 - mmengine - INFO - Iter(train) [ 53600/160000] base_lr: 6.7130e-05 lr: 2.4819e-07 eta: 1 day, 5:30:00 time: 0.9979 data_time: 0.0047 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0087 decode.acc_seg: 99.6069 aux.loss_ce: 0.0080 aux.acc_seg: 98.9807 +04/16 23:06:25 - mmengine - INFO - Iter(train) [ 53650/160000] base_lr: 6.7098e-05 lr: 2.4808e-07 eta: 1 day, 5:29:10 time: 0.9974 data_time: 0.0045 memory: 8462 loss: 0.0159 decode.loss_ce: 0.0079 decode.acc_seg: 99.6035 aux.loss_ce: 0.0080 aux.acc_seg: 99.0023 +04/16 23:07:15 - mmengine - INFO - Iter(train) [ 53700/160000] base_lr: 6.7067e-05 lr: 2.4796e-07 eta: 1 day, 5:28:20 time: 0.9980 data_time: 0.0048 memory: 8462 loss: 0.0165 decode.loss_ce: 0.0077 decode.acc_seg: 99.7391 aux.loss_ce: 0.0088 aux.acc_seg: 99.2634 +04/16 23:08:05 - mmengine - INFO - Iter(train) [ 53750/160000] base_lr: 6.7035e-05 lr: 2.4784e-07 eta: 1 day, 5:27:30 time: 0.9970 data_time: 0.0048 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0077 decode.acc_seg: 99.7259 aux.loss_ce: 0.0083 aux.acc_seg: 99.3168 +04/16 23:08:55 - mmengine - INFO - Iter(train) [ 53800/160000] base_lr: 6.7004e-05 lr: 2.4773e-07 eta: 1 day, 5:26:40 time: 0.9972 data_time: 0.0048 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0082 decode.acc_seg: 99.5207 aux.loss_ce: 0.0092 aux.acc_seg: 98.8226 +04/16 23:09:45 - mmengine - INFO - Iter(train) [ 53850/160000] base_lr: 6.6972e-05 lr: 2.4761e-07 eta: 1 day, 5:25:50 time: 0.9959 data_time: 0.0049 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0085 decode.acc_seg: 99.5327 aux.loss_ce: 0.0081 aux.acc_seg: 98.8750 +04/16 23:10:35 - mmengine - INFO - Iter(train) [ 53900/160000] base_lr: 6.6940e-05 lr: 2.4749e-07 eta: 1 day, 5:25:00 time: 0.9966 data_time: 0.0042 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0071 decode.acc_seg: 99.7131 aux.loss_ce: 0.0069 aux.acc_seg: 99.2233 +04/16 23:11:24 - mmengine - INFO - Iter(train) [ 53950/160000] base_lr: 6.6909e-05 lr: 2.4738e-07 eta: 1 day, 5:24:09 time: 0.9965 data_time: 0.0047 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0082 decode.acc_seg: 99.6389 aux.loss_ce: 0.0085 aux.acc_seg: 98.9637 +04/16 23:12:14 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 23:12:14 - mmengine - INFO - Iter(train) [ 54000/160000] base_lr: 6.6877e-05 lr: 2.4726e-07 eta: 1 day, 5:23:20 time: 0.9965 data_time: 0.0044 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0080 decode.acc_seg: 99.5647 aux.loss_ce: 0.0080 aux.acc_seg: 98.7936 +04/16 23:13:04 - mmengine - INFO - Iter(train) [ 54050/160000] base_lr: 6.6846e-05 lr: 2.4714e-07 eta: 1 day, 5:22:29 time: 0.9959 data_time: 0.0045 memory: 8462 loss: 0.0182 decode.loss_ce: 0.0096 decode.acc_seg: 99.6729 aux.loss_ce: 0.0087 aux.acc_seg: 98.9664 +04/16 23:13:54 - mmengine - INFO - Iter(train) [ 54100/160000] base_lr: 6.6814e-05 lr: 2.4703e-07 eta: 1 day, 5:21:39 time: 0.9961 data_time: 0.0048 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0095 decode.acc_seg: 99.4678 aux.loss_ce: 0.0083 aux.acc_seg: 98.9416 +04/16 23:14:44 - mmengine - INFO - Iter(train) [ 54150/160000] base_lr: 6.6783e-05 lr: 2.4691e-07 eta: 1 day, 5:20:49 time: 0.9969 data_time: 0.0045 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0075 decode.acc_seg: 99.7181 aux.loss_ce: 0.0080 aux.acc_seg: 99.0208 +04/16 23:15:34 - mmengine - INFO - Iter(train) [ 54200/160000] base_lr: 6.6751e-05 lr: 2.4679e-07 eta: 1 day, 5:19:59 time: 0.9958 data_time: 0.0045 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0079 decode.acc_seg: 99.6202 aux.loss_ce: 0.0081 aux.acc_seg: 98.9937 +04/16 23:16:23 - mmengine - INFO - Iter(train) [ 54250/160000] base_lr: 6.6720e-05 lr: 2.4668e-07 eta: 1 day, 5:19:09 time: 0.9970 data_time: 0.0045 memory: 8462 loss: 0.0159 decode.loss_ce: 0.0079 decode.acc_seg: 99.6475 aux.loss_ce: 0.0080 aux.acc_seg: 98.8007 +04/16 23:17:13 - mmengine - INFO - Iter(train) [ 54300/160000] base_lr: 6.6688e-05 lr: 2.4656e-07 eta: 1 day, 5:18:19 time: 0.9948 data_time: 0.0044 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0083 decode.acc_seg: 99.6634 aux.loss_ce: 0.0084 aux.acc_seg: 98.9866 +04/16 23:18:03 - mmengine - INFO - Iter(train) [ 54350/160000] base_lr: 6.6657e-05 lr: 2.4644e-07 eta: 1 day, 5:17:29 time: 0.9964 data_time: 0.0047 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0084 decode.acc_seg: 99.7025 aux.loss_ce: 0.0092 aux.acc_seg: 98.9573 +04/16 23:18:53 - mmengine - INFO - Iter(train) [ 54400/160000] base_lr: 6.6625e-05 lr: 2.4633e-07 eta: 1 day, 5:16:39 time: 0.9966 data_time: 0.0045 memory: 8462 loss: 0.0163 decode.loss_ce: 0.0078 decode.acc_seg: 99.7503 aux.loss_ce: 0.0085 aux.acc_seg: 99.2405 +04/16 23:19:43 - mmengine - INFO - Iter(train) [ 54450/160000] base_lr: 6.6593e-05 lr: 2.4621e-07 eta: 1 day, 5:15:49 time: 0.9972 data_time: 0.0051 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0077 decode.acc_seg: 99.6162 aux.loss_ce: 0.0080 aux.acc_seg: 99.0257 +04/16 23:20:33 - mmengine - INFO - Iter(train) [ 54500/160000] base_lr: 6.6562e-05 lr: 2.4609e-07 eta: 1 day, 5:14:59 time: 0.9969 data_time: 0.0045 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0097 decode.acc_seg: 99.5949 aux.loss_ce: 0.0090 aux.acc_seg: 99.1062 +04/16 23:21:22 - mmengine - INFO - Iter(train) [ 54550/160000] base_lr: 6.6530e-05 lr: 2.4598e-07 eta: 1 day, 5:14:09 time: 0.9967 data_time: 0.0051 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0085 decode.acc_seg: 99.5005 aux.loss_ce: 0.0090 aux.acc_seg: 99.0097 +04/16 23:22:12 - mmengine - INFO - Iter(train) [ 54600/160000] base_lr: 6.6499e-05 lr: 2.4586e-07 eta: 1 day, 5:13:19 time: 0.9958 data_time: 0.0048 memory: 8462 loss: 0.0180 decode.loss_ce: 0.0090 decode.acc_seg: 99.7503 aux.loss_ce: 0.0090 aux.acc_seg: 99.2399 +04/16 23:23:02 - mmengine - INFO - Iter(train) [ 54650/160000] base_lr: 6.6467e-05 lr: 2.4574e-07 eta: 1 day, 5:12:29 time: 0.9963 data_time: 0.0050 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0076 decode.acc_seg: 99.6700 aux.loss_ce: 0.0072 aux.acc_seg: 99.1671 +04/16 23:23:52 - mmengine - INFO - Iter(train) [ 54700/160000] base_lr: 6.6436e-05 lr: 2.4563e-07 eta: 1 day, 5:11:39 time: 0.9970 data_time: 0.0044 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0088 decode.acc_seg: 99.6424 aux.loss_ce: 0.0086 aux.acc_seg: 98.8768 +04/16 23:24:42 - mmengine - INFO - Iter(train) [ 54750/160000] base_lr: 6.6404e-05 lr: 2.4551e-07 eta: 1 day, 5:10:48 time: 0.9971 data_time: 0.0048 memory: 8462 loss: 0.0151 decode.loss_ce: 0.0071 decode.acc_seg: 99.7747 aux.loss_ce: 0.0080 aux.acc_seg: 99.3519 +04/16 23:25:32 - mmengine - INFO - Iter(train) [ 54800/160000] base_lr: 6.6373e-05 lr: 2.4539e-07 eta: 1 day, 5:09:58 time: 0.9972 data_time: 0.0043 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0082 decode.acc_seg: 99.5329 aux.loss_ce: 0.0085 aux.acc_seg: 98.6334 +04/16 23:26:21 - mmengine - INFO - Iter(train) [ 54850/160000] base_lr: 6.6341e-05 lr: 2.4528e-07 eta: 1 day, 5:09:08 time: 0.9965 data_time: 0.0048 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0082 decode.acc_seg: 99.6662 aux.loss_ce: 0.0085 aux.acc_seg: 99.0597 +04/16 23:27:11 - mmengine - INFO - Iter(train) [ 54900/160000] base_lr: 6.6310e-05 lr: 2.4516e-07 eta: 1 day, 5:08:18 time: 0.9964 data_time: 0.0046 memory: 8462 loss: 0.0159 decode.loss_ce: 0.0081 decode.acc_seg: 99.8238 aux.loss_ce: 0.0078 aux.acc_seg: 99.3221 +04/16 23:28:01 - mmengine - INFO - Iter(train) [ 54950/160000] base_lr: 6.6278e-05 lr: 2.4504e-07 eta: 1 day, 5:07:28 time: 0.9963 data_time: 0.0046 memory: 8462 loss: 0.0171 decode.loss_ce: 0.0083 decode.acc_seg: 99.7200 aux.loss_ce: 0.0088 aux.acc_seg: 99.0465 +04/16 23:28:51 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 23:28:51 - mmengine - INFO - Iter(train) [ 55000/160000] base_lr: 6.6246e-05 lr: 2.4493e-07 eta: 1 day, 5:06:38 time: 0.9960 data_time: 0.0045 memory: 8462 loss: 0.0162 decode.loss_ce: 0.0078 decode.acc_seg: 99.6517 aux.loss_ce: 0.0084 aux.acc_seg: 99.1795 +04/16 23:29:41 - mmengine - INFO - Iter(train) [ 55050/160000] base_lr: 6.6215e-05 lr: 2.4481e-07 eta: 1 day, 5:05:48 time: 0.9956 data_time: 0.0046 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0081 decode.acc_seg: 99.7381 aux.loss_ce: 0.0089 aux.acc_seg: 99.3469 +04/16 23:30:30 - mmengine - INFO - Iter(train) [ 55100/160000] base_lr: 6.6183e-05 lr: 2.4469e-07 eta: 1 day, 5:04:58 time: 0.9949 data_time: 0.0044 memory: 8462 loss: 0.0151 decode.loss_ce: 0.0073 decode.acc_seg: 99.5972 aux.loss_ce: 0.0078 aux.acc_seg: 98.9872 +04/16 23:31:20 - mmengine - INFO - Iter(train) [ 55150/160000] base_lr: 6.6152e-05 lr: 2.4458e-07 eta: 1 day, 5:04:08 time: 0.9965 data_time: 0.0047 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0074 decode.acc_seg: 99.7822 aux.loss_ce: 0.0078 aux.acc_seg: 99.3582 +04/16 23:32:10 - mmengine - INFO - Iter(train) [ 55200/160000] base_lr: 6.6120e-05 lr: 2.4446e-07 eta: 1 day, 5:03:18 time: 0.9967 data_time: 0.0050 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0084 decode.acc_seg: 99.7225 aux.loss_ce: 0.0090 aux.acc_seg: 99.2895 +04/16 23:33:00 - mmengine - INFO - Iter(train) [ 55250/160000] base_lr: 6.6089e-05 lr: 2.4434e-07 eta: 1 day, 5:02:28 time: 0.9970 data_time: 0.0047 memory: 8462 loss: 0.0180 decode.loss_ce: 0.0091 decode.acc_seg: 99.6620 aux.loss_ce: 0.0089 aux.acc_seg: 99.2310 +04/16 23:33:50 - mmengine - INFO - Iter(train) [ 55300/160000] base_lr: 6.6057e-05 lr: 2.4423e-07 eta: 1 day, 5:01:38 time: 0.9961 data_time: 0.0053 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0071 decode.acc_seg: 99.6387 aux.loss_ce: 0.0078 aux.acc_seg: 99.0450 +04/16 23:34:40 - mmengine - INFO - Iter(train) [ 55350/160000] base_lr: 6.6026e-05 lr: 2.4411e-07 eta: 1 day, 5:00:48 time: 0.9964 data_time: 0.0050 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0071 decode.acc_seg: 99.6101 aux.loss_ce: 0.0076 aux.acc_seg: 98.9889 +04/16 23:35:29 - mmengine - INFO - Iter(train) [ 55400/160000] base_lr: 6.5994e-05 lr: 2.4399e-07 eta: 1 day, 4:59:58 time: 0.9974 data_time: 0.0050 memory: 8462 loss: 0.0162 decode.loss_ce: 0.0078 decode.acc_seg: 99.6679 aux.loss_ce: 0.0084 aux.acc_seg: 99.0776 +04/16 23:36:19 - mmengine - INFO - Iter(train) [ 55450/160000] base_lr: 6.5963e-05 lr: 2.4388e-07 eta: 1 day, 4:59:08 time: 0.9961 data_time: 0.0049 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0076 decode.acc_seg: 99.6929 aux.loss_ce: 0.0084 aux.acc_seg: 99.0248 +04/16 23:37:09 - mmengine - INFO - Iter(train) [ 55500/160000] base_lr: 6.5931e-05 lr: 2.4376e-07 eta: 1 day, 4:58:18 time: 0.9954 data_time: 0.0044 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0085 decode.acc_seg: 99.5415 aux.loss_ce: 0.0085 aux.acc_seg: 99.0578 +04/16 23:37:59 - mmengine - INFO - Iter(train) [ 55550/160000] base_lr: 6.5899e-05 lr: 2.4364e-07 eta: 1 day, 4:57:27 time: 0.9950 data_time: 0.0047 memory: 8462 loss: 0.0162 decode.loss_ce: 0.0080 decode.acc_seg: 99.6861 aux.loss_ce: 0.0082 aux.acc_seg: 99.3368 +04/16 23:38:49 - mmengine - INFO - Iter(train) [ 55600/160000] base_lr: 6.5868e-05 lr: 2.4353e-07 eta: 1 day, 4:56:37 time: 0.9957 data_time: 0.0044 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0084 decode.acc_seg: 99.6862 aux.loss_ce: 0.0086 aux.acc_seg: 99.0326 +04/16 23:39:39 - mmengine - INFO - Iter(train) [ 55650/160000] base_lr: 6.5836e-05 lr: 2.4341e-07 eta: 1 day, 4:55:47 time: 0.9968 data_time: 0.0047 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0068 decode.acc_seg: 99.5714 aux.loss_ce: 0.0075 aux.acc_seg: 98.8604 +04/16 23:40:28 - mmengine - INFO - Iter(train) [ 55700/160000] base_lr: 6.5805e-05 lr: 2.4329e-07 eta: 1 day, 4:54:57 time: 0.9953 data_time: 0.0044 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0076 decode.acc_seg: 99.4999 aux.loss_ce: 0.0078 aux.acc_seg: 99.0545 +04/16 23:41:18 - mmengine - INFO - Iter(train) [ 55750/160000] base_lr: 6.5773e-05 lr: 2.4318e-07 eta: 1 day, 4:54:07 time: 0.9971 data_time: 0.0046 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0073 decode.acc_seg: 99.7549 aux.loss_ce: 0.0083 aux.acc_seg: 99.0911 +04/16 23:42:08 - mmengine - INFO - Iter(train) [ 55800/160000] base_lr: 6.5742e-05 lr: 2.4306e-07 eta: 1 day, 4:53:17 time: 0.9959 data_time: 0.0046 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0081 decode.acc_seg: 99.7007 aux.loss_ce: 0.0089 aux.acc_seg: 99.1953 +04/16 23:42:58 - mmengine - INFO - Iter(train) [ 55850/160000] base_lr: 6.5710e-05 lr: 2.4294e-07 eta: 1 day, 4:52:27 time: 0.9958 data_time: 0.0049 memory: 8462 loss: 0.0171 decode.loss_ce: 0.0080 decode.acc_seg: 99.6983 aux.loss_ce: 0.0090 aux.acc_seg: 99.0719 +04/16 23:43:48 - mmengine - INFO - Iter(train) [ 55900/160000] base_lr: 6.5679e-05 lr: 2.4283e-07 eta: 1 day, 4:51:37 time: 0.9956 data_time: 0.0045 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0092 decode.acc_seg: 99.6111 aux.loss_ce: 0.0089 aux.acc_seg: 99.0194 +04/16 23:44:37 - mmengine - INFO - Iter(train) [ 55950/160000] base_lr: 6.5647e-05 lr: 2.4271e-07 eta: 1 day, 4:50:47 time: 0.9981 data_time: 0.0049 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0076 decode.acc_seg: 99.6408 aux.loss_ce: 0.0078 aux.acc_seg: 98.8640 +04/16 23:45:27 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/16 23:45:27 - mmengine - INFO - Iter(train) [ 56000/160000] base_lr: 6.5616e-05 lr: 2.4259e-07 eta: 1 day, 4:49:57 time: 0.9962 data_time: 0.0048 memory: 8462 loss: 0.0163 decode.loss_ce: 0.0077 decode.acc_seg: 99.8175 aux.loss_ce: 0.0086 aux.acc_seg: 99.3641 +04/16 23:46:17 - mmengine - INFO - Iter(train) [ 56050/160000] base_lr: 6.5584e-05 lr: 2.4248e-07 eta: 1 day, 4:49:06 time: 0.9973 data_time: 0.0057 memory: 8462 loss: 0.0163 decode.loss_ce: 0.0079 decode.acc_seg: 99.6599 aux.loss_ce: 0.0084 aux.acc_seg: 99.1657 +04/16 23:47:07 - mmengine - INFO - Iter(train) [ 56100/160000] base_lr: 6.5552e-05 lr: 2.4236e-07 eta: 1 day, 4:48:16 time: 0.9959 data_time: 0.0048 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0082 decode.acc_seg: 99.6529 aux.loss_ce: 0.0096 aux.acc_seg: 98.9407 +04/16 23:47:57 - mmengine - INFO - Iter(train) [ 56150/160000] base_lr: 6.5521e-05 lr: 2.4224e-07 eta: 1 day, 4:47:26 time: 0.9966 data_time: 0.0049 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0078 decode.acc_seg: 99.7486 aux.loss_ce: 0.0080 aux.acc_seg: 99.3921 +04/16 23:48:46 - mmengine - INFO - Iter(train) [ 56200/160000] base_lr: 6.5489e-05 lr: 2.4213e-07 eta: 1 day, 4:46:36 time: 0.9985 data_time: 0.0051 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0090 decode.acc_seg: 99.6469 aux.loss_ce: 0.0081 aux.acc_seg: 99.1430 +04/16 23:49:36 - mmengine - INFO - Iter(train) [ 56250/160000] base_lr: 6.5458e-05 lr: 2.4201e-07 eta: 1 day, 4:45:46 time: 0.9964 data_time: 0.0045 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0075 decode.acc_seg: 99.6906 aux.loss_ce: 0.0079 aux.acc_seg: 99.3254 +04/16 23:50:26 - mmengine - INFO - Iter(train) [ 56300/160000] base_lr: 6.5426e-05 lr: 2.4189e-07 eta: 1 day, 4:44:56 time: 0.9963 data_time: 0.0048 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0077 decode.acc_seg: 99.6508 aux.loss_ce: 0.0083 aux.acc_seg: 99.0313 +04/16 23:51:16 - mmengine - INFO - Iter(train) [ 56350/160000] base_lr: 6.5395e-05 lr: 2.4178e-07 eta: 1 day, 4:44:06 time: 0.9965 data_time: 0.0045 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0086 decode.acc_seg: 99.6262 aux.loss_ce: 0.0087 aux.acc_seg: 99.1371 +04/16 23:52:06 - mmengine - INFO - Iter(train) [ 56400/160000] base_lr: 6.5363e-05 lr: 2.4166e-07 eta: 1 day, 4:43:16 time: 0.9954 data_time: 0.0045 memory: 8462 loss: 0.0164 decode.loss_ce: 0.0083 decode.acc_seg: 99.7545 aux.loss_ce: 0.0081 aux.acc_seg: 99.4267 +04/16 23:52:55 - mmengine - INFO - Iter(train) [ 56450/160000] base_lr: 6.5332e-05 lr: 2.4154e-07 eta: 1 day, 4:42:26 time: 0.9967 data_time: 0.0050 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0086 decode.acc_seg: 99.5520 aux.loss_ce: 0.0082 aux.acc_seg: 99.0448 +04/16 23:53:45 - mmengine - INFO - Iter(train) [ 56500/160000] base_lr: 6.5300e-05 lr: 2.4143e-07 eta: 1 day, 4:41:36 time: 0.9960 data_time: 0.0047 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0077 decode.acc_seg: 99.7601 aux.loss_ce: 0.0084 aux.acc_seg: 99.1243 +04/16 23:54:35 - mmengine - INFO - Iter(train) [ 56550/160000] base_lr: 6.5269e-05 lr: 2.4131e-07 eta: 1 day, 4:40:46 time: 0.9969 data_time: 0.0051 memory: 8462 loss: 0.0162 decode.loss_ce: 0.0085 decode.acc_seg: 99.2119 aux.loss_ce: 0.0077 aux.acc_seg: 98.3730 +04/16 23:55:25 - mmengine - INFO - Iter(train) [ 56600/160000] base_lr: 6.5237e-05 lr: 2.4119e-07 eta: 1 day, 4:39:56 time: 0.9976 data_time: 0.0049 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0077 decode.acc_seg: 99.7047 aux.loss_ce: 0.0078 aux.acc_seg: 99.2519 +04/16 23:56:15 - mmengine - INFO - Iter(train) [ 56650/160000] base_lr: 6.5205e-05 lr: 2.4108e-07 eta: 1 day, 4:39:06 time: 0.9964 data_time: 0.0049 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0084 decode.acc_seg: 99.4806 aux.loss_ce: 0.0089 aux.acc_seg: 98.9159 +04/16 23:57:05 - mmengine - INFO - Iter(train) [ 56700/160000] base_lr: 6.5174e-05 lr: 2.4096e-07 eta: 1 day, 4:38:16 time: 0.9960 data_time: 0.0052 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0074 decode.acc_seg: 99.7871 aux.loss_ce: 0.0080 aux.acc_seg: 99.4078 +04/16 23:57:54 - mmengine - INFO - Iter(train) [ 56750/160000] base_lr: 6.5142e-05 lr: 2.4084e-07 eta: 1 day, 4:37:26 time: 0.9954 data_time: 0.0047 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0086 decode.acc_seg: 99.5737 aux.loss_ce: 0.0095 aux.acc_seg: 99.0189 +04/16 23:58:44 - mmengine - INFO - Iter(train) [ 56800/160000] base_lr: 6.5111e-05 lr: 2.4073e-07 eta: 1 day, 4:36:36 time: 0.9968 data_time: 0.0047 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0076 decode.acc_seg: 99.6952 aux.loss_ce: 0.0083 aux.acc_seg: 99.1755 +04/16 23:59:34 - mmengine - INFO - Iter(train) [ 56850/160000] base_lr: 6.5079e-05 lr: 2.4061e-07 eta: 1 day, 4:35:45 time: 0.9952 data_time: 0.0046 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0071 decode.acc_seg: 99.5892 aux.loss_ce: 0.0079 aux.acc_seg: 98.5762 +04/17 00:00:24 - mmengine - INFO - Iter(train) [ 56900/160000] base_lr: 6.5048e-05 lr: 2.4049e-07 eta: 1 day, 4:34:55 time: 0.9960 data_time: 0.0048 memory: 8462 loss: 0.0171 decode.loss_ce: 0.0089 decode.acc_seg: 99.7309 aux.loss_ce: 0.0082 aux.acc_seg: 99.1968 +04/17 00:01:14 - mmengine - INFO - Iter(train) [ 56950/160000] base_lr: 6.5016e-05 lr: 2.4038e-07 eta: 1 day, 4:34:05 time: 0.9963 data_time: 0.0048 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0087 decode.acc_seg: 99.6555 aux.loss_ce: 0.0080 aux.acc_seg: 99.2270 +04/17 00:02:03 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/17 00:02:03 - mmengine - INFO - Iter(train) [ 57000/160000] base_lr: 6.4985e-05 lr: 2.4026e-07 eta: 1 day, 4:33:15 time: 0.9961 data_time: 0.0047 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0087 decode.acc_seg: 99.5787 aux.loss_ce: 0.0082 aux.acc_seg: 99.1844 +04/17 00:02:53 - mmengine - INFO - Iter(train) [ 57050/160000] base_lr: 6.4953e-05 lr: 2.4014e-07 eta: 1 day, 4:32:25 time: 0.9963 data_time: 0.0046 memory: 8462 loss: 0.0159 decode.loss_ce: 0.0075 decode.acc_seg: 99.6214 aux.loss_ce: 0.0084 aux.acc_seg: 98.8850 +04/17 00:03:43 - mmengine - INFO - Iter(train) [ 57100/160000] base_lr: 6.4922e-05 lr: 2.4003e-07 eta: 1 day, 4:31:35 time: 0.9960 data_time: 0.0048 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0085 decode.acc_seg: 99.7746 aux.loss_ce: 0.0090 aux.acc_seg: 99.2683 +04/17 00:04:33 - mmengine - INFO - Iter(train) [ 57150/160000] base_lr: 6.4890e-05 lr: 2.3991e-07 eta: 1 day, 4:30:45 time: 0.9953 data_time: 0.0048 memory: 8462 loss: 0.0169 decode.loss_ce: 0.0079 decode.acc_seg: 99.5529 aux.loss_ce: 0.0090 aux.acc_seg: 98.8056 +04/17 00:05:23 - mmengine - INFO - Iter(train) [ 57200/160000] base_lr: 6.4858e-05 lr: 2.3980e-07 eta: 1 day, 4:29:55 time: 0.9963 data_time: 0.0047 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0081 decode.acc_seg: 99.6891 aux.loss_ce: 0.0085 aux.acc_seg: 99.3155 +04/17 00:06:13 - mmengine - INFO - Iter(train) [ 57250/160000] base_lr: 6.4827e-05 lr: 2.3968e-07 eta: 1 day, 4:29:05 time: 0.9957 data_time: 0.0048 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0077 decode.acc_seg: 99.6056 aux.loss_ce: 0.0079 aux.acc_seg: 98.8895 +04/17 00:07:02 - mmengine - INFO - Iter(train) [ 57300/160000] base_lr: 6.4795e-05 lr: 2.3956e-07 eta: 1 day, 4:28:15 time: 0.9955 data_time: 0.0044 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0077 decode.acc_seg: 99.7478 aux.loss_ce: 0.0081 aux.acc_seg: 99.4514 +04/17 00:07:52 - mmengine - INFO - Iter(train) [ 57350/160000] base_lr: 6.4764e-05 lr: 2.3945e-07 eta: 1 day, 4:27:25 time: 0.9959 data_time: 0.0043 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0073 decode.acc_seg: 99.5729 aux.loss_ce: 0.0080 aux.acc_seg: 99.1472 +04/17 00:08:42 - mmengine - INFO - Iter(train) [ 57400/160000] base_lr: 6.4732e-05 lr: 2.3933e-07 eta: 1 day, 4:26:35 time: 0.9952 data_time: 0.0044 memory: 8462 loss: 0.0164 decode.loss_ce: 0.0082 decode.acc_seg: 99.7618 aux.loss_ce: 0.0082 aux.acc_seg: 99.1835 +04/17 00:09:32 - mmengine - INFO - Iter(train) [ 57450/160000] base_lr: 6.4701e-05 lr: 2.3921e-07 eta: 1 day, 4:25:45 time: 0.9958 data_time: 0.0045 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0083 decode.acc_seg: 99.6758 aux.loss_ce: 0.0085 aux.acc_seg: 99.1816 +04/17 00:10:22 - mmengine - INFO - Iter(train) [ 57500/160000] base_lr: 6.4669e-05 lr: 2.3910e-07 eta: 1 day, 4:24:54 time: 0.9954 data_time: 0.0048 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0082 decode.acc_seg: 99.6456 aux.loss_ce: 0.0088 aux.acc_seg: 98.8880 +04/17 00:11:11 - mmengine - INFO - Iter(train) [ 57550/160000] base_lr: 6.4638e-05 lr: 2.3898e-07 eta: 1 day, 4:24:04 time: 0.9959 data_time: 0.0043 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0089 decode.acc_seg: 99.5880 aux.loss_ce: 0.0086 aux.acc_seg: 99.0543 +04/17 00:12:01 - mmengine - INFO - Iter(train) [ 57600/160000] base_lr: 6.4606e-05 lr: 2.3886e-07 eta: 1 day, 4:23:14 time: 0.9957 data_time: 0.0045 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0074 decode.acc_seg: 99.5525 aux.loss_ce: 0.0086 aux.acc_seg: 98.7368 +04/17 00:12:51 - mmengine - INFO - Iter(train) [ 57650/160000] base_lr: 6.4575e-05 lr: 2.3875e-07 eta: 1 day, 4:22:24 time: 0.9967 data_time: 0.0051 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0075 decode.acc_seg: 99.6441 aux.loss_ce: 0.0077 aux.acc_seg: 99.0360 +04/17 00:13:41 - mmengine - INFO - Iter(train) [ 57700/160000] base_lr: 6.4543e-05 lr: 2.3863e-07 eta: 1 day, 4:21:34 time: 0.9959 data_time: 0.0046 memory: 8462 loss: 0.0180 decode.loss_ce: 0.0085 decode.acc_seg: 99.6099 aux.loss_ce: 0.0095 aux.acc_seg: 99.0791 +04/17 00:14:31 - mmengine - INFO - Iter(train) [ 57750/160000] base_lr: 6.4511e-05 lr: 2.3851e-07 eta: 1 day, 4:20:44 time: 0.9956 data_time: 0.0049 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0078 decode.acc_seg: 99.6040 aux.loss_ce: 0.0093 aux.acc_seg: 98.7823 +04/17 00:15:21 - mmengine - INFO - Iter(train) [ 57800/160000] base_lr: 6.4480e-05 lr: 2.3840e-07 eta: 1 day, 4:19:54 time: 0.9972 data_time: 0.0046 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0080 decode.acc_seg: 99.6916 aux.loss_ce: 0.0081 aux.acc_seg: 99.2733 +04/17 00:16:10 - mmengine - INFO - Iter(train) [ 57850/160000] base_lr: 6.4448e-05 lr: 2.3828e-07 eta: 1 day, 4:19:04 time: 0.9961 data_time: 0.0048 memory: 8462 loss: 0.0139 decode.loss_ce: 0.0066 decode.acc_seg: 99.6283 aux.loss_ce: 0.0073 aux.acc_seg: 99.1034 +04/17 00:17:00 - mmengine - INFO - Iter(train) [ 57900/160000] base_lr: 6.4417e-05 lr: 2.3816e-07 eta: 1 day, 4:18:14 time: 0.9969 data_time: 0.0050 memory: 8462 loss: 0.0162 decode.loss_ce: 0.0076 decode.acc_seg: 99.6788 aux.loss_ce: 0.0085 aux.acc_seg: 99.1293 +04/17 00:17:50 - mmengine - INFO - Iter(train) [ 57950/160000] base_lr: 6.4385e-05 lr: 2.3805e-07 eta: 1 day, 4:17:24 time: 0.9963 data_time: 0.0046 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0073 decode.acc_seg: 99.6782 aux.loss_ce: 0.0076 aux.acc_seg: 99.3078 +04/17 00:18:40 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/17 00:18:40 - mmengine - INFO - Iter(train) [ 58000/160000] base_lr: 6.4354e-05 lr: 2.3793e-07 eta: 1 day, 4:16:34 time: 0.9965 data_time: 0.0046 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0076 decode.acc_seg: 99.6408 aux.loss_ce: 0.0080 aux.acc_seg: 98.8129 +04/17 00:19:30 - mmengine - INFO - Iter(train) [ 58050/160000] base_lr: 6.4322e-05 lr: 2.3781e-07 eta: 1 day, 4:15:44 time: 0.9969 data_time: 0.0045 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0069 decode.acc_seg: 99.7543 aux.loss_ce: 0.0080 aux.acc_seg: 99.2516 +04/17 00:20:20 - mmengine - INFO - Iter(train) [ 58100/160000] base_lr: 6.4291e-05 lr: 2.3770e-07 eta: 1 day, 4:14:54 time: 0.9957 data_time: 0.0044 memory: 8462 loss: 0.0137 decode.loss_ce: 0.0063 decode.acc_seg: 99.7808 aux.loss_ce: 0.0074 aux.acc_seg: 99.3057 +04/17 00:21:09 - mmengine - INFO - Iter(train) [ 58150/160000] base_lr: 6.4259e-05 lr: 2.3758e-07 eta: 1 day, 4:14:04 time: 0.9984 data_time: 0.0047 memory: 8462 loss: 0.0162 decode.loss_ce: 0.0080 decode.acc_seg: 99.7034 aux.loss_ce: 0.0083 aux.acc_seg: 99.1354 +04/17 00:21:59 - mmengine - INFO - Iter(train) [ 58200/160000] base_lr: 6.4228e-05 lr: 2.3746e-07 eta: 1 day, 4:13:14 time: 0.9967 data_time: 0.0044 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0065 decode.acc_seg: 99.6964 aux.loss_ce: 0.0075 aux.acc_seg: 99.0650 +04/17 00:22:49 - mmengine - INFO - Iter(train) [ 58250/160000] base_lr: 6.4196e-05 lr: 2.3735e-07 eta: 1 day, 4:12:24 time: 0.9960 data_time: 0.0049 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0070 decode.acc_seg: 99.7053 aux.loss_ce: 0.0073 aux.acc_seg: 99.3534 +04/17 00:23:39 - mmengine - INFO - Iter(train) [ 58300/160000] base_lr: 6.4164e-05 lr: 2.3723e-07 eta: 1 day, 4:11:34 time: 0.9957 data_time: 0.0044 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0074 decode.acc_seg: 99.6477 aux.loss_ce: 0.0087 aux.acc_seg: 99.0463 +04/17 00:24:29 - mmengine - INFO - Iter(train) [ 58350/160000] base_lr: 6.4133e-05 lr: 2.3711e-07 eta: 1 day, 4:10:44 time: 0.9968 data_time: 0.0048 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0083 decode.acc_seg: 99.6548 aux.loss_ce: 0.0078 aux.acc_seg: 99.2752 +04/17 00:25:19 - mmengine - INFO - Iter(train) [ 58400/160000] base_lr: 6.4101e-05 lr: 2.3700e-07 eta: 1 day, 4:09:54 time: 0.9961 data_time: 0.0050 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0072 decode.acc_seg: 99.6813 aux.loss_ce: 0.0078 aux.acc_seg: 98.8253 +04/17 00:26:08 - mmengine - INFO - Iter(train) [ 58450/160000] base_lr: 6.4070e-05 lr: 2.3688e-07 eta: 1 day, 4:09:04 time: 0.9945 data_time: 0.0044 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0085 decode.acc_seg: 99.6655 aux.loss_ce: 0.0091 aux.acc_seg: 99.2115 +04/17 00:26:58 - mmengine - INFO - Iter(train) [ 58500/160000] base_lr: 6.4038e-05 lr: 2.3676e-07 eta: 1 day, 4:08:14 time: 0.9963 data_time: 0.0044 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0067 decode.acc_seg: 99.6471 aux.loss_ce: 0.0073 aux.acc_seg: 99.4682 +04/17 00:27:48 - mmengine - INFO - Iter(train) [ 58550/160000] base_lr: 6.4007e-05 lr: 2.3665e-07 eta: 1 day, 4:07:24 time: 0.9973 data_time: 0.0045 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0074 decode.acc_seg: 99.7215 aux.loss_ce: 0.0083 aux.acc_seg: 99.2233 +04/17 00:28:38 - mmengine - INFO - Iter(train) [ 58600/160000] base_lr: 6.3975e-05 lr: 2.3653e-07 eta: 1 day, 4:06:34 time: 0.9957 data_time: 0.0046 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0085 decode.acc_seg: 99.6063 aux.loss_ce: 0.0088 aux.acc_seg: 99.2188 +04/17 00:29:28 - mmengine - INFO - Iter(train) [ 58650/160000] base_lr: 6.3944e-05 lr: 2.3641e-07 eta: 1 day, 4:05:44 time: 0.9977 data_time: 0.0047 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0069 decode.acc_seg: 99.5914 aux.loss_ce: 0.0080 aux.acc_seg: 98.8672 +04/17 00:30:18 - mmengine - INFO - Iter(train) [ 58700/160000] base_lr: 6.3912e-05 lr: 2.3630e-07 eta: 1 day, 4:04:54 time: 0.9965 data_time: 0.0046 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0082 decode.acc_seg: 99.6300 aux.loss_ce: 0.0085 aux.acc_seg: 99.0694 +04/17 00:31:07 - mmengine - INFO - Iter(train) [ 58750/160000] base_lr: 6.3881e-05 lr: 2.3618e-07 eta: 1 day, 4:04:04 time: 0.9957 data_time: 0.0047 memory: 8462 loss: 0.0165 decode.loss_ce: 0.0082 decode.acc_seg: 99.5741 aux.loss_ce: 0.0083 aux.acc_seg: 98.8823 +04/17 00:31:57 - mmengine - INFO - Iter(train) [ 58800/160000] base_lr: 6.3849e-05 lr: 2.3606e-07 eta: 1 day, 4:03:14 time: 0.9965 data_time: 0.0047 memory: 8462 loss: 0.0165 decode.loss_ce: 0.0080 decode.acc_seg: 99.6418 aux.loss_ce: 0.0085 aux.acc_seg: 99.1936 +04/17 00:32:47 - mmengine - INFO - Iter(train) [ 58850/160000] base_lr: 6.3817e-05 lr: 2.3595e-07 eta: 1 day, 4:02:24 time: 0.9979 data_time: 0.0047 memory: 8462 loss: 0.0193 decode.loss_ce: 0.0096 decode.acc_seg: 99.5193 aux.loss_ce: 0.0098 aux.acc_seg: 98.9685 +04/17 00:33:37 - mmengine - INFO - Iter(train) [ 58900/160000] base_lr: 6.3786e-05 lr: 2.3583e-07 eta: 1 day, 4:01:34 time: 0.9960 data_time: 0.0050 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0073 decode.acc_seg: 99.6208 aux.loss_ce: 0.0082 aux.acc_seg: 99.2405 +04/17 00:34:27 - mmengine - INFO - Iter(train) [ 58950/160000] base_lr: 6.3754e-05 lr: 2.3571e-07 eta: 1 day, 4:00:44 time: 0.9971 data_time: 0.0047 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0091 decode.acc_seg: 99.7253 aux.loss_ce: 0.0087 aux.acc_seg: 99.2334 +04/17 00:35:17 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/17 00:35:17 - mmengine - INFO - Iter(train) [ 59000/160000] base_lr: 6.3723e-05 lr: 2.3560e-07 eta: 1 day, 3:59:54 time: 0.9960 data_time: 0.0047 memory: 8462 loss: 0.0180 decode.loss_ce: 0.0089 decode.acc_seg: 99.6449 aux.loss_ce: 0.0090 aux.acc_seg: 99.1611 +04/17 00:36:06 - mmengine - INFO - Iter(train) [ 59050/160000] base_lr: 6.3691e-05 lr: 2.3548e-07 eta: 1 day, 3:59:04 time: 0.9965 data_time: 0.0047 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0080 decode.acc_seg: 99.6113 aux.loss_ce: 0.0092 aux.acc_seg: 99.0860 +04/17 00:36:56 - mmengine - INFO - Iter(train) [ 59100/160000] base_lr: 6.3660e-05 lr: 2.3536e-07 eta: 1 day, 3:58:14 time: 0.9971 data_time: 0.0048 memory: 8462 loss: 0.0151 decode.loss_ce: 0.0074 decode.acc_seg: 99.7328 aux.loss_ce: 0.0077 aux.acc_seg: 99.1354 +04/17 00:37:46 - mmengine - INFO - Iter(train) [ 59150/160000] base_lr: 6.3628e-05 lr: 2.3525e-07 eta: 1 day, 3:57:24 time: 0.9970 data_time: 0.0048 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0080 decode.acc_seg: 99.4686 aux.loss_ce: 0.0085 aux.acc_seg: 98.6643 +04/17 00:38:36 - mmengine - INFO - Iter(train) [ 59200/160000] base_lr: 6.3597e-05 lr: 2.3513e-07 eta: 1 day, 3:56:34 time: 0.9957 data_time: 0.0044 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0074 decode.acc_seg: 99.6296 aux.loss_ce: 0.0084 aux.acc_seg: 99.1884 +04/17 00:39:26 - mmengine - INFO - Iter(train) [ 59250/160000] base_lr: 6.3565e-05 lr: 2.3501e-07 eta: 1 day, 3:55:44 time: 0.9964 data_time: 0.0046 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0081 decode.acc_seg: 99.5554 aux.loss_ce: 0.0089 aux.acc_seg: 98.7679 +04/17 00:40:16 - mmengine - INFO - Iter(train) [ 59300/160000] base_lr: 6.3534e-05 lr: 2.3490e-07 eta: 1 day, 3:54:54 time: 0.9971 data_time: 0.0047 memory: 8462 loss: 0.0159 decode.loss_ce: 0.0075 decode.acc_seg: 99.6595 aux.loss_ce: 0.0084 aux.acc_seg: 99.4104 +04/17 00:41:06 - mmengine - INFO - Iter(train) [ 59350/160000] base_lr: 6.3502e-05 lr: 2.3478e-07 eta: 1 day, 3:54:04 time: 0.9968 data_time: 0.0050 memory: 8462 loss: 0.0145 decode.loss_ce: 0.0073 decode.acc_seg: 99.6786 aux.loss_ce: 0.0072 aux.acc_seg: 99.0585 +04/17 00:41:55 - mmengine - INFO - Iter(train) [ 59400/160000] base_lr: 6.3470e-05 lr: 2.3466e-07 eta: 1 day, 3:53:14 time: 0.9970 data_time: 0.0047 memory: 8462 loss: 0.0164 decode.loss_ce: 0.0081 decode.acc_seg: 99.4442 aux.loss_ce: 0.0083 aux.acc_seg: 98.7026 +04/17 00:42:45 - mmengine - INFO - Iter(train) [ 59450/160000] base_lr: 6.3439e-05 lr: 2.3455e-07 eta: 1 day, 3:52:24 time: 0.9978 data_time: 0.0047 memory: 8462 loss: 0.0159 decode.loss_ce: 0.0078 decode.acc_seg: 99.6820 aux.loss_ce: 0.0081 aux.acc_seg: 99.5205 +04/17 00:43:35 - mmengine - INFO - Iter(train) [ 59500/160000] base_lr: 6.3407e-05 lr: 2.3443e-07 eta: 1 day, 3:51:34 time: 0.9974 data_time: 0.0046 memory: 8462 loss: 0.0163 decode.loss_ce: 0.0078 decode.acc_seg: 99.6511 aux.loss_ce: 0.0086 aux.acc_seg: 98.9683 +04/17 00:44:25 - mmengine - INFO - Iter(train) [ 59550/160000] base_lr: 6.3376e-05 lr: 2.3431e-07 eta: 1 day, 3:50:44 time: 0.9973 data_time: 0.0046 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0083 decode.acc_seg: 99.6803 aux.loss_ce: 0.0089 aux.acc_seg: 99.1005 +04/17 00:45:15 - mmengine - INFO - Iter(train) [ 59600/160000] base_lr: 6.3344e-05 lr: 2.3420e-07 eta: 1 day, 3:49:54 time: 0.9972 data_time: 0.0045 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0082 decode.acc_seg: 99.7608 aux.loss_ce: 0.0084 aux.acc_seg: 99.3568 +04/17 00:46:05 - mmengine - INFO - Iter(train) [ 59650/160000] base_lr: 6.3313e-05 lr: 2.3408e-07 eta: 1 day, 3:49:04 time: 0.9969 data_time: 0.0043 memory: 8462 loss: 0.0164 decode.loss_ce: 0.0078 decode.acc_seg: 99.6710 aux.loss_ce: 0.0086 aux.acc_seg: 99.2632 +04/17 00:46:55 - mmengine - INFO - Iter(train) [ 59700/160000] base_lr: 6.3281e-05 lr: 2.3396e-07 eta: 1 day, 3:48:14 time: 0.9961 data_time: 0.0045 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0085 decode.acc_seg: 99.6861 aux.loss_ce: 0.0070 aux.acc_seg: 99.3183 +04/17 00:47:45 - mmengine - INFO - Iter(train) [ 59750/160000] base_lr: 6.3250e-05 lr: 2.3385e-07 eta: 1 day, 3:47:24 time: 0.9978 data_time: 0.0044 memory: 8462 loss: 0.0162 decode.loss_ce: 0.0081 decode.acc_seg: 99.5653 aux.loss_ce: 0.0081 aux.acc_seg: 98.7288 +04/17 00:48:34 - mmengine - INFO - Iter(train) [ 59800/160000] base_lr: 6.3218e-05 lr: 2.3373e-07 eta: 1 day, 3:46:34 time: 0.9973 data_time: 0.0048 memory: 8462 loss: 0.0145 decode.loss_ce: 0.0071 decode.acc_seg: 99.8343 aux.loss_ce: 0.0074 aux.acc_seg: 99.6162 +04/17 00:49:24 - mmengine - INFO - Iter(train) [ 59850/160000] base_lr: 6.3187e-05 lr: 2.3361e-07 eta: 1 day, 3:45:44 time: 0.9963 data_time: 0.0045 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0089 decode.acc_seg: 99.6489 aux.loss_ce: 0.0092 aux.acc_seg: 98.8594 +04/17 00:50:14 - mmengine - INFO - Iter(train) [ 59900/160000] base_lr: 6.3155e-05 lr: 2.3350e-07 eta: 1 day, 3:44:55 time: 0.9980 data_time: 0.0047 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0086 decode.acc_seg: 99.6519 aux.loss_ce: 0.0081 aux.acc_seg: 99.2622 +04/17 00:51:04 - mmengine - INFO - Iter(train) [ 59950/160000] base_lr: 6.3123e-05 lr: 2.3338e-07 eta: 1 day, 3:44:05 time: 0.9964 data_time: 0.0045 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0089 decode.acc_seg: 99.4581 aux.loss_ce: 0.0084 aux.acc_seg: 98.9388 +04/17 00:51:54 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/17 00:51:54 - mmengine - INFO - Iter(train) [ 60000/160000] base_lr: 6.3092e-05 lr: 2.3326e-07 eta: 1 day, 3:43:15 time: 0.9968 data_time: 0.0048 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0087 decode.acc_seg: 99.7086 aux.loss_ce: 0.0086 aux.acc_seg: 99.1188 +04/17 00:51:54 - mmengine - INFO - Saving checkpoint at 60000 iterations +04/17 00:52:04 - mmengine - INFO - Iter(val) [ 50/355] eta: 0:00:35 time: 0.1150 data_time: 0.0014 memory: 4004 +04/17 00:52:10 - mmengine - INFO - Iter(val) [100/355] eta: 0:00:29 time: 0.1152 data_time: 0.0015 memory: 4004 +04/17 00:52:15 - mmengine - INFO - Iter(val) [150/355] eta: 0:00:23 time: 0.1155 data_time: 0.0016 memory: 4004 +04/17 00:52:21 - mmengine - INFO - Iter(val) [200/355] eta: 0:00:17 time: 0.1155 data_time: 0.0014 memory: 4004 +04/17 00:52:27 - mmengine - INFO - Iter(val) [250/355] eta: 0:00:12 time: 0.1157 data_time: 0.0014 memory: 4004 +04/17 00:52:33 - mmengine - INFO - Iter(val) [300/355] eta: 0:00:06 time: 0.1160 data_time: 0.0020 memory: 4004 +04/17 00:52:39 - mmengine - INFO - Iter(val) [350/355] eta: 0:00:00 time: 0.1156 data_time: 0.0015 memory: 4004 +04/17 00:52:40 - mmengine - INFO - per class results: +04/17 00:52:40 - mmengine - INFO - ++------------+-------+-------+ +| Class | IoU | Acc | ++------------+-------+-------+ +| background | 99.19 | 99.61 | +| contrast | 82.27 | 89.84 | ++------------+-------+-------+ +04/17 00:52:40 - mmengine - INFO - Iter(val) [355/355] aAcc: 99.2200 mIoU: 90.7300 mAcc: 94.7200 data_time: 0.0015 time: 0.1157 +04/17 00:53:29 - mmengine - INFO - Iter(train) [ 60050/160000] base_lr: 6.3060e-05 lr: 2.3315e-07 eta: 1 day, 3:42:25 time: 0.9983 data_time: 0.0053 memory: 8462 loss: 0.0182 decode.loss_ce: 0.0087 decode.acc_seg: 99.7343 aux.loss_ce: 0.0095 aux.acc_seg: 99.2016 +04/17 00:54:19 - mmengine - INFO - Iter(train) [ 60100/160000] base_lr: 6.3029e-05 lr: 2.3303e-07 eta: 1 day, 3:41:35 time: 0.9990 data_time: 0.0048 memory: 8462 loss: 0.0138 decode.loss_ce: 0.0068 decode.acc_seg: 99.5863 aux.loss_ce: 0.0071 aux.acc_seg: 99.0122 +04/17 00:55:09 - mmengine - INFO - Iter(train) [ 60150/160000] base_lr: 6.2997e-05 lr: 2.3291e-07 eta: 1 day, 3:40:45 time: 0.9980 data_time: 0.0048 memory: 8462 loss: 0.0164 decode.loss_ce: 0.0079 decode.acc_seg: 99.7379 aux.loss_ce: 0.0085 aux.acc_seg: 99.2889 +04/17 00:55:59 - mmengine - INFO - Iter(train) [ 60200/160000] base_lr: 6.2966e-05 lr: 2.3280e-07 eta: 1 day, 3:39:56 time: 0.9978 data_time: 0.0049 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0078 decode.acc_seg: 99.6826 aux.loss_ce: 0.0080 aux.acc_seg: 99.0971 +04/17 00:56:49 - mmengine - INFO - Iter(train) [ 60250/160000] base_lr: 6.2934e-05 lr: 2.3268e-07 eta: 1 day, 3:39:06 time: 0.9988 data_time: 0.0053 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0074 decode.acc_seg: 99.6826 aux.loss_ce: 0.0084 aux.acc_seg: 98.9386 +04/17 00:57:39 - mmengine - INFO - Iter(train) [ 60300/160000] base_lr: 6.2903e-05 lr: 2.3256e-07 eta: 1 day, 3:38:16 time: 0.9983 data_time: 0.0051 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0077 decode.acc_seg: 99.7467 aux.loss_ce: 0.0084 aux.acc_seg: 99.3000 +04/17 00:58:29 - mmengine - INFO - Iter(train) [ 60350/160000] base_lr: 6.2871e-05 lr: 2.3245e-07 eta: 1 day, 3:37:26 time: 0.9971 data_time: 0.0047 memory: 8462 loss: 0.0164 decode.loss_ce: 0.0079 decode.acc_seg: 99.7631 aux.loss_ce: 0.0085 aux.acc_seg: 99.4083 +04/17 00:59:19 - mmengine - INFO - Iter(train) [ 60400/160000] base_lr: 6.2840e-05 lr: 2.3233e-07 eta: 1 day, 3:36:36 time: 0.9978 data_time: 0.0050 memory: 8462 loss: 0.0164 decode.loss_ce: 0.0080 decode.acc_seg: 99.6956 aux.loss_ce: 0.0084 aux.acc_seg: 99.2638 +04/17 01:00:09 - mmengine - INFO - Iter(train) [ 60450/160000] base_lr: 6.2808e-05 lr: 2.3221e-07 eta: 1 day, 3:35:46 time: 0.9978 data_time: 0.0046 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0083 decode.acc_seg: 99.6384 aux.loss_ce: 0.0091 aux.acc_seg: 98.8962 +04/17 01:00:59 - mmengine - INFO - Iter(train) [ 60500/160000] base_lr: 6.2776e-05 lr: 2.3210e-07 eta: 1 day, 3:34:56 time: 0.9980 data_time: 0.0051 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0086 decode.acc_seg: 99.6979 aux.loss_ce: 0.0088 aux.acc_seg: 99.1194 +04/17 01:01:49 - mmengine - INFO - Iter(train) [ 60550/160000] base_lr: 6.2745e-05 lr: 2.3198e-07 eta: 1 day, 3:34:07 time: 0.9988 data_time: 0.0051 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0087 decode.acc_seg: 99.5541 aux.loss_ce: 0.0092 aux.acc_seg: 98.9962 +04/17 01:02:39 - mmengine - INFO - Iter(train) [ 60600/160000] base_lr: 6.2713e-05 lr: 2.3186e-07 eta: 1 day, 3:33:17 time: 0.9983 data_time: 0.0046 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0076 decode.acc_seg: 99.6687 aux.loss_ce: 0.0078 aux.acc_seg: 99.1209 +04/17 01:03:29 - mmengine - INFO - Iter(train) [ 60650/160000] base_lr: 6.2682e-05 lr: 2.3175e-07 eta: 1 day, 3:32:27 time: 0.9982 data_time: 0.0050 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0077 decode.acc_seg: 99.6916 aux.loss_ce: 0.0082 aux.acc_seg: 99.3086 +04/17 01:04:18 - mmengine - INFO - Iter(train) [ 60700/160000] base_lr: 6.2650e-05 lr: 2.3163e-07 eta: 1 day, 3:31:37 time: 0.9976 data_time: 0.0046 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0077 decode.acc_seg: 99.6794 aux.loss_ce: 0.0083 aux.acc_seg: 99.2924 +04/17 01:05:08 - mmengine - INFO - Iter(train) [ 60750/160000] base_lr: 6.2619e-05 lr: 2.3151e-07 eta: 1 day, 3:30:47 time: 0.9983 data_time: 0.0050 memory: 8462 loss: 0.0187 decode.loss_ce: 0.0093 decode.acc_seg: 99.6153 aux.loss_ce: 0.0094 aux.acc_seg: 99.0814 +04/17 01:05:58 - mmengine - INFO - Iter(train) [ 60800/160000] base_lr: 6.2587e-05 lr: 2.3140e-07 eta: 1 day, 3:29:57 time: 0.9979 data_time: 0.0051 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0081 decode.acc_seg: 99.6361 aux.loss_ce: 0.0085 aux.acc_seg: 99.0313 +04/17 01:06:48 - mmengine - INFO - Iter(train) [ 60850/160000] base_lr: 6.2556e-05 lr: 2.3128e-07 eta: 1 day, 3:29:07 time: 0.9997 data_time: 0.0048 memory: 8462 loss: 0.0135 decode.loss_ce: 0.0067 decode.acc_seg: 99.7324 aux.loss_ce: 0.0068 aux.acc_seg: 99.2758 +04/17 01:07:38 - mmengine - INFO - Iter(train) [ 60900/160000] base_lr: 6.2524e-05 lr: 2.3116e-07 eta: 1 day, 3:28:18 time: 0.9987 data_time: 0.0052 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0081 decode.acc_seg: 99.5985 aux.loss_ce: 0.0089 aux.acc_seg: 98.8792 +04/17 01:08:28 - mmengine - INFO - Iter(train) [ 60950/160000] base_lr: 6.2493e-05 lr: 2.3105e-07 eta: 1 day, 3:27:28 time: 0.9985 data_time: 0.0049 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0073 decode.acc_seg: 99.6412 aux.loss_ce: 0.0080 aux.acc_seg: 98.9189 +04/17 01:09:18 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/17 01:09:18 - mmengine - INFO - Iter(train) [ 61000/160000] base_lr: 6.2461e-05 lr: 2.3093e-07 eta: 1 day, 3:26:38 time: 0.9987 data_time: 0.0046 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0075 decode.acc_seg: 99.5504 aux.loss_ce: 0.0081 aux.acc_seg: 98.7961 +04/17 01:10:08 - mmengine - INFO - Iter(train) [ 61050/160000] base_lr: 6.2429e-05 lr: 2.3081e-07 eta: 1 day, 3:25:48 time: 0.9991 data_time: 0.0046 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0076 decode.acc_seg: 99.7383 aux.loss_ce: 0.0084 aux.acc_seg: 99.1812 +04/17 01:10:58 - mmengine - INFO - Iter(train) [ 61100/160000] base_lr: 6.2398e-05 lr: 2.3070e-07 eta: 1 day, 3:24:58 time: 0.9997 data_time: 0.0050 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0072 decode.acc_seg: 99.7255 aux.loss_ce: 0.0082 aux.acc_seg: 99.1756 +04/17 01:11:48 - mmengine - INFO - Iter(train) [ 61150/160000] base_lr: 6.2366e-05 lr: 2.3058e-07 eta: 1 day, 3:24:09 time: 0.9997 data_time: 0.0046 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0079 decode.acc_seg: 99.6849 aux.loss_ce: 0.0092 aux.acc_seg: 98.9201 +04/17 01:12:38 - mmengine - INFO - Iter(train) [ 61200/160000] base_lr: 6.2335e-05 lr: 2.3046e-07 eta: 1 day, 3:23:19 time: 0.9984 data_time: 0.0052 memory: 8462 loss: 0.0169 decode.loss_ce: 0.0081 decode.acc_seg: 99.5611 aux.loss_ce: 0.0088 aux.acc_seg: 99.2458 +04/17 01:13:28 - mmengine - INFO - Iter(train) [ 61250/160000] base_lr: 6.2303e-05 lr: 2.3035e-07 eta: 1 day, 3:22:29 time: 0.9987 data_time: 0.0047 memory: 8462 loss: 0.0144 decode.loss_ce: 0.0073 decode.acc_seg: 99.6452 aux.loss_ce: 0.0072 aux.acc_seg: 98.8508 +04/17 01:14:18 - mmengine - INFO - Iter(train) [ 61300/160000] base_lr: 6.2272e-05 lr: 2.3023e-07 eta: 1 day, 3:21:39 time: 0.9986 data_time: 0.0048 memory: 8462 loss: 0.0136 decode.loss_ce: 0.0065 decode.acc_seg: 99.7452 aux.loss_ce: 0.0072 aux.acc_seg: 99.3576 +04/17 01:15:08 - mmengine - INFO - Iter(train) [ 61350/160000] base_lr: 6.2240e-05 lr: 2.3011e-07 eta: 1 day, 3:20:49 time: 0.9990 data_time: 0.0046 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0079 decode.acc_seg: 99.6916 aux.loss_ce: 0.0082 aux.acc_seg: 99.2538 +04/17 01:15:58 - mmengine - INFO - Iter(train) [ 61400/160000] base_lr: 6.2209e-05 lr: 2.3000e-07 eta: 1 day, 3:20:00 time: 0.9994 data_time: 0.0052 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0078 decode.acc_seg: 99.7406 aux.loss_ce: 0.0083 aux.acc_seg: 99.2584 +04/17 01:16:48 - mmengine - INFO - Iter(train) [ 61450/160000] base_lr: 6.2177e-05 lr: 2.2988e-07 eta: 1 day, 3:19:10 time: 0.9998 data_time: 0.0046 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0075 decode.acc_seg: 99.8035 aux.loss_ce: 0.0078 aux.acc_seg: 99.3460 +04/17 01:17:38 - mmengine - INFO - Iter(train) [ 61500/160000] base_lr: 6.2146e-05 lr: 2.2976e-07 eta: 1 day, 3:18:20 time: 0.9998 data_time: 0.0047 memory: 8462 loss: 0.0169 decode.loss_ce: 0.0084 decode.acc_seg: 99.4026 aux.loss_ce: 0.0084 aux.acc_seg: 98.6305 +04/17 01:18:28 - mmengine - INFO - Iter(train) [ 61550/160000] base_lr: 6.2114e-05 lr: 2.2965e-07 eta: 1 day, 3:17:30 time: 1.0007 data_time: 0.0052 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0071 decode.acc_seg: 99.5100 aux.loss_ce: 0.0084 aux.acc_seg: 98.5268 +04/17 01:19:18 - mmengine - INFO - Iter(train) [ 61600/160000] base_lr: 6.2082e-05 lr: 2.2953e-07 eta: 1 day, 3:16:40 time: 0.9999 data_time: 0.0050 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0066 decode.acc_seg: 99.7517 aux.loss_ce: 0.0076 aux.acc_seg: 99.1806 +04/17 01:20:08 - mmengine - INFO - Iter(train) [ 61650/160000] base_lr: 6.2051e-05 lr: 2.2941e-07 eta: 1 day, 3:15:51 time: 1.0008 data_time: 0.0046 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0084 decode.acc_seg: 99.6758 aux.loss_ce: 0.0082 aux.acc_seg: 98.9677 +04/17 01:20:58 - mmengine - INFO - Iter(train) [ 61700/160000] base_lr: 6.2019e-05 lr: 2.2930e-07 eta: 1 day, 3:15:01 time: 1.0001 data_time: 0.0047 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0069 decode.acc_seg: 99.6979 aux.loss_ce: 0.0080 aux.acc_seg: 99.2336 +04/17 01:21:48 - mmengine - INFO - Iter(train) [ 61750/160000] base_lr: 6.1988e-05 lr: 2.2918e-07 eta: 1 day, 3:14:11 time: 1.0006 data_time: 0.0045 memory: 8462 loss: 0.0139 decode.loss_ce: 0.0066 decode.acc_seg: 99.6271 aux.loss_ce: 0.0073 aux.acc_seg: 98.8829 +04/17 01:22:38 - mmengine - INFO - Iter(train) [ 61800/160000] base_lr: 6.1956e-05 lr: 2.2906e-07 eta: 1 day, 3:13:21 time: 1.0002 data_time: 0.0047 memory: 8462 loss: 0.0184 decode.loss_ce: 0.0092 decode.acc_seg: 99.6683 aux.loss_ce: 0.0093 aux.acc_seg: 99.0225 +04/17 01:23:28 - mmengine - INFO - Iter(train) [ 61850/160000] base_lr: 6.1925e-05 lr: 2.2895e-07 eta: 1 day, 3:12:32 time: 0.9998 data_time: 0.0045 memory: 8462 loss: 0.0171 decode.loss_ce: 0.0081 decode.acc_seg: 99.6580 aux.loss_ce: 0.0090 aux.acc_seg: 99.1571 +04/17 01:24:18 - mmengine - INFO - Iter(train) [ 61900/160000] base_lr: 6.1893e-05 lr: 2.2883e-07 eta: 1 day, 3:11:42 time: 0.9984 data_time: 0.0046 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0075 decode.acc_seg: 99.6153 aux.loss_ce: 0.0083 aux.acc_seg: 99.0864 +04/17 01:25:08 - mmengine - INFO - Iter(train) [ 61950/160000] base_lr: 6.1862e-05 lr: 2.2872e-07 eta: 1 day, 3:10:52 time: 0.9996 data_time: 0.0045 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0081 decode.acc_seg: 99.7885 aux.loss_ce: 0.0087 aux.acc_seg: 99.2813 +04/17 01:25:58 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/17 01:25:58 - mmengine - INFO - Iter(train) [ 62000/160000] base_lr: 6.1830e-05 lr: 2.2860e-07 eta: 1 day, 3:10:02 time: 0.9986 data_time: 0.0049 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0075 decode.acc_seg: 99.6367 aux.loss_ce: 0.0083 aux.acc_seg: 99.0023 +04/17 01:26:48 - mmengine - INFO - Iter(train) [ 62050/160000] base_lr: 6.1798e-05 lr: 2.2848e-07 eta: 1 day, 3:09:13 time: 0.9991 data_time: 0.0045 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0073 decode.acc_seg: 99.5485 aux.loss_ce: 0.0076 aux.acc_seg: 98.6851 +04/17 01:27:38 - mmengine - INFO - Iter(train) [ 62100/160000] base_lr: 6.1767e-05 lr: 2.2837e-07 eta: 1 day, 3:08:23 time: 0.9993 data_time: 0.0046 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0073 decode.acc_seg: 99.6651 aux.loss_ce: 0.0074 aux.acc_seg: 99.2819 +04/17 01:28:28 - mmengine - INFO - Iter(train) [ 62150/160000] base_lr: 6.1735e-05 lr: 2.2825e-07 eta: 1 day, 3:07:33 time: 0.9986 data_time: 0.0045 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0070 decode.acc_seg: 99.7620 aux.loss_ce: 0.0078 aux.acc_seg: 99.2718 +04/17 01:29:18 - mmengine - INFO - Iter(train) [ 62200/160000] base_lr: 6.1704e-05 lr: 2.2813e-07 eta: 1 day, 3:06:43 time: 0.9992 data_time: 0.0045 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0073 decode.acc_seg: 99.5941 aux.loss_ce: 0.0081 aux.acc_seg: 98.6378 +04/17 01:30:08 - mmengine - INFO - Iter(train) [ 62250/160000] base_lr: 6.1672e-05 lr: 2.2802e-07 eta: 1 day, 3:05:53 time: 1.0011 data_time: 0.0048 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0075 decode.acc_seg: 99.4978 aux.loss_ce: 0.0082 aux.acc_seg: 98.8611 +04/17 01:30:57 - mmengine - INFO - Iter(train) [ 62300/160000] base_lr: 6.1641e-05 lr: 2.2790e-07 eta: 1 day, 3:05:04 time: 1.0008 data_time: 0.0043 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0075 decode.acc_seg: 99.5865 aux.loss_ce: 0.0085 aux.acc_seg: 99.0210 +04/17 01:31:48 - mmengine - INFO - Iter(train) [ 62350/160000] base_lr: 6.1609e-05 lr: 2.2778e-07 eta: 1 day, 3:04:14 time: 0.9998 data_time: 0.0044 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0078 decode.acc_seg: 99.8123 aux.loss_ce: 0.0076 aux.acc_seg: 99.4509 +04/17 01:32:38 - mmengine - INFO - Iter(train) [ 62400/160000] base_lr: 6.1578e-05 lr: 2.2767e-07 eta: 1 day, 3:03:24 time: 0.9989 data_time: 0.0043 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0071 decode.acc_seg: 99.7084 aux.loss_ce: 0.0082 aux.acc_seg: 99.1547 +04/17 01:33:28 - mmengine - INFO - Iter(train) [ 62450/160000] base_lr: 6.1546e-05 lr: 2.2755e-07 eta: 1 day, 3:02:34 time: 0.9991 data_time: 0.0047 memory: 8462 loss: 0.0165 decode.loss_ce: 0.0077 decode.acc_seg: 99.7868 aux.loss_ce: 0.0088 aux.acc_seg: 99.2298 +04/17 01:34:18 - mmengine - INFO - Iter(train) [ 62500/160000] base_lr: 6.1515e-05 lr: 2.2743e-07 eta: 1 day, 3:01:45 time: 1.0020 data_time: 0.0046 memory: 8462 loss: 0.0144 decode.loss_ce: 0.0071 decode.acc_seg: 99.7049 aux.loss_ce: 0.0073 aux.acc_seg: 99.2956 +04/17 01:35:08 - mmengine - INFO - Iter(train) [ 62550/160000] base_lr: 6.1483e-05 lr: 2.2732e-07 eta: 1 day, 3:00:55 time: 0.9994 data_time: 0.0049 memory: 8462 loss: 0.0132 decode.loss_ce: 0.0067 decode.acc_seg: 99.6576 aux.loss_ce: 0.0065 aux.acc_seg: 99.1966 +04/17 01:35:58 - mmengine - INFO - Iter(train) [ 62600/160000] base_lr: 6.1451e-05 lr: 2.2720e-07 eta: 1 day, 3:00:05 time: 1.0003 data_time: 0.0049 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0072 decode.acc_seg: 99.7746 aux.loss_ce: 0.0076 aux.acc_seg: 99.3265 +04/17 01:36:48 - mmengine - INFO - Iter(train) [ 62650/160000] base_lr: 6.1420e-05 lr: 2.2708e-07 eta: 1 day, 2:59:16 time: 0.9988 data_time: 0.0044 memory: 8462 loss: 0.0134 decode.loss_ce: 0.0063 decode.acc_seg: 99.8293 aux.loss_ce: 0.0071 aux.acc_seg: 99.4669 +04/17 01:37:38 - mmengine - INFO - Iter(train) [ 62700/160000] base_lr: 6.1388e-05 lr: 2.2697e-07 eta: 1 day, 2:58:26 time: 1.0001 data_time: 0.0046 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0070 decode.acc_seg: 99.5947 aux.loss_ce: 0.0080 aux.acc_seg: 99.0145 +04/17 01:38:28 - mmengine - INFO - Iter(train) [ 62750/160000] base_lr: 6.1357e-05 lr: 2.2685e-07 eta: 1 day, 2:57:36 time: 1.0006 data_time: 0.0046 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0072 decode.acc_seg: 99.6117 aux.loss_ce: 0.0076 aux.acc_seg: 99.1707 +04/17 01:39:18 - mmengine - INFO - Iter(train) [ 62800/160000] base_lr: 6.1325e-05 lr: 2.2673e-07 eta: 1 day, 2:56:46 time: 1.0013 data_time: 0.0047 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0071 decode.acc_seg: 99.6681 aux.loss_ce: 0.0078 aux.acc_seg: 99.3587 +04/17 01:40:08 - mmengine - INFO - Iter(train) [ 62850/160000] base_lr: 6.1294e-05 lr: 2.2662e-07 eta: 1 day, 2:55:57 time: 0.9993 data_time: 0.0045 memory: 8462 loss: 0.0189 decode.loss_ce: 0.0086 decode.acc_seg: 99.5710 aux.loss_ce: 0.0102 aux.acc_seg: 98.3358 +04/17 01:40:58 - mmengine - INFO - Iter(train) [ 62900/160000] base_lr: 6.1262e-05 lr: 2.2650e-07 eta: 1 day, 2:55:07 time: 0.9990 data_time: 0.0047 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0072 decode.acc_seg: 99.6313 aux.loss_ce: 0.0080 aux.acc_seg: 99.2073 +04/17 01:41:48 - mmengine - INFO - Iter(train) [ 62950/160000] base_lr: 6.1231e-05 lr: 2.2638e-07 eta: 1 day, 2:54:17 time: 0.9996 data_time: 0.0046 memory: 8462 loss: 0.0139 decode.loss_ce: 0.0071 decode.acc_seg: 99.7908 aux.loss_ce: 0.0069 aux.acc_seg: 99.6157 +04/17 01:42:38 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240416_080959 +04/17 01:42:38 - mmengine - INFO - Iter(train) [ 63000/160000] base_lr: 6.1199e-05 lr: 2.2627e-07 eta: 1 day, 2:53:27 time: 0.9990 data_time: 0.0047 memory: 8462 loss: 0.0145 decode.loss_ce: 0.0072 decode.acc_seg: 99.6899 aux.loss_ce: 0.0073 aux.acc_seg: 99.3097 +04/17 01:43:28 - mmengine - INFO - Iter(train) [ 63050/160000] base_lr: 6.1168e-05 lr: 2.2615e-07 eta: 1 day, 2:52:38 time: 0.9994 data_time: 0.0049 memory: 8462 loss: 0.0138 decode.loss_ce: 0.0069 decode.acc_seg: 99.6851 aux.loss_ce: 0.0069 aux.acc_seg: 99.4026 +04/17 01:44:18 - mmengine - INFO - Iter(train) [ 63100/160000] base_lr: 6.1136e-05 lr: 2.2603e-07 eta: 1 day, 2:51:48 time: 0.9997 data_time: 0.0048 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0074 decode.acc_seg: 99.6614 aux.loss_ce: 0.0080 aux.acc_seg: 99.0128 +04/17 01:45:08 - mmengine - INFO - Iter(train) [ 63150/160000] base_lr: 6.1104e-05 lr: 2.2592e-07 eta: 1 day, 2:50:58 time: 1.0001 data_time: 0.0047 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0072 decode.acc_seg: 99.7526 aux.loss_ce: 0.0086 aux.acc_seg: 99.2950 +04/17 01:45:58 - mmengine - INFO - Iter(train) [ 63200/160000] base_lr: 6.1073e-05 lr: 2.2580e-07 eta: 1 day, 2:50:08 time: 0.9990 data_time: 0.0045 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0073 decode.acc_seg: 99.7175 aux.loss_ce: 0.0079 aux.acc_seg: 99.3681 +04/17 01:46:48 - mmengine - INFO - Iter(train) [ 63250/160000] base_lr: 6.1041e-05 lr: 2.2568e-07 eta: 1 day, 2:49:18 time: 1.0005 data_time: 0.0047 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0070 decode.acc_seg: 99.7387 aux.loss_ce: 0.0086 aux.acc_seg: 99.3063 +04/17 01:47:38 - mmengine - INFO - Iter(train) [ 63300/160000] base_lr: 6.1010e-05 lr: 2.2557e-07 eta: 1 day, 2:48:29 time: 1.0007 data_time: 0.0046 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0072 decode.acc_seg: 99.6525 aux.loss_ce: 0.0077 aux.acc_seg: 99.2231 +04/17 01:48:28 - mmengine - INFO - Iter(train) [ 63350/160000] base_lr: 6.0978e-05 lr: 2.2545e-07 eta: 1 day, 2:47:39 time: 0.9999 data_time: 0.0045 memory: 8462 loss: 0.0169 decode.loss_ce: 0.0078 decode.acc_seg: 99.7160 aux.loss_ce: 0.0091 aux.acc_seg: 98.8708 +04/17 01:49:18 - mmengine - INFO - Iter(train) [ 63400/160000] base_lr: 6.0947e-05 lr: 2.2533e-07 eta: 1 day, 2:46:49 time: 1.0000 data_time: 0.0048 memory: 8462 loss: 0.0136 decode.loss_ce: 0.0065 decode.acc_seg: 99.7473 aux.loss_ce: 0.0071 aux.acc_seg: 99.4810 +04/17 01:50:08 - mmengine - INFO - Iter(train) [ 63450/160000] base_lr: 6.0915e-05 lr: 2.2522e-07 eta: 1 day, 2:45:59 time: 1.0002 data_time: 0.0045 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0076 decode.acc_seg: 99.6769 aux.loss_ce: 0.0080 aux.acc_seg: 99.2762 +04/17 01:50:58 - mmengine - INFO - Iter(train) [ 63500/160000] base_lr: 6.0884e-05 lr: 2.2510e-07 eta: 1 day, 2:45:10 time: 1.0004 data_time: 0.0050 memory: 8462 loss: 0.0151 decode.loss_ce: 0.0071 decode.acc_seg: 99.8365 aux.loss_ce: 0.0080 aux.acc_seg: 99.3488 +04/17 01:51:48 - mmengine - INFO - Iter(train) [ 63550/160000] base_lr: 6.0852e-05 lr: 2.2498e-07 eta: 1 day, 2:44:20 time: 1.0020 data_time: 0.0046 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0067 decode.acc_seg: 99.8329 aux.loss_ce: 0.0073 aux.acc_seg: 99.4867 +04/17 01:52:38 - mmengine - INFO - Iter(train) [ 63600/160000] base_lr: 6.0821e-05 lr: 2.2487e-07 eta: 1 day, 2:43:30 time: 0.9993 data_time: 0.0047 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0068 decode.acc_seg: 99.7368 aux.loss_ce: 0.0072 aux.acc_seg: 99.3559 +04/17 01:53:28 - mmengine - INFO - Iter(train) [ 63650/160000] base_lr: 6.0789e-05 lr: 2.2475e-07 eta: 1 day, 2:42:40 time: 0.9991 data_time: 0.0050 memory: 8462 loss: 0.0164 decode.loss_ce: 0.0082 decode.acc_seg: 99.6851 aux.loss_ce: 0.0082 aux.acc_seg: 99.2630 +04/17 01:54:18 - mmengine - INFO - Iter(train) [ 63700/160000] base_lr: 6.0757e-05 lr: 2.2463e-07 eta: 1 day, 2:41:51 time: 0.9990 data_time: 0.0045 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0070 decode.acc_seg: 99.5785 aux.loss_ce: 0.0080 aux.acc_seg: 99.0826 +04/17 01:55:08 - mmengine - INFO - Iter(train) [ 63750/160000] base_lr: 6.0726e-05 lr: 2.2452e-07 eta: 1 day, 2:41:01 time: 1.0010 data_time: 0.0049 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0083 decode.acc_seg: 99.6666 aux.loss_ce: 0.0088 aux.acc_seg: 98.8733 +WARNING:torch.distributed.elastic.agent.server.api:Received 2 death signal, shutting down workers +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 366197 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 366198 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 366199 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 366200 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 366197 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 366198 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 366199 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 366200 closing signal SIGTERM +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 709, in run + result = self._invoke_run(role) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 850, in _invoke_run + time.sleep(monitor_interval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 366166 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 716, in run + self._shutdown(e.sigval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 366166 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 236, in launch_agent + result = agent.run() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/metrics/api.py", line 125, in wrapper + result = f(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 721, in run + self._shutdown() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 366166 got signal: 2 +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/17 01:58:49 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 1256867340 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 1256867340 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/17 01:58:50 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = None +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=1024, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + num_convs=1, + type='FCNHead'), + backbone=dict( + contract_dilation=True, + depth=101, + dilations=( + 1, + 1, + 1, + 1, + ), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + norm_eval=False, + num_stages=4, + out_indices=( + 0, + 1, + 2, + 3, + ), + strides=( + 1, + 2, + 2, + 2, + ), + style='pytorch', + type='ResNetV1c'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=512, + dropout_ratio=0.1, + in_channels=[ + 256, + 512, + 1024, + 2048, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + pretrained='open-mmlab://resnet101_v1c', + test_cfg=dict(mode='whole'), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + clip_grad=None, + optimizer=dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, + by_epoch=False, + end=160000, + eta_min=0.0001, + power=0.9, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=4, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = './work_dirs/upernet_r101_4xb4-160k_cag-512x512' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/17 01:58:52 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +04/17 01:58:53 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +04/17 01:58:53 - mmengine - INFO - load model from: open-mmlab://resnet101_v1c +04/17 01:58:53 - mmengine - INFO - Loads checkpoint by openmmlab backend from path: open-mmlab://resnet101_v1c +Downloading: "https://download.openmmlab.com/pretrain/third_party/resnet101_v1c-e67eebb6.pth" to /root/.cache/torch/hub/checkpoints/resnet101_v1c-e67eebb6.pth +04/17 01:59:09 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: fc.weight, fc.bias + +04/17 01:59:09 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +04/17 01:59:09 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +04/17 01:59:09 - mmengine - INFO - Checkpoints will be saved to /workspaces/mmsegmentation-1/work_dirs/upernet_r101_4xb4-160k_cag-512x512. +04/17 01:59:52 - mmengine - INFO - Iter(train) [ 50/160000] lr: 9.9973e-03 eta: 1 day, 14:09:24 time: 0.7661 data_time: 0.0057 memory: 8166 loss: 0.2475 decode.loss_ce: 0.1574 decode.acc_seg: 96.1514 aux.loss_ce: 0.0902 aux.acc_seg: 96.1514 +04/17 02:00:31 - mmengine - INFO - Iter(train) [ 100/160000] lr: 9.9945e-03 eta: 1 day, 12:25:46 time: 0.7943 data_time: 0.0054 memory: 8166 loss: 0.1151 decode.loss_ce: 0.0504 decode.acc_seg: 98.2868 aux.loss_ce: 0.0647 aux.acc_seg: 96.7345 +04/17 02:01:11 - mmengine - INFO - Iter(train) [ 150/160000] lr: 9.9917e-03 eta: 1 day, 12:04:58 time: 0.7971 data_time: 0.0059 memory: 8166 loss: 0.0849 decode.loss_ce: 0.0463 decode.acc_seg: 98.0924 aux.loss_ce: 0.0387 aux.acc_seg: 97.1909 +WARNING:torch.distributed.elastic.agent.server.api:Received 2 death signal, shutting down workers +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 789665 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 789666 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 789667 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 789668 closing signal SIGINT +Traceback (most recent call last): +Traceback (most recent call last): + File "tools/train.py", line 104, in + File "tools/train.py", line 104, in +Traceback (most recent call last): +Traceback (most recent call last): + File "tools/train.py", line 104, in + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train +runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + main() +main() + File "tools/train.py", line 100, in main + File "tools/train.py", line 100, in main + runner.train()runner.train() + + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + model = self.train_loop.run() # type: ignore + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + model = self.train_loop.run() # type: ignore + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + self.run_iter(data_batch) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + self.run_iter(data_batch) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + outputs = self.runner.model.train_step( + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + outputs = self.runner.model.train_step(model = self.train_loop.run() # type: ignoremodel = self.train_loop.run() # type: ignore + + + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + optim_wrapper.update_params(parsed_loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 196, in update_params + optim_wrapper.update_params(parsed_loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 196, in update_params + self.backward(loss) +self.run_iter(data_batch)self.run_iter(data_batch) File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 220, in backward + + + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter +self.backward(loss) File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 220, in backward + loss.backward(**kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/_tensor.py", line 363, in backward + loss.backward(**kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/_tensor.py", line 363, in backward + outputs = self.runner.model.train_step( +outputs = self.runner.model.train_step( File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + optim_wrapper.update_params(parsed_loss)torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs)optim_wrapper.update_params(parsed_loss)torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) + + + + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 196, in update_params + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py", line 173, in backward + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 196, in update_params + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py", line 173, in backward + Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass + self.backward(loss) self.backward(loss)KeyboardInterrupt +Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 220, in backward + + + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 220, in backward +KeyboardInterrupt + loss.backward(**kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/_tensor.py", line 363, in backward +loss.backward(**kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/_tensor.py", line 363, in backward + torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) +torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) File "/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py", line 173, in backward + + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py", line 173, in backward + Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward passVariable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass + +KeyboardInterruptKeyboardInterrupt + +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 789665 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 789666 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 789667 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 789668 closing signal SIGTERM +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 709, in run + result = self._invoke_run(role) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 850, in _invoke_run + time.sleep(monitor_interval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 789635 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 716, in run + self._shutdown(e.sigval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 789635 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 236, in launch_agent + result = agent.run() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/metrics/api.py", line 125, in wrapper + result = f(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 721, in run + self._shutdown() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 789635 got signal: 2 +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/17 02:01:24 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 259084046 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 259084046 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/17 02:01:24 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = None +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=1024, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=2, + num_convs=1, + type='FCNHead'), + backbone=dict( + contract_dilation=True, + depth=101, + dilations=( + 1, + 1, + 1, + 1, + ), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + norm_eval=False, + num_stages=4, + out_indices=( + 0, + 1, + 2, + 3, + ), + strides=( + 1, + 2, + 2, + 2, + ), + style='pytorch', + type='ResNetV1c'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=512, + dropout_ratio=0.1, + in_channels=[ + 256, + 512, + 1024, + 2048, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=2, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + pretrained='open-mmlab://resnet101_v1c', + test_cfg=dict(mode='whole'), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + clip_grad=None, + optimizer=dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, + by_epoch=False, + end=160000, + eta_min=0.0001, + power=0.9, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=4, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = './work_dirs/upernet_r101_4xb4-160k_cag-512x512' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/17 02:01:26 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +04/17 02:01:27 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +04/17 02:01:27 - mmengine - INFO - load model from: open-mmlab://resnet101_v1c +04/17 02:01:27 - mmengine - INFO - Loads checkpoint by openmmlab backend from path: open-mmlab://resnet101_v1c +04/17 02:01:27 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: fc.weight, fc.bias + +04/17 02:01:27 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +04/17 02:01:27 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +04/17 02:01:27 - mmengine - INFO - Checkpoints will be saved to /workspaces/mmsegmentation-1/work_dirs/upernet_r101_4xb4-160k_cag-512x512. +WARNING:torch.distributed.elastic.agent.server.api:Received 2 death signal, shutting down workers +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 792179 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 792180 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 792181 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 792182 closing signal SIGINT +Traceback (most recent call last): + File "tools/train.py", line 104, in +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main +Traceback (most recent call last): + runner.train() File "tools/train.py", line 104, in + + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + main() +main() File "tools/train.py", line 100, in main + + File "tools/train.py", line 100, in main +runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + runner.train()model = self.train_loop.run() # type: ignore + +runner.train() File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + self.run_iter(data_batch) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + outputs = self.runner.model.train_step( + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + model = self.train_loop.run() # type: ignoreoptim_wrapper.update_params(parsed_loss) + + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 196, in update_params + model = self.train_loop.run() # type: ignore + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + model = self.train_loop.run() # type: ignore + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + self.backward(loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 220, in backward +self.run_iter(data_batch) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + self.run_iter(data_batch) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + self.run_iter(data_batch)loss.backward(**kwargs) + + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + File "/opt/conda/lib/python3.8/site-packages/torch/_tensor.py", line 363, in backward +outputs = self.runner.model.train_step( + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + outputs = self.runner.model.train_step( +optim_wrapper.update_params(parsed_loss) File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + +outputs = self.runner.model.train_step( File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 196, in update_params + +torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py", line 173, in backward + optim_wrapper.update_params(parsed_loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 196, in update_params + self.backward(loss)optim_wrapper.update_params(parsed_loss) + + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 220, in backward + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 196, in update_params + Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass +self.backward(loss) +KeyboardInterrupt File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 220, in backward + + loss.backward(**kwargs) +self.backward(loss) File "/opt/conda/lib/python3.8/site-packages/torch/_tensor.py", line 363, in backward + + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 220, in backward + loss.backward(**kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/_tensor.py", line 363, in backward + loss.backward(**kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/_tensor.py", line 363, in backward + torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py", line 173, in backward + torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py", line 173, in backward + torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py", line 173, in backward +Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass +KeyboardInterrupt + Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass +KeyboardInterrupt + Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass +KeyboardInterrupt +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 792179 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 792180 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 792181 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 792182 closing signal SIGTERM +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 709, in run + result = self._invoke_run(role) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 850, in _invoke_run + time.sleep(monitor_interval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 792148 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 716, in run + self._shutdown(e.sigval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 792148 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 236, in launch_agent + result = agent.run() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/metrics/api.py", line 125, in wrapper + result = f(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 721, in run + self._shutdown() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 792148 got signal: 2 +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/17 02:02:09 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 1622490299 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 1622490299 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/17 02:02:10 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = None +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=1024, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=2, + num_convs=1, + type='FCNHead'), + backbone=dict( + contract_dilation=True, + depth=101, + dilations=( + 1, + 1, + 1, + 1, + ), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + norm_eval=False, + num_stages=4, + out_indices=( + 0, + 1, + 2, + 3, + ), + strides=( + 1, + 2, + 2, + 2, + ), + style='pytorch', + type='ResNetV1c'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=512, + dropout_ratio=0.1, + in_channels=[ + 256, + 512, + 1024, + 2048, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=2, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + pretrained='open-mmlab://resnet101_v1c', + test_cfg=dict(mode='whole'), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + clip_grad=None, + optimizer=dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, + by_epoch=False, + end=160000, + eta_min=0.0001, + power=0.9, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=4, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = './work_dirs/upernet_r101_4xb4-160k_cag-512x512' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/17 02:02:12 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +04/17 02:02:13 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +04/17 02:02:13 - mmengine - INFO - load model from: open-mmlab://resnet101_v1c +04/17 02:02:13 - mmengine - INFO - Loads checkpoint by openmmlab backend from path: open-mmlab://resnet101_v1c +04/17 02:02:13 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: fc.weight, fc.bias + +04/17 02:02:13 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +04/17 02:02:13 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +04/17 02:02:13 - mmengine - INFO - Checkpoints will be saved to /workspaces/mmsegmentation-1/work_dirs/upernet_r101_4xb4-160k_cag-512x512. +WARNING:torch.distributed.elastic.agent.server.api:Received 2 death signal, shutting down workers +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 793328 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 793329 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 793330 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 793331 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 793328 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 793329 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 793330 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 793331 closing signal SIGTERM +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 709, in run + result = self._invoke_run(role) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 850, in _invoke_run + time.sleep(monitor_interval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 793297 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 716, in run + self._shutdown(e.sigval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 793297 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 236, in launch_agent + result = agent.run() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/metrics/api.py", line 125, in wrapper + result = f(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 721, in run + self._shutdown() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 793297 got signal: 2 +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/17 02:02:22 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 968316655 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 968316655 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/17 02:02:23 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = None +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=1024, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=2, + num_convs=1, + type='FCNHead'), + backbone=dict( + contract_dilation=True, + depth=101, + dilations=( + 1, + 1, + 1, + 1, + ), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + norm_eval=False, + num_stages=4, + out_indices=( + 0, + 1, + 2, + 3, + ), + strides=( + 1, + 2, + 2, + 2, + ), + style='pytorch', + type='ResNetV1c'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=512, + dropout_ratio=0.1, + in_channels=[ + 256, + 512, + 1024, + 2048, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=2, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + pretrained='open-mmlab://resnet101_v1c', + test_cfg=dict(mode='whole'), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + clip_grad=None, + optimizer=dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, + by_epoch=False, + end=160000, + eta_min=0.0001, + power=0.9, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=4, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = './work_dirs/upernet_r101_4xb4-160k_cag-512x512' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/17 02:02:24 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +04/17 02:02:26 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +04/17 02:02:26 - mmengine - INFO - load model from: open-mmlab://resnet101_v1c +04/17 02:02:26 - mmengine - INFO - Loads checkpoint by openmmlab backend from path: open-mmlab://resnet101_v1c +04/17 02:02:26 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: fc.weight, fc.bias + +04/17 02:02:26 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +04/17 02:02:26 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +04/17 02:02:26 - mmengine - INFO - Checkpoints will be saved to /workspaces/mmsegmentation-1/work_dirs/upernet_r101_4xb4-160k_cag-512x512. +04/17 02:02:58 - mmengine - INFO - Iter(train) [ 50/160000] lr: 9.9973e-03 eta: 1 day, 4:45:42 time: 0.5454 data_time: 0.0064 memory: 7635 loss: 0.1422 decode.loss_ce: 0.0900 decode.acc_seg: 96.6448 aux.loss_ce: 0.0522 aux.acc_seg: 96.6413 +04/17 02:03:26 - mmengine - INFO - Iter(train) [ 100/160000] lr: 9.9945e-03 eta: 1 day, 2:31:41 time: 0.5494 data_time: 0.0055 memory: 7635 loss: 0.0984 decode.loss_ce: 0.0598 decode.acc_seg: 97.3445 aux.loss_ce: 0.0386 aux.acc_seg: 96.4732 +04/17 02:03:53 - mmengine - INFO - Iter(train) [ 150/160000] lr: 9.9917e-03 eta: 1 day, 1:47:48 time: 0.5473 data_time: 0.0064 memory: 7635 loss: 0.0804 decode.loss_ce: 0.0477 decode.acc_seg: 98.2474 aux.loss_ce: 0.0327 aux.acc_seg: 96.4436 +04/17 02:04:21 - mmengine - INFO - Iter(train) [ 200/160000] lr: 9.9889e-03 eta: 1 day, 1:26:04 time: 0.5494 data_time: 0.0062 memory: 7635 loss: 0.0758 decode.loss_ce: 0.0457 decode.acc_seg: 97.5277 aux.loss_ce: 0.0302 aux.acc_seg: 95.7481 +04/17 02:04:48 - mmengine - INFO - Iter(train) [ 250/160000] lr: 9.9861e-03 eta: 1 day, 1:12:38 time: 0.5488 data_time: 0.0062 memory: 7635 loss: 0.0767 decode.loss_ce: 0.0468 decode.acc_seg: 98.4130 aux.loss_ce: 0.0299 aux.acc_seg: 97.0962 +04/17 02:05:15 - mmengine - INFO - Iter(train) [ 300/160000] lr: 9.9833e-03 eta: 1 day, 1:04:02 time: 0.5501 data_time: 0.0056 memory: 7635 loss: 0.0743 decode.loss_ce: 0.0468 decode.acc_seg: 98.1305 aux.loss_ce: 0.0275 aux.acc_seg: 97.3926 +04/17 02:05:43 - mmengine - INFO - Iter(train) [ 350/160000] lr: 9.9806e-03 eta: 1 day, 0:57:55 time: 0.5519 data_time: 0.0056 memory: 7635 loss: 0.0737 decode.loss_ce: 0.0474 decode.acc_seg: 97.6963 aux.loss_ce: 0.0263 aux.acc_seg: 97.4849 +04/17 02:06:11 - mmengine - INFO - Iter(train) [ 400/160000] lr: 9.9778e-03 eta: 1 day, 0:53:18 time: 0.5512 data_time: 0.0060 memory: 7635 loss: 0.0647 decode.loss_ce: 0.0413 decode.acc_seg: 98.7028 aux.loss_ce: 0.0234 aux.acc_seg: 98.0618 +04/17 02:06:38 - mmengine - INFO - Iter(train) [ 450/160000] lr: 9.9750e-03 eta: 1 day, 0:49:44 time: 0.5509 data_time: 0.0060 memory: 7635 loss: 0.0556 decode.loss_ce: 0.0345 decode.acc_seg: 98.9654 aux.loss_ce: 0.0210 aux.acc_seg: 98.1362 +04/17 02:07:06 - mmengine - INFO - Iter(train) [ 500/160000] lr: 9.9722e-03 eta: 1 day, 0:47:28 time: 0.5506 data_time: 0.0060 memory: 7635 loss: 0.0546 decode.loss_ce: 0.0342 decode.acc_seg: 98.3095 aux.loss_ce: 0.0204 aux.acc_seg: 97.5809 +04/17 02:07:33 - mmengine - INFO - Iter(train) [ 550/160000] lr: 9.9694e-03 eta: 1 day, 0:44:58 time: 0.5514 data_time: 0.0059 memory: 7635 loss: 0.0577 decode.loss_ce: 0.0372 decode.acc_seg: 99.0083 aux.loss_ce: 0.0205 aux.acc_seg: 98.3421 +04/17 02:08:01 - mmengine - INFO - Iter(train) [ 600/160000] lr: 9.9666e-03 eta: 1 day, 0:42:50 time: 0.5509 data_time: 0.0055 memory: 7635 loss: 0.0616 decode.loss_ce: 0.0397 decode.acc_seg: 98.0560 aux.loss_ce: 0.0219 aux.acc_seg: 97.6258 +04/17 02:08:28 - mmengine - INFO - Iter(train) [ 650/160000] lr: 9.9639e-03 eta: 1 day, 0:41:06 time: 0.5517 data_time: 0.0053 memory: 7635 loss: 0.0578 decode.loss_ce: 0.0370 decode.acc_seg: 98.5957 aux.loss_ce: 0.0208 aux.acc_seg: 97.8125 +04/17 02:08:56 - mmengine - INFO - Iter(train) [ 700/160000] lr: 9.9611e-03 eta: 1 day, 0:39:15 time: 0.5503 data_time: 0.0059 memory: 7635 loss: 0.0532 decode.loss_ce: 0.0340 decode.acc_seg: 98.7547 aux.loss_ce: 0.0192 aux.acc_seg: 98.1804 +04/17 02:09:24 - mmengine - INFO - Iter(train) [ 750/160000] lr: 9.9583e-03 eta: 1 day, 0:37:42 time: 0.5514 data_time: 0.0061 memory: 7635 loss: 0.0522 decode.loss_ce: 0.0336 decode.acc_seg: 99.0025 aux.loss_ce: 0.0186 aux.acc_seg: 98.7077 +WARNING:torch.distributed.elastic.agent.server.api:Received 1 death signal, shutting down workers +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 794125 closing signal SIGHUP +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 794126 closing signal SIGHUP +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 794127 closing signal SIGHUP +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 794128 closing signal SIGHUP +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/17 02:22:25 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 536263187 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 536263187 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/17 02:22:25 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = None +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=1024, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=2, + num_convs=1, + type='FCNHead'), + backbone=dict( + contract_dilation=True, + depth=101, + dilations=( + 1, + 1, + 1, + 1, + ), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + norm_eval=False, + num_stages=4, + out_indices=( + 0, + 1, + 2, + 3, + ), + strides=( + 1, + 2, + 2, + 2, + ), + style='pytorch', + type='ResNetV1c'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=512, + dropout_ratio=0.1, + in_channels=[ + 256, + 512, + 1024, + 2048, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=2, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + pretrained='open-mmlab://resnet101_v1c', + test_cfg=dict(mode='whole'), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + clip_grad=None, + optimizer=dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, + by_epoch=False, + end=160000, + eta_min=0.0001, + power=0.9, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=4, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = './work_dirs/upernet_r101_4xb4-160k_cag-512x512' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/17 02:22:27 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +04/17 02:22:28 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +04/17 02:22:28 - mmengine - INFO - load model from: open-mmlab://resnet101_v1c +04/17 02:22:28 - mmengine - INFO - Loads checkpoint by openmmlab backend from path: open-mmlab://resnet101_v1c +04/17 02:22:28 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: fc.weight, fc.bias + +04/17 02:22:28 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +04/17 02:22:28 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +04/17 02:22:28 - mmengine - INFO - Checkpoints will be saved to /workspaces/mmsegmentation-1/work_dirs/upernet_r101_4xb4-160k_cag-512x512. +04/17 02:23:00 - mmengine - INFO - Iter(train) [ 50/160000] lr: 9.9973e-03 eta: 1 day, 4:26:13 time: 0.5398 data_time: 0.0063 memory: 7635 loss: 0.1440 decode.loss_ce: 0.0952 decode.acc_seg: 97.1099 aux.loss_ce: 0.0488 aux.acc_seg: 97.1099 +04/17 02:23:28 - mmengine - INFO - Iter(train) [ 100/160000] lr: 9.9945e-03 eta: 1 day, 2:15:52 time: 0.5444 data_time: 0.0059 memory: 7635 loss: 0.0943 decode.loss_ce: 0.0589 decode.acc_seg: 97.2680 aux.loss_ce: 0.0355 aux.acc_seg: 96.6311 +04/17 02:23:55 - mmengine - INFO - Iter(train) [ 150/160000] lr: 9.9917e-03 eta: 1 day, 1:33:56 time: 0.5451 data_time: 0.0062 memory: 7635 loss: 0.0767 decode.loss_ce: 0.0459 decode.acc_seg: 98.7021 aux.loss_ce: 0.0307 aux.acc_seg: 96.7049 +04/17 02:24:22 - mmengine - INFO - Iter(train) [ 200/160000] lr: 9.9889e-03 eta: 1 day, 1:13:36 time: 0.5473 data_time: 0.0056 memory: 7635 loss: 0.0753 decode.loss_ce: 0.0463 decode.acc_seg: 98.7144 aux.loss_ce: 0.0291 aux.acc_seg: 97.6154 +04/17 02:24:49 - mmengine - INFO - Iter(train) [ 250/160000] lr: 9.9861e-03 eta: 1 day, 1:01:50 time: 0.5468 data_time: 0.0057 memory: 7635 loss: 0.0751 decode.loss_ce: 0.0457 decode.acc_seg: 98.6154 aux.loss_ce: 0.0294 aux.acc_seg: 97.7401 +04/17 02:25:17 - mmengine - INFO - Iter(train) [ 300/160000] lr: 9.9833e-03 eta: 1 day, 0:54:02 time: 0.5479 data_time: 0.0060 memory: 7635 loss: 0.0692 decode.loss_ce: 0.0432 decode.acc_seg: 98.8323 aux.loss_ce: 0.0260 aux.acc_seg: 98.2899 +04/17 02:25:44 - mmengine - INFO - Iter(train) [ 350/160000] lr: 9.9806e-03 eta: 1 day, 0:48:39 time: 0.5470 data_time: 0.0057 memory: 7635 loss: 0.0672 decode.loss_ce: 0.0424 decode.acc_seg: 98.5642 aux.loss_ce: 0.0249 aux.acc_seg: 97.9949 +04/17 02:26:12 - mmengine - INFO - Iter(train) [ 400/160000] lr: 9.9778e-03 eta: 1 day, 0:44:29 time: 0.5486 data_time: 0.0065 memory: 7635 loss: 0.0723 decode.loss_ce: 0.0463 decode.acc_seg: 97.5659 aux.loss_ce: 0.0260 aux.acc_seg: 97.4424 +04/17 02:26:39 - mmengine - INFO - Iter(train) [ 450/160000] lr: 9.9750e-03 eta: 1 day, 0:41:24 time: 0.5492 data_time: 0.0058 memory: 7635 loss: 0.0604 decode.loss_ce: 0.0377 decode.acc_seg: 98.4617 aux.loss_ce: 0.0228 aux.acc_seg: 97.6000 +04/17 02:27:07 - mmengine - INFO - Iter(train) [ 500/160000] lr: 9.9722e-03 eta: 1 day, 0:39:25 time: 0.5489 data_time: 0.0061 memory: 7635 loss: 0.0621 decode.loss_ce: 0.0397 decode.acc_seg: 98.8098 aux.loss_ce: 0.0224 aux.acc_seg: 98.3232 +04/17 02:27:34 - mmengine - INFO - Iter(train) [ 550/160000] lr: 9.9694e-03 eta: 1 day, 0:37:21 time: 0.5496 data_time: 0.0061 memory: 7635 loss: 0.0637 decode.loss_ce: 0.0407 decode.acc_seg: 99.2899 aux.loss_ce: 0.0230 aux.acc_seg: 98.3852 +04/17 02:28:02 - mmengine - INFO - Iter(train) [ 600/160000] lr: 9.9666e-03 eta: 1 day, 0:35:40 time: 0.5503 data_time: 0.0058 memory: 7635 loss: 0.0580 decode.loss_ce: 0.0368 decode.acc_seg: 98.2452 aux.loss_ce: 0.0213 aux.acc_seg: 96.9857 +04/17 02:28:29 - mmengine - INFO - Iter(train) [ 650/160000] lr: 9.9639e-03 eta: 1 day, 0:34:01 time: 0.5498 data_time: 0.0057 memory: 7635 loss: 0.0550 decode.loss_ce: 0.0349 decode.acc_seg: 98.5855 aux.loss_ce: 0.0201 aux.acc_seg: 98.1388 +04/17 02:28:57 - mmengine - INFO - Iter(train) [ 700/160000] lr: 9.9611e-03 eta: 1 day, 0:32:33 time: 0.5508 data_time: 0.0066 memory: 7635 loss: 0.0603 decode.loss_ce: 0.0393 decode.acc_seg: 98.4143 aux.loss_ce: 0.0210 aux.acc_seg: 97.8218 +04/17 02:29:24 - mmengine - INFO - Iter(train) [ 750/160000] lr: 9.9583e-03 eta: 1 day, 0:31:20 time: 0.5510 data_time: 0.0058 memory: 7635 loss: 0.0479 decode.loss_ce: 0.0299 decode.acc_seg: 98.2230 aux.loss_ce: 0.0180 aux.acc_seg: 97.5971 +04/17 02:29:52 - mmengine - INFO - Iter(train) [ 800/160000] lr: 9.9555e-03 eta: 1 day, 0:30:12 time: 0.5502 data_time: 0.0056 memory: 7635 loss: 0.0497 decode.loss_ce: 0.0315 decode.acc_seg: 99.0558 aux.loss_ce: 0.0182 aux.acc_seg: 98.1543 +04/17 02:30:19 - mmengine - INFO - Iter(train) [ 850/160000] lr: 9.9527e-03 eta: 1 day, 0:29:05 time: 0.5504 data_time: 0.0065 memory: 7635 loss: 0.0490 decode.loss_ce: 0.0306 decode.acc_seg: 99.3666 aux.loss_ce: 0.0184 aux.acc_seg: 98.7853 +04/17 02:30:47 - mmengine - INFO - Iter(train) [ 900/160000] lr: 9.9499e-03 eta: 1 day, 0:28:09 time: 0.5507 data_time: 0.0059 memory: 7635 loss: 0.0533 decode.loss_ce: 0.0340 decode.acc_seg: 98.0306 aux.loss_ce: 0.0193 aux.acc_seg: 97.3399 +04/17 02:31:14 - mmengine - INFO - Iter(train) [ 950/160000] lr: 9.9471e-03 eta: 1 day, 0:27:18 time: 0.5504 data_time: 0.0066 memory: 7635 loss: 0.0532 decode.loss_ce: 0.0340 decode.acc_seg: 98.8028 aux.loss_ce: 0.0191 aux.acc_seg: 98.3513 +04/17 02:31:42 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 02:31:42 - mmengine - INFO - Iter(train) [ 1000/160000] lr: 9.9444e-03 eta: 1 day, 0:26:27 time: 0.5500 data_time: 0.0064 memory: 7635 loss: 0.0465 decode.loss_ce: 0.0290 decode.acc_seg: 99.1015 aux.loss_ce: 0.0175 aux.acc_seg: 98.5930 +04/17 02:32:09 - mmengine - INFO - Iter(train) [ 1050/160000] lr: 9.9416e-03 eta: 1 day, 0:25:40 time: 0.5515 data_time: 0.0063 memory: 7635 loss: 0.0456 decode.loss_ce: 0.0287 decode.acc_seg: 98.5889 aux.loss_ce: 0.0168 aux.acc_seg: 98.2008 +04/17 02:32:18 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 02:32:37 - mmengine - INFO - Iter(train) [ 1100/160000] lr: 9.9388e-03 eta: 1 day, 0:24:54 time: 0.5515 data_time: 0.0055 memory: 7635 loss: 0.0503 decode.loss_ce: 0.0322 decode.acc_seg: 99.0838 aux.loss_ce: 0.0181 aux.acc_seg: 98.5331 +04/17 02:33:04 - mmengine - INFO - Iter(train) [ 1150/160000] lr: 9.9360e-03 eta: 1 day, 0:24:12 time: 0.5511 data_time: 0.0052 memory: 7635 loss: 0.0447 decode.loss_ce: 0.0280 decode.acc_seg: 99.0372 aux.loss_ce: 0.0166 aux.acc_seg: 98.5503 +04/17 02:33:32 - mmengine - INFO - Iter(train) [ 1200/160000] lr: 9.9332e-03 eta: 1 day, 0:23:34 time: 0.5521 data_time: 0.0065 memory: 7635 loss: 0.0595 decode.loss_ce: 0.0387 decode.acc_seg: 99.2927 aux.loss_ce: 0.0208 aux.acc_seg: 98.7716 +04/17 02:34:00 - mmengine - INFO - Iter(train) [ 1250/160000] lr: 9.9304e-03 eta: 1 day, 0:22:54 time: 0.5505 data_time: 0.0067 memory: 7635 loss: 0.0443 decode.loss_ce: 0.0279 decode.acc_seg: 98.7502 aux.loss_ce: 0.0164 aux.acc_seg: 98.2639 +04/17 02:34:27 - mmengine - INFO - Iter(train) [ 1300/160000] lr: 9.9276e-03 eta: 1 day, 0:22:16 time: 0.5506 data_time: 0.0056 memory: 7635 loss: 0.0481 decode.loss_ce: 0.0303 decode.acc_seg: 98.9054 aux.loss_ce: 0.0178 aux.acc_seg: 98.3751 +04/17 02:34:55 - mmengine - INFO - Iter(train) [ 1350/160000] lr: 9.9248e-03 eta: 1 day, 0:21:40 time: 0.5507 data_time: 0.0061 memory: 7635 loss: 0.0418 decode.loss_ce: 0.0262 decode.acc_seg: 98.9240 aux.loss_ce: 0.0156 aux.acc_seg: 98.3135 +04/17 02:35:22 - mmengine - INFO - Iter(train) [ 1400/160000] lr: 9.9221e-03 eta: 1 day, 0:21:02 time: 0.5507 data_time: 0.0059 memory: 7635 loss: 0.0399 decode.loss_ce: 0.0246 decode.acc_seg: 99.2918 aux.loss_ce: 0.0153 aux.acc_seg: 98.6363 +04/17 02:35:50 - mmengine - INFO - Iter(train) [ 1450/160000] lr: 9.9193e-03 eta: 1 day, 0:20:24 time: 0.5517 data_time: 0.0063 memory: 7635 loss: 0.0392 decode.loss_ce: 0.0243 decode.acc_seg: 98.7610 aux.loss_ce: 0.0149 aux.acc_seg: 98.0757 +04/17 02:36:17 - mmengine - INFO - Iter(train) [ 1500/160000] lr: 9.9165e-03 eta: 1 day, 0:19:50 time: 0.5509 data_time: 0.0056 memory: 7635 loss: 0.0439 decode.loss_ce: 0.0272 decode.acc_seg: 99.0294 aux.loss_ce: 0.0168 aux.acc_seg: 98.4800 +04/17 02:36:45 - mmengine - INFO - Iter(train) [ 1550/160000] lr: 9.9137e-03 eta: 1 day, 0:19:17 time: 0.5527 data_time: 0.0060 memory: 7635 loss: 0.0410 decode.loss_ce: 0.0255 decode.acc_seg: 97.9154 aux.loss_ce: 0.0156 aux.acc_seg: 97.3303 +04/17 02:37:13 - mmengine - INFO - Iter(train) [ 1600/160000] lr: 9.9109e-03 eta: 1 day, 0:18:54 time: 0.5505 data_time: 0.0061 memory: 7635 loss: 0.0396 decode.loss_ce: 0.0247 decode.acc_seg: 99.0260 aux.loss_ce: 0.0149 aux.acc_seg: 98.5363 +04/17 02:37:40 - mmengine - INFO - Iter(train) [ 1650/160000] lr: 9.9081e-03 eta: 1 day, 0:18:22 time: 0.5512 data_time: 0.0055 memory: 7635 loss: 0.0462 decode.loss_ce: 0.0291 decode.acc_seg: 99.0538 aux.loss_ce: 0.0170 aux.acc_seg: 98.4682 +04/17 02:38:08 - mmengine - INFO - Iter(train) [ 1700/160000] lr: 9.9053e-03 eta: 1 day, 0:17:50 time: 0.5519 data_time: 0.0068 memory: 7635 loss: 0.0442 decode.loss_ce: 0.0286 decode.acc_seg: 99.2168 aux.loss_ce: 0.0157 aux.acc_seg: 98.7749 +04/17 02:38:35 - mmengine - INFO - Iter(train) [ 1750/160000] lr: 9.9025e-03 eta: 1 day, 0:17:19 time: 0.5530 data_time: 0.0058 memory: 7635 loss: 0.0415 decode.loss_ce: 0.0259 decode.acc_seg: 99.2253 aux.loss_ce: 0.0156 aux.acc_seg: 98.6609 +04/17 02:39:03 - mmengine - INFO - Iter(train) [ 1800/160000] lr: 9.8998e-03 eta: 1 day, 0:16:46 time: 0.5506 data_time: 0.0059 memory: 7635 loss: 0.0408 decode.loss_ce: 0.0256 decode.acc_seg: 98.8564 aux.loss_ce: 0.0152 aux.acc_seg: 98.4221 +04/17 02:39:30 - mmengine - INFO - Iter(train) [ 1850/160000] lr: 9.8970e-03 eta: 1 day, 0:16:13 time: 0.5518 data_time: 0.0054 memory: 7635 loss: 0.0420 decode.loss_ce: 0.0264 decode.acc_seg: 98.8280 aux.loss_ce: 0.0156 aux.acc_seg: 98.5531 +04/17 02:39:58 - mmengine - INFO - Iter(train) [ 1900/160000] lr: 9.8942e-03 eta: 1 day, 0:15:43 time: 0.5517 data_time: 0.0054 memory: 7635 loss: 0.0354 decode.loss_ce: 0.0217 decode.acc_seg: 98.9604 aux.loss_ce: 0.0137 aux.acc_seg: 98.6360 +04/17 02:40:26 - mmengine - INFO - Iter(train) [ 1950/160000] lr: 9.8914e-03 eta: 1 day, 0:15:12 time: 0.5521 data_time: 0.0056 memory: 7635 loss: 0.0415 decode.loss_ce: 0.0258 decode.acc_seg: 98.5333 aux.loss_ce: 0.0157 aux.acc_seg: 97.9955 +04/17 02:40:53 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 02:40:53 - mmengine - INFO - Iter(train) [ 2000/160000] lr: 9.8886e-03 eta: 1 day, 0:14:40 time: 0.5519 data_time: 0.0055 memory: 7635 loss: 0.0353 decode.loss_ce: 0.0216 decode.acc_seg: 99.3915 aux.loss_ce: 0.0138 aux.acc_seg: 98.9317 +04/17 02:41:21 - mmengine - INFO - Iter(train) [ 2050/160000] lr: 9.8858e-03 eta: 1 day, 0:14:11 time: 0.5522 data_time: 0.0059 memory: 7635 loss: 0.0385 decode.loss_ce: 0.0235 decode.acc_seg: 99.2126 aux.loss_ce: 0.0149 aux.acc_seg: 98.5108 +04/17 02:41:48 - mmengine - INFO - Iter(train) [ 2100/160000] lr: 9.8830e-03 eta: 1 day, 0:13:44 time: 0.5515 data_time: 0.0056 memory: 7635 loss: 0.0391 decode.loss_ce: 0.0241 decode.acc_seg: 99.0423 aux.loss_ce: 0.0150 aux.acc_seg: 98.5409 +04/17 02:42:16 - mmengine - INFO - Iter(train) [ 2150/160000] lr: 9.8802e-03 eta: 1 day, 0:13:17 time: 0.5526 data_time: 0.0057 memory: 7635 loss: 0.0359 decode.loss_ce: 0.0218 decode.acc_seg: 99.0562 aux.loss_ce: 0.0141 aux.acc_seg: 98.5904 +04/17 02:42:44 - mmengine - INFO - Iter(train) [ 2200/160000] lr: 9.8775e-03 eta: 1 day, 0:12:47 time: 0.5511 data_time: 0.0059 memory: 7635 loss: 0.0396 decode.loss_ce: 0.0248 decode.acc_seg: 99.3354 aux.loss_ce: 0.0148 aux.acc_seg: 98.8929 +04/17 02:43:11 - mmengine - INFO - Iter(train) [ 2250/160000] lr: 9.8747e-03 eta: 1 day, 0:12:18 time: 0.5520 data_time: 0.0053 memory: 7635 loss: 0.0348 decode.loss_ce: 0.0217 decode.acc_seg: 99.0128 aux.loss_ce: 0.0131 aux.acc_seg: 98.8256 +04/17 02:43:39 - mmengine - INFO - Iter(train) [ 2300/160000] lr: 9.8719e-03 eta: 1 day, 0:11:49 time: 0.5513 data_time: 0.0064 memory: 7635 loss: 0.0336 decode.loss_ce: 0.0204 decode.acc_seg: 99.2620 aux.loss_ce: 0.0132 aux.acc_seg: 98.8499 +04/17 02:44:06 - mmengine - INFO - Iter(train) [ 2350/160000] lr: 9.8691e-03 eta: 1 day, 0:11:20 time: 0.5517 data_time: 0.0060 memory: 7635 loss: 0.0422 decode.loss_ce: 0.0266 decode.acc_seg: 99.2365 aux.loss_ce: 0.0156 aux.acc_seg: 98.7563 +04/17 02:44:34 - mmengine - INFO - Iter(train) [ 2400/160000] lr: 9.8663e-03 eta: 1 day, 0:10:52 time: 0.5512 data_time: 0.0055 memory: 7635 loss: 0.0345 decode.loss_ce: 0.0212 decode.acc_seg: 99.4081 aux.loss_ce: 0.0133 aux.acc_seg: 98.9314 +04/17 02:45:02 - mmengine - INFO - Iter(train) [ 2450/160000] lr: 9.8635e-03 eta: 1 day, 0:10:22 time: 0.5509 data_time: 0.0058 memory: 7635 loss: 0.0352 decode.loss_ce: 0.0220 decode.acc_seg: 99.0789 aux.loss_ce: 0.0132 aux.acc_seg: 98.7149 +04/17 02:45:29 - mmengine - INFO - Iter(train) [ 2500/160000] lr: 9.8607e-03 eta: 1 day, 0:09:52 time: 0.5530 data_time: 0.0055 memory: 7635 loss: 0.0353 decode.loss_ce: 0.0216 decode.acc_seg: 99.0977 aux.loss_ce: 0.0137 aux.acc_seg: 98.6426 +04/17 02:45:57 - mmengine - INFO - Iter(train) [ 2550/160000] lr: 9.8579e-03 eta: 1 day, 0:09:22 time: 0.5518 data_time: 0.0061 memory: 7635 loss: 0.0361 decode.loss_ce: 0.0225 decode.acc_seg: 99.1934 aux.loss_ce: 0.0136 aux.acc_seg: 98.7804 +04/17 02:46:24 - mmengine - INFO - Iter(train) [ 2600/160000] lr: 9.8551e-03 eta: 1 day, 0:08:52 time: 0.5525 data_time: 0.0057 memory: 7635 loss: 0.0341 decode.loss_ce: 0.0213 decode.acc_seg: 99.3256 aux.loss_ce: 0.0128 aux.acc_seg: 98.8477 +04/17 02:46:52 - mmengine - INFO - Iter(train) [ 2650/160000] lr: 9.8524e-03 eta: 1 day, 0:08:31 time: 0.5618 data_time: 0.0058 memory: 7635 loss: 0.0315 decode.loss_ce: 0.0190 decode.acc_seg: 99.3283 aux.loss_ce: 0.0125 aux.acc_seg: 98.7658 +04/17 02:47:20 - mmengine - INFO - Iter(train) [ 2700/160000] lr: 9.8496e-03 eta: 1 day, 0:08:03 time: 0.5523 data_time: 0.0055 memory: 7635 loss: 0.0339 decode.loss_ce: 0.0208 decode.acc_seg: 99.1889 aux.loss_ce: 0.0131 aux.acc_seg: 98.8381 +04/17 02:47:47 - mmengine - INFO - Iter(train) [ 2750/160000] lr: 9.8468e-03 eta: 1 day, 0:07:34 time: 0.5512 data_time: 0.0057 memory: 7635 loss: 0.0327 decode.loss_ce: 0.0198 decode.acc_seg: 99.3170 aux.loss_ce: 0.0129 aux.acc_seg: 98.8644 +04/17 02:48:15 - mmengine - INFO - Iter(train) [ 2800/160000] lr: 9.8440e-03 eta: 1 day, 0:07:07 time: 0.5529 data_time: 0.0056 memory: 7635 loss: 0.0387 decode.loss_ce: 0.0238 decode.acc_seg: 99.2949 aux.loss_ce: 0.0149 aux.acc_seg: 98.6554 +04/17 02:48:43 - mmengine - INFO - Iter(train) [ 2850/160000] lr: 9.8412e-03 eta: 1 day, 0:06:40 time: 0.5528 data_time: 0.0058 memory: 7635 loss: 0.0401 decode.loss_ce: 0.0253 decode.acc_seg: 99.1680 aux.loss_ce: 0.0148 aux.acc_seg: 98.7084 +04/17 02:49:10 - mmengine - INFO - Iter(train) [ 2900/160000] lr: 9.8384e-03 eta: 1 day, 0:06:12 time: 0.5518 data_time: 0.0068 memory: 7635 loss: 0.0348 decode.loss_ce: 0.0211 decode.acc_seg: 99.0792 aux.loss_ce: 0.0136 aux.acc_seg: 98.3262 +04/17 02:49:38 - mmengine - INFO - Iter(train) [ 2950/160000] lr: 9.8356e-03 eta: 1 day, 0:05:42 time: 0.5506 data_time: 0.0062 memory: 7635 loss: 0.0316 decode.loss_ce: 0.0192 decode.acc_seg: 99.1305 aux.loss_ce: 0.0124 aux.acc_seg: 98.6819 +04/17 02:50:05 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 02:50:05 - mmengine - INFO - Iter(train) [ 3000/160000] lr: 9.8328e-03 eta: 1 day, 0:05:14 time: 0.5524 data_time: 0.0065 memory: 7635 loss: 0.0376 decode.loss_ce: 0.0230 decode.acc_seg: 99.1773 aux.loss_ce: 0.0145 aux.acc_seg: 98.6571 +04/17 02:50:33 - mmengine - INFO - Iter(train) [ 3050/160000] lr: 9.8300e-03 eta: 1 day, 0:04:44 time: 0.5519 data_time: 0.0060 memory: 7635 loss: 0.0339 decode.loss_ce: 0.0208 decode.acc_seg: 99.2224 aux.loss_ce: 0.0131 aux.acc_seg: 98.7208 +04/17 02:51:01 - mmengine - INFO - Iter(train) [ 3100/160000] lr: 9.8273e-03 eta: 1 day, 0:04:16 time: 0.5527 data_time: 0.0060 memory: 7635 loss: 0.0426 decode.loss_ce: 0.0267 decode.acc_seg: 99.3018 aux.loss_ce: 0.0159 aux.acc_seg: 98.7821 +04/17 02:51:28 - mmengine - INFO - Iter(train) [ 3150/160000] lr: 9.8245e-03 eta: 1 day, 0:03:48 time: 0.5506 data_time: 0.0056 memory: 7635 loss: 0.0350 decode.loss_ce: 0.0212 decode.acc_seg: 99.3638 aux.loss_ce: 0.0138 aux.acc_seg: 98.8532 +04/17 02:51:56 - mmengine - INFO - Iter(train) [ 3200/160000] lr: 9.8217e-03 eta: 1 day, 0:03:19 time: 0.5509 data_time: 0.0067 memory: 7635 loss: 0.0366 decode.loss_ce: 0.0225 decode.acc_seg: 99.2182 aux.loss_ce: 0.0141 aux.acc_seg: 98.4915 +04/17 02:52:23 - mmengine - INFO - Iter(train) [ 3250/160000] lr: 9.8189e-03 eta: 1 day, 0:02:51 time: 0.5520 data_time: 0.0066 memory: 7635 loss: 0.0358 decode.loss_ce: 0.0221 decode.acc_seg: 98.7513 aux.loss_ce: 0.0137 aux.acc_seg: 98.3668 +04/17 02:52:51 - mmengine - INFO - Iter(train) [ 3300/160000] lr: 9.8161e-03 eta: 1 day, 0:02:23 time: 0.5529 data_time: 0.0059 memory: 7635 loss: 0.0363 decode.loss_ce: 0.0224 decode.acc_seg: 99.3017 aux.loss_ce: 0.0139 aux.acc_seg: 98.7112 +04/17 02:53:19 - mmengine - INFO - Iter(train) [ 3350/160000] lr: 9.8133e-03 eta: 1 day, 0:01:56 time: 0.5523 data_time: 0.0058 memory: 7635 loss: 0.0313 decode.loss_ce: 0.0188 decode.acc_seg: 99.4477 aux.loss_ce: 0.0125 aux.acc_seg: 98.9939 +04/17 02:53:46 - mmengine - INFO - Iter(train) [ 3400/160000] lr: 9.8105e-03 eta: 1 day, 0:01:28 time: 0.5535 data_time: 0.0059 memory: 7635 loss: 0.0349 decode.loss_ce: 0.0220 decode.acc_seg: 99.0543 aux.loss_ce: 0.0129 aux.acc_seg: 98.5743 +04/17 02:54:14 - mmengine - INFO - Iter(train) [ 3450/160000] lr: 9.8077e-03 eta: 1 day, 0:01:00 time: 0.5534 data_time: 0.0063 memory: 7635 loss: 0.0341 decode.loss_ce: 0.0209 decode.acc_seg: 99.4157 aux.loss_ce: 0.0132 aux.acc_seg: 98.9601 +04/17 02:54:41 - mmengine - INFO - Iter(train) [ 3500/160000] lr: 9.8049e-03 eta: 1 day, 0:00:32 time: 0.5527 data_time: 0.0062 memory: 7635 loss: 0.0352 decode.loss_ce: 0.0215 decode.acc_seg: 99.2278 aux.loss_ce: 0.0137 aux.acc_seg: 98.4088 +04/17 02:55:09 - mmengine - INFO - Iter(train) [ 3550/160000] lr: 9.8021e-03 eta: 1 day, 0:00:05 time: 0.5530 data_time: 0.0058 memory: 7635 loss: 0.0360 decode.loss_ce: 0.0221 decode.acc_seg: 99.0523 aux.loss_ce: 0.0139 aux.acc_seg: 98.5312 +04/17 02:55:37 - mmengine - INFO - Iter(train) [ 3600/160000] lr: 9.7994e-03 eta: 23:59:37 time: 0.5511 data_time: 0.0054 memory: 7635 loss: 0.0382 decode.loss_ce: 0.0237 decode.acc_seg: 98.8614 aux.loss_ce: 0.0145 aux.acc_seg: 98.4686 +04/17 02:56:04 - mmengine - INFO - Iter(train) [ 3650/160000] lr: 9.7966e-03 eta: 23:59:09 time: 0.5531 data_time: 0.0066 memory: 7635 loss: 0.0331 decode.loss_ce: 0.0204 decode.acc_seg: 99.1750 aux.loss_ce: 0.0127 aux.acc_seg: 98.7031 +04/17 02:56:32 - mmengine - INFO - Iter(train) [ 3700/160000] lr: 9.7938e-03 eta: 23:58:41 time: 0.5533 data_time: 0.0064 memory: 7635 loss: 0.0381 decode.loss_ce: 0.0238 decode.acc_seg: 98.8438 aux.loss_ce: 0.0143 aux.acc_seg: 98.4982 +04/17 02:57:00 - mmengine - INFO - Iter(train) [ 3750/160000] lr: 9.7910e-03 eta: 23:58:17 time: 0.5513 data_time: 0.0060 memory: 7635 loss: 0.0341 decode.loss_ce: 0.0204 decode.acc_seg: 99.2230 aux.loss_ce: 0.0137 aux.acc_seg: 98.6295 +04/17 02:57:27 - mmengine - INFO - Iter(train) [ 3800/160000] lr: 9.7882e-03 eta: 23:57:48 time: 0.5516 data_time: 0.0067 memory: 7635 loss: 0.0292 decode.loss_ce: 0.0178 decode.acc_seg: 99.3898 aux.loss_ce: 0.0115 aux.acc_seg: 98.7861 +04/17 02:57:55 - mmengine - INFO - Iter(train) [ 3850/160000] lr: 9.7854e-03 eta: 23:57:20 time: 0.5525 data_time: 0.0069 memory: 7635 loss: 0.0350 decode.loss_ce: 0.0216 decode.acc_seg: 99.2817 aux.loss_ce: 0.0134 aux.acc_seg: 98.6565 +04/17 02:58:22 - mmengine - INFO - Iter(train) [ 3900/160000] lr: 9.7826e-03 eta: 23:56:52 time: 0.5518 data_time: 0.0065 memory: 7635 loss: 0.0318 decode.loss_ce: 0.0194 decode.acc_seg: 99.3762 aux.loss_ce: 0.0125 aux.acc_seg: 98.8199 +04/17 02:58:50 - mmengine - INFO - Iter(train) [ 3950/160000] lr: 9.7798e-03 eta: 23:56:23 time: 0.5524 data_time: 0.0063 memory: 7635 loss: 0.0335 decode.loss_ce: 0.0205 decode.acc_seg: 99.2721 aux.loss_ce: 0.0130 aux.acc_seg: 98.8877 +04/17 02:59:18 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 02:59:18 - mmengine - INFO - Iter(train) [ 4000/160000] lr: 9.7770e-03 eta: 23:55:56 time: 0.5515 data_time: 0.0060 memory: 7635 loss: 0.0334 decode.loss_ce: 0.0204 decode.acc_seg: 98.9753 aux.loss_ce: 0.0129 aux.acc_seg: 98.5274 +04/17 02:59:45 - mmengine - INFO - Iter(train) [ 4050/160000] lr: 9.7742e-03 eta: 23:55:27 time: 0.5510 data_time: 0.0054 memory: 7635 loss: 0.0334 decode.loss_ce: 0.0201 decode.acc_seg: 99.1365 aux.loss_ce: 0.0133 aux.acc_seg: 98.4941 +04/17 03:00:13 - mmengine - INFO - Iter(train) [ 4100/160000] lr: 9.7714e-03 eta: 23:54:59 time: 0.5514 data_time: 0.0059 memory: 7635 loss: 0.0341 decode.loss_ce: 0.0207 decode.acc_seg: 99.2615 aux.loss_ce: 0.0134 aux.acc_seg: 98.8306 +04/17 03:00:40 - mmengine - INFO - Iter(train) [ 4150/160000] lr: 9.7686e-03 eta: 23:54:30 time: 0.5524 data_time: 0.0066 memory: 7635 loss: 0.0373 decode.loss_ce: 0.0232 decode.acc_seg: 99.0534 aux.loss_ce: 0.0141 aux.acc_seg: 98.2186 +04/17 03:01:08 - mmengine - INFO - Iter(train) [ 4200/160000] lr: 9.7659e-03 eta: 23:54:02 time: 0.5517 data_time: 0.0061 memory: 7635 loss: 0.0330 decode.loss_ce: 0.0200 decode.acc_seg: 98.9285 aux.loss_ce: 0.0130 aux.acc_seg: 98.4211 +04/17 03:01:36 - mmengine - INFO - Iter(train) [ 4250/160000] lr: 9.7631e-03 eta: 23:53:33 time: 0.5509 data_time: 0.0059 memory: 7635 loss: 0.0309 decode.loss_ce: 0.0187 decode.acc_seg: 99.2372 aux.loss_ce: 0.0122 aux.acc_seg: 98.9645 +04/17 03:02:03 - mmengine - INFO - Iter(train) [ 4300/160000] lr: 9.7603e-03 eta: 23:53:05 time: 0.5517 data_time: 0.0056 memory: 7635 loss: 0.0338 decode.loss_ce: 0.0206 decode.acc_seg: 99.0699 aux.loss_ce: 0.0133 aux.acc_seg: 98.4479 +04/17 03:02:31 - mmengine - INFO - Iter(train) [ 4350/160000] lr: 9.7575e-03 eta: 23:52:36 time: 0.5519 data_time: 0.0059 memory: 7635 loss: 0.0325 decode.loss_ce: 0.0198 decode.acc_seg: 99.1191 aux.loss_ce: 0.0127 aux.acc_seg: 98.4472 +04/17 03:02:58 - mmengine - INFO - Iter(train) [ 4400/160000] lr: 9.7547e-03 eta: 23:52:07 time: 0.5521 data_time: 0.0057 memory: 7635 loss: 0.0337 decode.loss_ce: 0.0209 decode.acc_seg: 99.3491 aux.loss_ce: 0.0128 aux.acc_seg: 98.8690 +04/17 03:03:26 - mmengine - INFO - Iter(train) [ 4450/160000] lr: 9.7519e-03 eta: 23:51:39 time: 0.5513 data_time: 0.0069 memory: 7635 loss: 0.0325 decode.loss_ce: 0.0197 decode.acc_seg: 98.8806 aux.loss_ce: 0.0127 aux.acc_seg: 98.5091 +04/17 03:03:53 - mmengine - INFO - Iter(train) [ 4500/160000] lr: 9.7491e-03 eta: 23:51:11 time: 0.5511 data_time: 0.0057 memory: 7635 loss: 0.0321 decode.loss_ce: 0.0196 decode.acc_seg: 99.3941 aux.loss_ce: 0.0125 aux.acc_seg: 98.9326 +04/17 03:04:21 - mmengine - INFO - Iter(train) [ 4550/160000] lr: 9.7463e-03 eta: 23:50:43 time: 0.5530 data_time: 0.0068 memory: 7635 loss: 0.0306 decode.loss_ce: 0.0186 decode.acc_seg: 99.2918 aux.loss_ce: 0.0120 aux.acc_seg: 98.9423 +04/17 03:04:49 - mmengine - INFO - Iter(train) [ 4600/160000] lr: 9.7435e-03 eta: 23:50:16 time: 0.5526 data_time: 0.0058 memory: 7635 loss: 0.0372 decode.loss_ce: 0.0230 decode.acc_seg: 99.0516 aux.loss_ce: 0.0142 aux.acc_seg: 98.4576 +04/17 03:05:16 - mmengine - INFO - Iter(train) [ 4650/160000] lr: 9.7407e-03 eta: 23:49:48 time: 0.5522 data_time: 0.0056 memory: 7635 loss: 0.0345 decode.loss_ce: 0.0215 decode.acc_seg: 99.6372 aux.loss_ce: 0.0130 aux.acc_seg: 99.1232 +04/17 03:05:44 - mmengine - INFO - Iter(train) [ 4700/160000] lr: 9.7379e-03 eta: 23:49:20 time: 0.5523 data_time: 0.0063 memory: 7635 loss: 0.0294 decode.loss_ce: 0.0178 decode.acc_seg: 98.9411 aux.loss_ce: 0.0117 aux.acc_seg: 98.4021 +04/17 03:06:12 - mmengine - INFO - Iter(train) [ 4750/160000] lr: 9.7351e-03 eta: 23:48:53 time: 0.5528 data_time: 0.0054 memory: 7635 loss: 0.0304 decode.loss_ce: 0.0187 decode.acc_seg: 99.2499 aux.loss_ce: 0.0117 aux.acc_seg: 98.7282 +04/17 03:06:39 - mmengine - INFO - Iter(train) [ 4800/160000] lr: 9.7323e-03 eta: 23:48:29 time: 0.5627 data_time: 0.0060 memory: 7635 loss: 0.0325 decode.loss_ce: 0.0196 decode.acc_seg: 99.1263 aux.loss_ce: 0.0129 aux.acc_seg: 98.6772 +04/17 03:07:07 - mmengine - INFO - Iter(train) [ 4850/160000] lr: 9.7296e-03 eta: 23:48:00 time: 0.5506 data_time: 0.0064 memory: 7635 loss: 0.0281 decode.loss_ce: 0.0165 decode.acc_seg: 99.5572 aux.loss_ce: 0.0117 aux.acc_seg: 99.1561 +04/17 03:07:34 - mmengine - INFO - Iter(train) [ 4900/160000] lr: 9.7268e-03 eta: 23:47:31 time: 0.5515 data_time: 0.0066 memory: 7635 loss: 0.0314 decode.loss_ce: 0.0194 decode.acc_seg: 99.3351 aux.loss_ce: 0.0119 aux.acc_seg: 98.7594 +04/17 03:08:02 - mmengine - INFO - Iter(train) [ 4950/160000] lr: 9.7240e-03 eta: 23:47:03 time: 0.5512 data_time: 0.0055 memory: 7635 loss: 0.0288 decode.loss_ce: 0.0180 decode.acc_seg: 99.1686 aux.loss_ce: 0.0108 aux.acc_seg: 98.8390 +04/17 03:08:30 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 03:08:30 - mmengine - INFO - Iter(train) [ 5000/160000] lr: 9.7212e-03 eta: 23:46:35 time: 0.5515 data_time: 0.0064 memory: 7635 loss: 0.0286 decode.loss_ce: 0.0173 decode.acc_seg: 99.5490 aux.loss_ce: 0.0114 aux.acc_seg: 99.1513 +04/17 03:08:57 - mmengine - INFO - Iter(train) [ 5050/160000] lr: 9.7184e-03 eta: 23:46:07 time: 0.5525 data_time: 0.0062 memory: 7635 loss: 0.0337 decode.loss_ce: 0.0212 decode.acc_seg: 99.1278 aux.loss_ce: 0.0125 aux.acc_seg: 98.8377 +04/17 03:09:25 - mmengine - INFO - Iter(train) [ 5100/160000] lr: 9.7156e-03 eta: 23:45:39 time: 0.5514 data_time: 0.0059 memory: 7635 loss: 0.0282 decode.loss_ce: 0.0168 decode.acc_seg: 99.4632 aux.loss_ce: 0.0114 aux.acc_seg: 98.9523 +04/17 03:09:52 - mmengine - INFO - Iter(train) [ 5150/160000] lr: 9.7128e-03 eta: 23:45:12 time: 0.5514 data_time: 0.0054 memory: 7635 loss: 0.0318 decode.loss_ce: 0.0193 decode.acc_seg: 99.0087 aux.loss_ce: 0.0125 aux.acc_seg: 98.6403 +04/17 03:10:20 - mmengine - INFO - Iter(train) [ 5200/160000] lr: 9.7100e-03 eta: 23:44:44 time: 0.5509 data_time: 0.0053 memory: 7635 loss: 0.0306 decode.loss_ce: 0.0185 decode.acc_seg: 99.2336 aux.loss_ce: 0.0121 aux.acc_seg: 98.7778 +04/17 03:10:48 - mmengine - INFO - Iter(train) [ 5250/160000] lr: 9.7072e-03 eta: 23:44:16 time: 0.5518 data_time: 0.0056 memory: 7635 loss: 0.0302 decode.loss_ce: 0.0185 decode.acc_seg: 99.0332 aux.loss_ce: 0.0118 aux.acc_seg: 98.3864 +04/17 03:11:15 - mmengine - INFO - Iter(train) [ 5300/160000] lr: 9.7044e-03 eta: 23:43:46 time: 0.5511 data_time: 0.0065 memory: 7635 loss: 0.0294 decode.loss_ce: 0.0179 decode.acc_seg: 99.0035 aux.loss_ce: 0.0115 aux.acc_seg: 98.5707 +04/17 03:11:43 - mmengine - INFO - Iter(train) [ 5350/160000] lr: 9.7016e-03 eta: 23:43:19 time: 0.5541 data_time: 0.0057 memory: 7635 loss: 0.0297 decode.loss_ce: 0.0180 decode.acc_seg: 99.4402 aux.loss_ce: 0.0118 aux.acc_seg: 98.9441 +04/17 03:12:10 - mmengine - INFO - Iter(train) [ 5400/160000] lr: 9.6988e-03 eta: 23:42:52 time: 0.5518 data_time: 0.0070 memory: 7635 loss: 0.0278 decode.loss_ce: 0.0167 decode.acc_seg: 99.0986 aux.loss_ce: 0.0111 aux.acc_seg: 98.6318 +04/17 03:12:38 - mmengine - INFO - Iter(train) [ 5450/160000] lr: 9.6960e-03 eta: 23:42:24 time: 0.5511 data_time: 0.0056 memory: 7635 loss: 0.0297 decode.loss_ce: 0.0182 decode.acc_seg: 99.2891 aux.loss_ce: 0.0115 aux.acc_seg: 98.8083 +04/17 03:13:06 - mmengine - INFO - Iter(train) [ 5500/160000] lr: 9.6932e-03 eta: 23:41:56 time: 0.5529 data_time: 0.0058 memory: 7635 loss: 0.0324 decode.loss_ce: 0.0201 decode.acc_seg: 99.2442 aux.loss_ce: 0.0123 aux.acc_seg: 98.7219 +04/17 03:13:33 - mmengine - INFO - Iter(train) [ 5550/160000] lr: 9.6904e-03 eta: 23:41:28 time: 0.5511 data_time: 0.0057 memory: 7635 loss: 0.0346 decode.loss_ce: 0.0215 decode.acc_seg: 99.3938 aux.loss_ce: 0.0131 aux.acc_seg: 98.7620 +04/17 03:14:01 - mmengine - INFO - Iter(train) [ 5600/160000] lr: 9.6877e-03 eta: 23:41:01 time: 0.5527 data_time: 0.0060 memory: 7635 loss: 0.0288 decode.loss_ce: 0.0173 decode.acc_seg: 99.5282 aux.loss_ce: 0.0115 aux.acc_seg: 99.1069 +04/17 03:14:28 - mmengine - INFO - Iter(train) [ 5650/160000] lr: 9.6849e-03 eta: 23:40:33 time: 0.5496 data_time: 0.0059 memory: 7635 loss: 0.0275 decode.loss_ce: 0.0164 decode.acc_seg: 99.3816 aux.loss_ce: 0.0111 aux.acc_seg: 99.0400 +04/17 03:14:56 - mmengine - INFO - Iter(train) [ 5700/160000] lr: 9.6821e-03 eta: 23:40:06 time: 0.5523 data_time: 0.0056 memory: 7635 loss: 0.0277 decode.loss_ce: 0.0166 decode.acc_seg: 99.2375 aux.loss_ce: 0.0111 aux.acc_seg: 98.7439 +04/17 03:15:24 - mmengine - INFO - Iter(train) [ 5750/160000] lr: 9.6793e-03 eta: 23:39:39 time: 0.5535 data_time: 0.0062 memory: 7635 loss: 0.0312 decode.loss_ce: 0.0186 decode.acc_seg: 98.9881 aux.loss_ce: 0.0126 aux.acc_seg: 98.3813 +04/17 03:15:51 - mmengine - INFO - Iter(train) [ 5800/160000] lr: 9.6765e-03 eta: 23:39:12 time: 0.5531 data_time: 0.0058 memory: 7635 loss: 0.0297 decode.loss_ce: 0.0179 decode.acc_seg: 99.3886 aux.loss_ce: 0.0118 aux.acc_seg: 98.9565 +04/17 03:16:19 - mmengine - INFO - Iter(train) [ 5850/160000] lr: 9.6737e-03 eta: 23:38:44 time: 0.5518 data_time: 0.0070 memory: 7635 loss: 0.0289 decode.loss_ce: 0.0171 decode.acc_seg: 99.4504 aux.loss_ce: 0.0119 aux.acc_seg: 98.9906 +04/17 03:16:47 - mmengine - INFO - Iter(train) [ 5900/160000] lr: 9.6709e-03 eta: 23:38:20 time: 0.5522 data_time: 0.0061 memory: 7635 loss: 0.0364 decode.loss_ce: 0.0226 decode.acc_seg: 98.7685 aux.loss_ce: 0.0138 aux.acc_seg: 98.1620 +04/17 03:17:14 - mmengine - INFO - Iter(train) [ 5950/160000] lr: 9.6681e-03 eta: 23:37:53 time: 0.5535 data_time: 0.0061 memory: 7635 loss: 0.0271 decode.loss_ce: 0.0166 decode.acc_seg: 99.4514 aux.loss_ce: 0.0104 aux.acc_seg: 99.1319 +04/17 03:17:42 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 03:17:42 - mmengine - INFO - Iter(train) [ 6000/160000] lr: 9.6653e-03 eta: 23:37:26 time: 0.5519 data_time: 0.0057 memory: 7635 loss: 0.0291 decode.loss_ce: 0.0174 decode.acc_seg: 99.1006 aux.loss_ce: 0.0118 aux.acc_seg: 98.7028 +04/17 03:18:10 - mmengine - INFO - Iter(train) [ 6050/160000] lr: 9.6625e-03 eta: 23:36:58 time: 0.5511 data_time: 0.0062 memory: 7635 loss: 0.0312 decode.loss_ce: 0.0186 decode.acc_seg: 99.3483 aux.loss_ce: 0.0126 aux.acc_seg: 98.9719 +04/17 03:18:37 - mmengine - INFO - Iter(train) [ 6100/160000] lr: 9.6597e-03 eta: 23:36:31 time: 0.5528 data_time: 0.0056 memory: 7635 loss: 0.0310 decode.loss_ce: 0.0192 decode.acc_seg: 99.4406 aux.loss_ce: 0.0118 aux.acc_seg: 99.0075 +04/17 03:19:05 - mmengine - INFO - Iter(train) [ 6150/160000] lr: 9.6569e-03 eta: 23:36:03 time: 0.5513 data_time: 0.0055 memory: 7635 loss: 0.0270 decode.loss_ce: 0.0163 decode.acc_seg: 98.9075 aux.loss_ce: 0.0106 aux.acc_seg: 98.4277 +04/17 03:19:32 - mmengine - INFO - Iter(train) [ 6200/160000] lr: 9.6541e-03 eta: 23:35:36 time: 0.5532 data_time: 0.0060 memory: 7635 loss: 0.0322 decode.loss_ce: 0.0196 decode.acc_seg: 98.6740 aux.loss_ce: 0.0126 aux.acc_seg: 98.2050 +04/17 03:20:00 - mmengine - INFO - Iter(train) [ 6250/160000] lr: 9.6513e-03 eta: 23:35:08 time: 0.5516 data_time: 0.0060 memory: 7635 loss: 0.0260 decode.loss_ce: 0.0154 decode.acc_seg: 99.5450 aux.loss_ce: 0.0106 aux.acc_seg: 99.1028 +04/17 03:20:28 - mmengine - INFO - Iter(train) [ 6300/160000] lr: 9.6485e-03 eta: 23:34:41 time: 0.5524 data_time: 0.0061 memory: 7635 loss: 0.0299 decode.loss_ce: 0.0178 decode.acc_seg: 99.3025 aux.loss_ce: 0.0121 aux.acc_seg: 98.9420 +04/17 03:20:55 - mmengine - INFO - Iter(train) [ 6350/160000] lr: 9.6457e-03 eta: 23:34:14 time: 0.5526 data_time: 0.0057 memory: 7635 loss: 0.0296 decode.loss_ce: 0.0176 decode.acc_seg: 99.2557 aux.loss_ce: 0.0120 aux.acc_seg: 98.8164 +04/17 03:21:23 - mmengine - INFO - Iter(train) [ 6400/160000] lr: 9.6429e-03 eta: 23:33:46 time: 0.5516 data_time: 0.0057 memory: 7635 loss: 0.0338 decode.loss_ce: 0.0206 decode.acc_seg: 99.1667 aux.loss_ce: 0.0132 aux.acc_seg: 98.7056 +04/17 03:21:51 - mmengine - INFO - Iter(train) [ 6450/160000] lr: 9.6401e-03 eta: 23:33:19 time: 0.5544 data_time: 0.0062 memory: 7635 loss: 0.0317 decode.loss_ce: 0.0193 decode.acc_seg: 99.4510 aux.loss_ce: 0.0124 aux.acc_seg: 98.9374 +04/17 03:22:18 - mmengine - INFO - Iter(train) [ 6500/160000] lr: 9.6373e-03 eta: 23:32:52 time: 0.5529 data_time: 0.0069 memory: 7635 loss: 0.0294 decode.loss_ce: 0.0177 decode.acc_seg: 98.8971 aux.loss_ce: 0.0117 aux.acc_seg: 98.3251 +04/17 03:22:46 - mmengine - INFO - Iter(train) [ 6550/160000] lr: 9.6345e-03 eta: 23:32:25 time: 0.5551 data_time: 0.0058 memory: 7635 loss: 0.0325 decode.loss_ce: 0.0202 decode.acc_seg: 99.1228 aux.loss_ce: 0.0123 aux.acc_seg: 98.6965 +04/17 03:23:13 - mmengine - INFO - Iter(train) [ 6600/160000] lr: 9.6317e-03 eta: 23:31:58 time: 0.5529 data_time: 0.0055 memory: 7635 loss: 0.0299 decode.loss_ce: 0.0184 decode.acc_seg: 99.4324 aux.loss_ce: 0.0115 aux.acc_seg: 98.8950 +04/17 03:23:41 - mmengine - INFO - Iter(train) [ 6650/160000] lr: 9.6290e-03 eta: 23:31:30 time: 0.5528 data_time: 0.0058 memory: 7635 loss: 0.0275 decode.loss_ce: 0.0167 decode.acc_seg: 99.4091 aux.loss_ce: 0.0109 aux.acc_seg: 98.9687 +04/17 03:24:09 - mmengine - INFO - Iter(train) [ 6700/160000] lr: 9.6262e-03 eta: 23:31:03 time: 0.5526 data_time: 0.0059 memory: 7635 loss: 0.0307 decode.loss_ce: 0.0186 decode.acc_seg: 99.3761 aux.loss_ce: 0.0121 aux.acc_seg: 99.0348 +04/17 03:24:36 - mmengine - INFO - Iter(train) [ 6750/160000] lr: 9.6234e-03 eta: 23:30:35 time: 0.5523 data_time: 0.0055 memory: 7635 loss: 0.0294 decode.loss_ce: 0.0175 decode.acc_seg: 99.4396 aux.loss_ce: 0.0119 aux.acc_seg: 98.8461 +04/17 03:25:04 - mmengine - INFO - Iter(train) [ 6800/160000] lr: 9.6206e-03 eta: 23:30:07 time: 0.5510 data_time: 0.0060 memory: 7635 loss: 0.0246 decode.loss_ce: 0.0145 decode.acc_seg: 99.3783 aux.loss_ce: 0.0101 aux.acc_seg: 98.7988 +04/17 03:25:32 - mmengine - INFO - Iter(train) [ 6850/160000] lr: 9.6178e-03 eta: 23:29:40 time: 0.5525 data_time: 0.0065 memory: 7635 loss: 0.0257 decode.loss_ce: 0.0152 decode.acc_seg: 99.1746 aux.loss_ce: 0.0105 aux.acc_seg: 98.7113 +04/17 03:25:59 - mmengine - INFO - Iter(train) [ 6900/160000] lr: 9.6150e-03 eta: 23:29:12 time: 0.5517 data_time: 0.0059 memory: 7635 loss: 0.0277 decode.loss_ce: 0.0165 decode.acc_seg: 99.0493 aux.loss_ce: 0.0112 aux.acc_seg: 98.6212 +04/17 03:26:27 - mmengine - INFO - Iter(train) [ 6950/160000] lr: 9.6122e-03 eta: 23:28:46 time: 0.5517 data_time: 0.0056 memory: 7635 loss: 0.0263 decode.loss_ce: 0.0153 decode.acc_seg: 99.5555 aux.loss_ce: 0.0110 aux.acc_seg: 99.0334 +04/17 03:26:54 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 03:26:54 - mmengine - INFO - Iter(train) [ 7000/160000] lr: 9.6094e-03 eta: 23:28:18 time: 0.5529 data_time: 0.0063 memory: 7635 loss: 0.0264 decode.loss_ce: 0.0157 decode.acc_seg: 99.3204 aux.loss_ce: 0.0107 aux.acc_seg: 98.9942 +04/17 03:27:22 - mmengine - INFO - Iter(train) [ 7050/160000] lr: 9.6066e-03 eta: 23:27:51 time: 0.5532 data_time: 0.0065 memory: 7635 loss: 0.0311 decode.loss_ce: 0.0192 decode.acc_seg: 99.1666 aux.loss_ce: 0.0119 aux.acc_seg: 98.8470 +04/17 03:27:50 - mmengine - INFO - Iter(train) [ 7100/160000] lr: 9.6038e-03 eta: 23:27:24 time: 0.5513 data_time: 0.0063 memory: 7635 loss: 0.0267 decode.loss_ce: 0.0158 decode.acc_seg: 99.3400 aux.loss_ce: 0.0109 aux.acc_seg: 98.8838 +04/17 03:28:17 - mmengine - INFO - Iter(train) [ 7150/160000] lr: 9.6010e-03 eta: 23:26:57 time: 0.5530 data_time: 0.0054 memory: 7635 loss: 0.0306 decode.loss_ce: 0.0189 decode.acc_seg: 99.6149 aux.loss_ce: 0.0117 aux.acc_seg: 99.1050 +04/17 03:28:45 - mmengine - INFO - Iter(train) [ 7200/160000] lr: 9.5982e-03 eta: 23:26:29 time: 0.5523 data_time: 0.0061 memory: 7635 loss: 0.0276 decode.loss_ce: 0.0165 decode.acc_seg: 98.9423 aux.loss_ce: 0.0111 aux.acc_seg: 98.6954 +04/17 03:29:13 - mmengine - INFO - Iter(train) [ 7250/160000] lr: 9.5954e-03 eta: 23:26:01 time: 0.5539 data_time: 0.0060 memory: 7635 loss: 0.0280 decode.loss_ce: 0.0169 decode.acc_seg: 99.4061 aux.loss_ce: 0.0111 aux.acc_seg: 98.7851 +04/17 03:29:40 - mmengine - INFO - Iter(train) [ 7300/160000] lr: 9.5926e-03 eta: 23:25:34 time: 0.5526 data_time: 0.0061 memory: 7635 loss: 0.0241 decode.loss_ce: 0.0140 decode.acc_seg: 99.5446 aux.loss_ce: 0.0101 aux.acc_seg: 99.1001 +04/17 03:30:08 - mmengine - INFO - Iter(train) [ 7350/160000] lr: 9.5898e-03 eta: 23:25:06 time: 0.5530 data_time: 0.0060 memory: 7635 loss: 0.0252 decode.loss_ce: 0.0150 decode.acc_seg: 99.4038 aux.loss_ce: 0.0103 aux.acc_seg: 98.9410 +04/17 03:30:35 - mmengine - INFO - Iter(train) [ 7400/160000] lr: 9.5870e-03 eta: 23:24:38 time: 0.5507 data_time: 0.0061 memory: 7635 loss: 0.0272 decode.loss_ce: 0.0165 decode.acc_seg: 99.5064 aux.loss_ce: 0.0107 aux.acc_seg: 99.0931 +04/17 03:31:03 - mmengine - INFO - Iter(train) [ 7450/160000] lr: 9.5842e-03 eta: 23:24:11 time: 0.5527 data_time: 0.0063 memory: 7635 loss: 0.0241 decode.loss_ce: 0.0142 decode.acc_seg: 99.5798 aux.loss_ce: 0.0099 aux.acc_seg: 99.0539 +04/17 03:31:31 - mmengine - INFO - Iter(train) [ 7500/160000] lr: 9.5814e-03 eta: 23:23:43 time: 0.5523 data_time: 0.0059 memory: 7635 loss: 0.0274 decode.loss_ce: 0.0161 decode.acc_seg: 99.2188 aux.loss_ce: 0.0113 aux.acc_seg: 98.9532 +04/17 03:31:58 - mmengine - INFO - Iter(train) [ 7550/160000] lr: 9.5786e-03 eta: 23:23:16 time: 0.5516 data_time: 0.0061 memory: 7635 loss: 0.0270 decode.loss_ce: 0.0161 decode.acc_seg: 99.1962 aux.loss_ce: 0.0109 aux.acc_seg: 98.7874 +04/17 03:32:26 - mmengine - INFO - Iter(train) [ 7600/160000] lr: 9.5758e-03 eta: 23:22:49 time: 0.5527 data_time: 0.0062 memory: 7635 loss: 0.0253 decode.loss_ce: 0.0148 decode.acc_seg: 99.4922 aux.loss_ce: 0.0105 aux.acc_seg: 98.8668 +04/17 03:32:54 - mmengine - INFO - Iter(train) [ 7650/160000] lr: 9.5730e-03 eta: 23:22:22 time: 0.5539 data_time: 0.0061 memory: 7635 loss: 0.0279 decode.loss_ce: 0.0166 decode.acc_seg: 99.3082 aux.loss_ce: 0.0113 aux.acc_seg: 98.6896 +04/17 03:33:21 - mmengine - INFO - Iter(train) [ 7700/160000] lr: 9.5702e-03 eta: 23:21:54 time: 0.5522 data_time: 0.0060 memory: 7635 loss: 0.0324 decode.loss_ce: 0.0199 decode.acc_seg: 99.2034 aux.loss_ce: 0.0125 aux.acc_seg: 98.7519 +04/17 03:33:49 - mmengine - INFO - Iter(train) [ 7750/160000] lr: 9.5674e-03 eta: 23:21:26 time: 0.5511 data_time: 0.0061 memory: 7635 loss: 0.0269 decode.loss_ce: 0.0159 decode.acc_seg: 99.3290 aux.loss_ce: 0.0110 aux.acc_seg: 98.9721 +04/17 03:34:16 - mmengine - INFO - Iter(train) [ 7800/160000] lr: 9.5646e-03 eta: 23:20:59 time: 0.5518 data_time: 0.0056 memory: 7635 loss: 0.0258 decode.loss_ce: 0.0151 decode.acc_seg: 99.6602 aux.loss_ce: 0.0107 aux.acc_seg: 99.2991 +04/17 03:34:44 - mmengine - INFO - Iter(train) [ 7850/160000] lr: 9.5618e-03 eta: 23:20:32 time: 0.5516 data_time: 0.0063 memory: 7635 loss: 0.0287 decode.loss_ce: 0.0172 decode.acc_seg: 98.8665 aux.loss_ce: 0.0115 aux.acc_seg: 98.2397 +04/17 03:35:12 - mmengine - INFO - Iter(train) [ 7900/160000] lr: 9.5590e-03 eta: 23:20:04 time: 0.5533 data_time: 0.0060 memory: 7635 loss: 0.0266 decode.loss_ce: 0.0162 decode.acc_seg: 99.3392 aux.loss_ce: 0.0104 aux.acc_seg: 98.9814 +04/17 03:35:39 - mmengine - INFO - Iter(train) [ 7950/160000] lr: 9.5562e-03 eta: 23:19:37 time: 0.5534 data_time: 0.0063 memory: 7635 loss: 0.0267 decode.loss_ce: 0.0160 decode.acc_seg: 99.5321 aux.loss_ce: 0.0107 aux.acc_seg: 99.1002 +04/17 03:36:07 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 03:36:07 - mmengine - INFO - Iter(train) [ 8000/160000] lr: 9.5534e-03 eta: 23:19:10 time: 0.5527 data_time: 0.0067 memory: 7635 loss: 0.0295 decode.loss_ce: 0.0178 decode.acc_seg: 99.3558 aux.loss_ce: 0.0117 aux.acc_seg: 98.8987 +04/17 03:36:35 - mmengine - INFO - Iter(train) [ 8050/160000] lr: 9.5506e-03 eta: 23:18:44 time: 0.5527 data_time: 0.0061 memory: 7635 loss: 0.0251 decode.loss_ce: 0.0147 decode.acc_seg: 99.4716 aux.loss_ce: 0.0104 aux.acc_seg: 98.8608 +04/17 03:37:02 - mmengine - INFO - Iter(train) [ 8100/160000] lr: 9.5478e-03 eta: 23:18:17 time: 0.5523 data_time: 0.0063 memory: 7635 loss: 0.0285 decode.loss_ce: 0.0173 decode.acc_seg: 99.4266 aux.loss_ce: 0.0112 aux.acc_seg: 98.9121 +04/17 03:37:30 - mmengine - INFO - Iter(train) [ 8150/160000] lr: 9.5450e-03 eta: 23:17:49 time: 0.5522 data_time: 0.0061 memory: 7635 loss: 0.0281 decode.loss_ce: 0.0166 decode.acc_seg: 99.0971 aux.loss_ce: 0.0115 aux.acc_seg: 98.5067 +04/17 03:37:58 - mmengine - INFO - Iter(train) [ 8200/160000] lr: 9.5422e-03 eta: 23:17:22 time: 0.5527 data_time: 0.0068 memory: 7635 loss: 0.0264 decode.loss_ce: 0.0160 decode.acc_seg: 99.3627 aux.loss_ce: 0.0104 aux.acc_seg: 98.9357 +04/17 03:38:25 - mmengine - INFO - Iter(train) [ 8250/160000] lr: 9.5394e-03 eta: 23:16:54 time: 0.5516 data_time: 0.0059 memory: 7635 loss: 0.0266 decode.loss_ce: 0.0158 decode.acc_seg: 99.4013 aux.loss_ce: 0.0108 aux.acc_seg: 98.8547 +04/17 03:38:53 - mmengine - INFO - Iter(train) [ 8300/160000] lr: 9.5366e-03 eta: 23:16:26 time: 0.5532 data_time: 0.0057 memory: 7635 loss: 0.0235 decode.loss_ce: 0.0134 decode.acc_seg: 99.5517 aux.loss_ce: 0.0100 aux.acc_seg: 99.0553 +04/17 03:39:20 - mmengine - INFO - Iter(train) [ 8350/160000] lr: 9.5338e-03 eta: 23:15:59 time: 0.5520 data_time: 0.0060 memory: 7635 loss: 0.0293 decode.loss_ce: 0.0174 decode.acc_seg: 99.0095 aux.loss_ce: 0.0119 aux.acc_seg: 98.5011 +04/17 03:39:48 - mmengine - INFO - Iter(train) [ 8400/160000] lr: 9.5310e-03 eta: 23:15:31 time: 0.5519 data_time: 0.0060 memory: 7635 loss: 0.0258 decode.loss_ce: 0.0151 decode.acc_seg: 99.3837 aux.loss_ce: 0.0107 aux.acc_seg: 98.9029 +04/17 03:40:16 - mmengine - INFO - Iter(train) [ 8450/160000] lr: 9.5282e-03 eta: 23:15:03 time: 0.5530 data_time: 0.0062 memory: 7635 loss: 0.0244 decode.loss_ce: 0.0144 decode.acc_seg: 99.1686 aux.loss_ce: 0.0100 aux.acc_seg: 98.6112 +04/17 03:40:43 - mmengine - INFO - Iter(train) [ 8500/160000] lr: 9.5254e-03 eta: 23:14:36 time: 0.5515 data_time: 0.0056 memory: 7635 loss: 0.0298 decode.loss_ce: 0.0180 decode.acc_seg: 99.2611 aux.loss_ce: 0.0118 aux.acc_seg: 98.9025 +04/17 03:41:11 - mmengine - INFO - Iter(train) [ 8550/160000] lr: 9.5226e-03 eta: 23:14:08 time: 0.5522 data_time: 0.0058 memory: 7635 loss: 0.0266 decode.loss_ce: 0.0157 decode.acc_seg: 99.4311 aux.loss_ce: 0.0109 aux.acc_seg: 99.0005 +04/17 03:41:38 - mmengine - INFO - Iter(train) [ 8600/160000] lr: 9.5198e-03 eta: 23:13:40 time: 0.5507 data_time: 0.0071 memory: 7635 loss: 0.0259 decode.loss_ce: 0.0153 decode.acc_seg: 99.2587 aux.loss_ce: 0.0106 aux.acc_seg: 98.9434 +04/17 03:42:06 - mmengine - INFO - Iter(train) [ 8650/160000] lr: 9.5170e-03 eta: 23:13:12 time: 0.5528 data_time: 0.0061 memory: 7635 loss: 0.0249 decode.loss_ce: 0.0145 decode.acc_seg: 99.4470 aux.loss_ce: 0.0104 aux.acc_seg: 98.9309 +04/17 03:42:34 - mmengine - INFO - Iter(train) [ 8700/160000] lr: 9.5142e-03 eta: 23:12:45 time: 0.5530 data_time: 0.0064 memory: 7635 loss: 0.0278 decode.loss_ce: 0.0165 decode.acc_seg: 99.1575 aux.loss_ce: 0.0113 aux.acc_seg: 98.7628 +04/17 03:43:01 - mmengine - INFO - Iter(train) [ 8750/160000] lr: 9.5114e-03 eta: 23:12:18 time: 0.5524 data_time: 0.0057 memory: 7635 loss: 0.0281 decode.loss_ce: 0.0168 decode.acc_seg: 99.1311 aux.loss_ce: 0.0113 aux.acc_seg: 98.7585 +04/17 03:43:29 - mmengine - INFO - Iter(train) [ 8800/160000] lr: 9.5086e-03 eta: 23:11:51 time: 0.5521 data_time: 0.0059 memory: 7635 loss: 0.0232 decode.loss_ce: 0.0135 decode.acc_seg: 99.5893 aux.loss_ce: 0.0096 aux.acc_seg: 99.2972 +04/17 03:43:57 - mmengine - INFO - Iter(train) [ 8850/160000] lr: 9.5058e-03 eta: 23:11:23 time: 0.5520 data_time: 0.0059 memory: 7635 loss: 0.0263 decode.loss_ce: 0.0157 decode.acc_seg: 99.4049 aux.loss_ce: 0.0106 aux.acc_seg: 98.8473 +04/17 03:44:24 - mmengine - INFO - Iter(train) [ 8900/160000] lr: 9.5030e-03 eta: 23:10:56 time: 0.5532 data_time: 0.0061 memory: 7635 loss: 0.0241 decode.loss_ce: 0.0139 decode.acc_seg: 99.5001 aux.loss_ce: 0.0102 aux.acc_seg: 99.1658 +04/17 03:44:52 - mmengine - INFO - Iter(train) [ 8950/160000] lr: 9.5002e-03 eta: 23:10:28 time: 0.5510 data_time: 0.0063 memory: 7635 loss: 0.0274 decode.loss_ce: 0.0162 decode.acc_seg: 99.3296 aux.loss_ce: 0.0112 aux.acc_seg: 98.7326 +04/17 03:45:19 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 03:45:19 - mmengine - INFO - Iter(train) [ 9000/160000] lr: 9.4974e-03 eta: 23:10:01 time: 0.5531 data_time: 0.0057 memory: 7635 loss: 0.0235 decode.loss_ce: 0.0137 decode.acc_seg: 99.5480 aux.loss_ce: 0.0098 aux.acc_seg: 99.1562 +04/17 03:45:47 - mmengine - INFO - Iter(train) [ 9050/160000] lr: 9.4946e-03 eta: 23:09:34 time: 0.5523 data_time: 0.0058 memory: 7635 loss: 0.0311 decode.loss_ce: 0.0187 decode.acc_seg: 99.3399 aux.loss_ce: 0.0124 aux.acc_seg: 98.9336 +04/17 03:46:15 - mmengine - INFO - Iter(train) [ 9100/160000] lr: 9.4918e-03 eta: 23:09:08 time: 0.5631 data_time: 0.0062 memory: 7635 loss: 0.0253 decode.loss_ce: 0.0149 decode.acc_seg: 99.4165 aux.loss_ce: 0.0104 aux.acc_seg: 98.9483 +04/17 03:46:42 - mmengine - INFO - Iter(train) [ 9150/160000] lr: 9.4890e-03 eta: 23:08:41 time: 0.5510 data_time: 0.0054 memory: 7635 loss: 0.0254 decode.loss_ce: 0.0148 decode.acc_seg: 99.1962 aux.loss_ce: 0.0106 aux.acc_seg: 98.4969 +04/17 03:47:10 - mmengine - INFO - Iter(train) [ 9200/160000] lr: 9.4862e-03 eta: 23:08:13 time: 0.5518 data_time: 0.0058 memory: 7635 loss: 0.0254 decode.loss_ce: 0.0148 decode.acc_seg: 99.3896 aux.loss_ce: 0.0106 aux.acc_seg: 98.8047 +04/17 03:47:38 - mmengine - INFO - Iter(train) [ 9250/160000] lr: 9.4834e-03 eta: 23:07:45 time: 0.5521 data_time: 0.0063 memory: 7635 loss: 0.0248 decode.loss_ce: 0.0146 decode.acc_seg: 99.2444 aux.loss_ce: 0.0102 aux.acc_seg: 98.7537 +04/17 03:48:05 - mmengine - INFO - Iter(train) [ 9300/160000] lr: 9.4806e-03 eta: 23:07:18 time: 0.5514 data_time: 0.0058 memory: 7635 loss: 0.0232 decode.loss_ce: 0.0137 decode.acc_seg: 99.3546 aux.loss_ce: 0.0095 aux.acc_seg: 98.9738 +04/17 03:48:33 - mmengine - INFO - Iter(train) [ 9350/160000] lr: 9.4778e-03 eta: 23:06:50 time: 0.5502 data_time: 0.0061 memory: 7635 loss: 0.0275 decode.loss_ce: 0.0161 decode.acc_seg: 99.5212 aux.loss_ce: 0.0114 aux.acc_seg: 99.1037 +04/17 03:49:01 - mmengine - INFO - Iter(train) [ 9400/160000] lr: 9.4750e-03 eta: 23:06:22 time: 0.5527 data_time: 0.0059 memory: 7635 loss: 0.0268 decode.loss_ce: 0.0158 decode.acc_seg: 99.4898 aux.loss_ce: 0.0109 aux.acc_seg: 98.9123 +04/17 03:49:28 - mmengine - INFO - Iter(train) [ 9450/160000] lr: 9.4722e-03 eta: 23:05:55 time: 0.5518 data_time: 0.0060 memory: 7635 loss: 0.0246 decode.loss_ce: 0.0143 decode.acc_seg: 99.2866 aux.loss_ce: 0.0103 aux.acc_seg: 98.7805 +04/17 03:49:56 - mmengine - INFO - Iter(train) [ 9500/160000] lr: 9.4694e-03 eta: 23:05:26 time: 0.5505 data_time: 0.0059 memory: 7635 loss: 0.0258 decode.loss_ce: 0.0152 decode.acc_seg: 99.2098 aux.loss_ce: 0.0106 aux.acc_seg: 98.6047 +04/17 03:50:23 - mmengine - INFO - Iter(train) [ 9550/160000] lr: 9.4666e-03 eta: 23:04:58 time: 0.5512 data_time: 0.0059 memory: 7635 loss: 0.0269 decode.loss_ce: 0.0160 decode.acc_seg: 99.4492 aux.loss_ce: 0.0108 aux.acc_seg: 98.9509 +04/17 03:50:51 - mmengine - INFO - Iter(train) [ 9600/160000] lr: 9.4638e-03 eta: 23:04:30 time: 0.5515 data_time: 0.0064 memory: 7635 loss: 0.0250 decode.loss_ce: 0.0148 decode.acc_seg: 99.5263 aux.loss_ce: 0.0102 aux.acc_seg: 99.0936 +04/17 03:51:19 - mmengine - INFO - Iter(train) [ 9650/160000] lr: 9.4610e-03 eta: 23:04:02 time: 0.5525 data_time: 0.0064 memory: 7635 loss: 0.0279 decode.loss_ce: 0.0165 decode.acc_seg: 99.4065 aux.loss_ce: 0.0114 aux.acc_seg: 98.9489 +04/17 03:51:46 - mmengine - INFO - Iter(train) [ 9700/160000] lr: 9.4582e-03 eta: 23:03:34 time: 0.5518 data_time: 0.0066 memory: 7635 loss: 0.0279 decode.loss_ce: 0.0166 decode.acc_seg: 99.4917 aux.loss_ce: 0.0113 aux.acc_seg: 98.9057 +04/17 03:52:14 - mmengine - INFO - Iter(train) [ 9750/160000] lr: 9.4554e-03 eta: 23:03:06 time: 0.5510 data_time: 0.0060 memory: 7635 loss: 0.0278 decode.loss_ce: 0.0166 decode.acc_seg: 99.3966 aux.loss_ce: 0.0112 aux.acc_seg: 98.7986 +04/17 03:52:41 - mmengine - INFO - Iter(train) [ 9800/160000] lr: 9.4526e-03 eta: 23:02:37 time: 0.5512 data_time: 0.0061 memory: 7635 loss: 0.0260 decode.loss_ce: 0.0153 decode.acc_seg: 99.3086 aux.loss_ce: 0.0107 aux.acc_seg: 98.8056 +04/17 03:53:09 - mmengine - INFO - Iter(train) [ 9850/160000] lr: 9.4498e-03 eta: 23:02:09 time: 0.5513 data_time: 0.0062 memory: 7635 loss: 0.0280 decode.loss_ce: 0.0166 decode.acc_seg: 99.5903 aux.loss_ce: 0.0114 aux.acc_seg: 99.0402 +04/17 03:53:36 - mmengine - INFO - Iter(train) [ 9900/160000] lr: 9.4470e-03 eta: 23:01:41 time: 0.5510 data_time: 0.0054 memory: 7635 loss: 0.0258 decode.loss_ce: 0.0151 decode.acc_seg: 99.4496 aux.loss_ce: 0.0106 aux.acc_seg: 99.0826 +04/17 03:54:04 - mmengine - INFO - Iter(train) [ 9950/160000] lr: 9.4442e-03 eta: 23:01:14 time: 0.5533 data_time: 0.0069 memory: 7635 loss: 0.0241 decode.loss_ce: 0.0145 decode.acc_seg: 99.4751 aux.loss_ce: 0.0096 aux.acc_seg: 99.0772 +04/17 03:54:32 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 03:54:32 - mmengine - INFO - Iter(train) [ 10000/160000] lr: 9.4414e-03 eta: 23:00:46 time: 0.5528 data_time: 0.0062 memory: 7635 loss: 0.0244 decode.loss_ce: 0.0144 decode.acc_seg: 99.4905 aux.loss_ce: 0.0100 aux.acc_seg: 99.2376 +04/17 03:54:32 - mmengine - INFO - Saving checkpoint at 10000 iterations +04/17 03:54:37 - mmengine - INFO - Iter(val) [ 50/355] eta: 0:00:20 time: 0.0465 data_time: 0.0014 memory: 5542 +04/17 03:54:39 - mmengine - INFO - Iter(val) [100/355] eta: 0:00:14 time: 0.0467 data_time: 0.0015 memory: 1657 +04/17 03:54:41 - mmengine - INFO - Iter(val) [150/355] eta: 0:00:10 time: 0.0465 data_time: 0.0014 memory: 1657 +04/17 03:54:44 - mmengine - INFO - Iter(val) [200/355] eta: 0:00:08 time: 0.0468 data_time: 0.0014 memory: 1657 +04/17 03:54:46 - mmengine - INFO - Iter(val) [250/355] eta: 0:00:05 time: 0.0469 data_time: 0.0015 memory: 1657 +04/17 03:54:48 - mmengine - INFO - Iter(val) [300/355] eta: 0:00:02 time: 0.0466 data_time: 0.0015 memory: 1657 +04/17 03:54:51 - mmengine - INFO - Iter(val) [350/355] eta: 0:00:00 time: 0.0454 data_time: 0.0013 memory: 1657 +04/17 03:54:51 - mmengine - INFO - per class results: +04/17 03:54:51 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 99.08 | 99.52 | 99.54 | 99.55 | 99.52 | +| contrast | 80.15 | 89.32 | 88.98 | 88.64 | 89.32 | ++------------+-------+-------+--------+-----------+--------+ +04/17 03:54:51 - mmengine - INFO - Iter(val) [355/355] aAcc: 99.1100 mIoU: 89.6100 mAcc: 94.4200 mFscore: 94.2600 mPrecision: 94.0900 mRecall: 94.4200 data_time: 0.0019 time: 0.0495 +04/17 03:55:19 - mmengine - INFO - Iter(train) [ 10050/160000] lr: 9.4386e-03 eta: 23:00:22 time: 0.5518 data_time: 0.0058 memory: 7635 loss: 0.0284 decode.loss_ce: 0.0174 decode.acc_seg: 98.8690 aux.loss_ce: 0.0111 aux.acc_seg: 98.5484 +04/17 03:55:46 - mmengine - INFO - Iter(train) [ 10100/160000] lr: 9.4358e-03 eta: 22:59:56 time: 0.5513 data_time: 0.0059 memory: 7635 loss: 0.0265 decode.loss_ce: 0.0157 decode.acc_seg: 99.4200 aux.loss_ce: 0.0107 aux.acc_seg: 98.9168 +04/17 03:56:14 - mmengine - INFO - Iter(train) [ 10150/160000] lr: 9.4330e-03 eta: 22:59:27 time: 0.5519 data_time: 0.0069 memory: 7635 loss: 0.0231 decode.loss_ce: 0.0135 decode.acc_seg: 99.4089 aux.loss_ce: 0.0096 aux.acc_seg: 99.0520 +04/17 03:56:42 - mmengine - INFO - Iter(train) [ 10200/160000] lr: 9.4302e-03 eta: 22:58:59 time: 0.5508 data_time: 0.0060 memory: 7635 loss: 0.0247 decode.loss_ce: 0.0144 decode.acc_seg: 99.5257 aux.loss_ce: 0.0104 aux.acc_seg: 99.1284 +04/17 03:57:09 - mmengine - INFO - Iter(train) [ 10250/160000] lr: 9.4274e-03 eta: 22:58:32 time: 0.5504 data_time: 0.0059 memory: 7635 loss: 0.0238 decode.loss_ce: 0.0136 decode.acc_seg: 99.5479 aux.loss_ce: 0.0101 aux.acc_seg: 99.0498 +04/17 03:57:37 - mmengine - INFO - Iter(train) [ 10300/160000] lr: 9.4246e-03 eta: 22:58:04 time: 0.5519 data_time: 0.0062 memory: 7635 loss: 0.0271 decode.loss_ce: 0.0162 decode.acc_seg: 99.2863 aux.loss_ce: 0.0109 aux.acc_seg: 98.8252 +04/17 03:58:05 - mmengine - INFO - Iter(train) [ 10350/160000] lr: 9.4218e-03 eta: 22:57:36 time: 0.5526 data_time: 0.0063 memory: 7635 loss: 0.0246 decode.loss_ce: 0.0142 decode.acc_seg: 99.6452 aux.loss_ce: 0.0104 aux.acc_seg: 99.1855 +04/17 03:58:32 - mmengine - INFO - Iter(train) [ 10400/160000] lr: 9.4190e-03 eta: 22:57:08 time: 0.5511 data_time: 0.0059 memory: 7635 loss: 0.0267 decode.loss_ce: 0.0160 decode.acc_seg: 98.8905 aux.loss_ce: 0.0107 aux.acc_seg: 98.4540 +04/17 03:59:00 - mmengine - INFO - Iter(train) [ 10450/160000] lr: 9.4162e-03 eta: 22:56:40 time: 0.5530 data_time: 0.0076 memory: 7635 loss: 0.0260 decode.loss_ce: 0.0151 decode.acc_seg: 99.4430 aux.loss_ce: 0.0109 aux.acc_seg: 98.9142 +04/17 03:59:27 - mmengine - INFO - Iter(train) [ 10500/160000] lr: 9.4134e-03 eta: 22:56:12 time: 0.5516 data_time: 0.0060 memory: 7635 loss: 0.0256 decode.loss_ce: 0.0151 decode.acc_seg: 99.1611 aux.loss_ce: 0.0105 aux.acc_seg: 98.7815 +04/17 03:59:55 - mmengine - INFO - Iter(train) [ 10550/160000] lr: 9.4106e-03 eta: 22:55:45 time: 0.5521 data_time: 0.0062 memory: 7635 loss: 0.0242 decode.loss_ce: 0.0142 decode.acc_seg: 99.1981 aux.loss_ce: 0.0100 aux.acc_seg: 98.7757 +04/17 04:00:22 - mmengine - INFO - Iter(train) [ 10600/160000] lr: 9.4078e-03 eta: 22:55:17 time: 0.5527 data_time: 0.0060 memory: 7635 loss: 0.0260 decode.loss_ce: 0.0154 decode.acc_seg: 99.2702 aux.loss_ce: 0.0106 aux.acc_seg: 98.8341 +04/17 04:00:50 - mmengine - INFO - Iter(train) [ 10650/160000] lr: 9.4050e-03 eta: 22:54:49 time: 0.5524 data_time: 0.0061 memory: 7635 loss: 0.0251 decode.loss_ce: 0.0149 decode.acc_seg: 99.3166 aux.loss_ce: 0.0102 aux.acc_seg: 98.8374 +04/17 04:01:18 - mmengine - INFO - Iter(train) [ 10700/160000] lr: 9.4022e-03 eta: 22:54:21 time: 0.5526 data_time: 0.0061 memory: 7635 loss: 0.0258 decode.loss_ce: 0.0156 decode.acc_seg: 99.3894 aux.loss_ce: 0.0103 aux.acc_seg: 98.9603 +04/17 04:01:45 - mmengine - INFO - Iter(train) [ 10750/160000] lr: 9.3993e-03 eta: 22:53:54 time: 0.5525 data_time: 0.0064 memory: 7635 loss: 0.0258 decode.loss_ce: 0.0154 decode.acc_seg: 99.3429 aux.loss_ce: 0.0105 aux.acc_seg: 98.8722 +04/17 04:02:13 - mmengine - INFO - Iter(train) [ 10800/160000] lr: 9.3965e-03 eta: 22:53:25 time: 0.5509 data_time: 0.0061 memory: 7635 loss: 0.0266 decode.loss_ce: 0.0158 decode.acc_seg: 99.4295 aux.loss_ce: 0.0107 aux.acc_seg: 98.9844 +04/17 04:02:41 - mmengine - INFO - Iter(train) [ 10850/160000] lr: 9.3937e-03 eta: 22:52:58 time: 0.5520 data_time: 0.0058 memory: 7635 loss: 0.0242 decode.loss_ce: 0.0144 decode.acc_seg: 99.2278 aux.loss_ce: 0.0099 aux.acc_seg: 98.6084 +04/17 04:03:08 - mmengine - INFO - Iter(train) [ 10900/160000] lr: 9.3909e-03 eta: 22:52:30 time: 0.5526 data_time: 0.0063 memory: 7635 loss: 0.0256 decode.loss_ce: 0.0153 decode.acc_seg: 99.3282 aux.loss_ce: 0.0102 aux.acc_seg: 98.9741 +04/17 04:03:36 - mmengine - INFO - Iter(train) [ 10950/160000] lr: 9.3881e-03 eta: 22:52:03 time: 0.5532 data_time: 0.0062 memory: 7635 loss: 0.0275 decode.loss_ce: 0.0163 decode.acc_seg: 99.2203 aux.loss_ce: 0.0113 aux.acc_seg: 98.8931 +04/17 04:04:03 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 04:04:03 - mmengine - INFO - Iter(train) [ 11000/160000] lr: 9.3853e-03 eta: 22:51:35 time: 0.5522 data_time: 0.0059 memory: 7635 loss: 0.0231 decode.loss_ce: 0.0133 decode.acc_seg: 99.4202 aux.loss_ce: 0.0098 aux.acc_seg: 98.9544 +04/17 04:04:31 - mmengine - INFO - Iter(train) [ 11050/160000] lr: 9.3825e-03 eta: 22:51:08 time: 0.5533 data_time: 0.0067 memory: 7635 loss: 0.0237 decode.loss_ce: 0.0138 decode.acc_seg: 99.5981 aux.loss_ce: 0.0099 aux.acc_seg: 99.0320 +04/17 04:04:59 - mmengine - INFO - Iter(train) [ 11100/160000] lr: 9.3797e-03 eta: 22:50:40 time: 0.5518 data_time: 0.0057 memory: 7635 loss: 0.0283 decode.loss_ce: 0.0170 decode.acc_seg: 99.3212 aux.loss_ce: 0.0113 aux.acc_seg: 98.7568 +04/17 04:05:26 - mmengine - INFO - Iter(train) [ 11150/160000] lr: 9.3769e-03 eta: 22:50:13 time: 0.5516 data_time: 0.0061 memory: 7635 loss: 0.0263 decode.loss_ce: 0.0154 decode.acc_seg: 99.5035 aux.loss_ce: 0.0108 aux.acc_seg: 99.0602 +04/17 04:05:54 - mmengine - INFO - Iter(train) [ 11200/160000] lr: 9.3741e-03 eta: 22:49:46 time: 0.5533 data_time: 0.0062 memory: 7635 loss: 0.0271 decode.loss_ce: 0.0162 decode.acc_seg: 99.4478 aux.loss_ce: 0.0109 aux.acc_seg: 98.8791 +04/17 04:06:22 - mmengine - INFO - Iter(train) [ 11250/160000] lr: 9.3713e-03 eta: 22:49:19 time: 0.5510 data_time: 0.0062 memory: 7635 loss: 0.0231 decode.loss_ce: 0.0132 decode.acc_seg: 99.4971 aux.loss_ce: 0.0099 aux.acc_seg: 99.0219 +04/17 04:06:49 - mmengine - INFO - Iter(train) [ 11300/160000] lr: 9.3685e-03 eta: 22:48:53 time: 0.5528 data_time: 0.0060 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0120 decode.acc_seg: 99.4635 aux.loss_ce: 0.0087 aux.acc_seg: 99.0871 +04/17 04:07:17 - mmengine - INFO - Iter(train) [ 11350/160000] lr: 9.3657e-03 eta: 22:48:25 time: 0.5520 data_time: 0.0061 memory: 7635 loss: 0.0241 decode.loss_ce: 0.0139 decode.acc_seg: 99.3774 aux.loss_ce: 0.0102 aux.acc_seg: 98.7114 +04/17 04:07:45 - mmengine - INFO - Iter(train) [ 11400/160000] lr: 9.3629e-03 eta: 22:47:57 time: 0.5510 data_time: 0.0054 memory: 7635 loss: 0.0266 decode.loss_ce: 0.0156 decode.acc_seg: 99.5908 aux.loss_ce: 0.0110 aux.acc_seg: 99.2347 +04/17 04:08:12 - mmengine - INFO - Iter(train) [ 11450/160000] lr: 9.3601e-03 eta: 22:47:30 time: 0.5527 data_time: 0.0062 memory: 7635 loss: 0.0233 decode.loss_ce: 0.0136 decode.acc_seg: 99.3744 aux.loss_ce: 0.0098 aux.acc_seg: 99.0390 +04/17 04:08:40 - mmengine - INFO - Iter(train) [ 11500/160000] lr: 9.3573e-03 eta: 22:47:02 time: 0.5527 data_time: 0.0061 memory: 7635 loss: 0.0227 decode.loss_ce: 0.0130 decode.acc_seg: 99.5013 aux.loss_ce: 0.0096 aux.acc_seg: 99.0967 +04/17 04:09:07 - mmengine - INFO - Iter(train) [ 11550/160000] lr: 9.3545e-03 eta: 22:46:35 time: 0.5534 data_time: 0.0055 memory: 7635 loss: 0.0236 decode.loss_ce: 0.0137 decode.acc_seg: 99.4680 aux.loss_ce: 0.0099 aux.acc_seg: 99.0490 +04/17 04:09:35 - mmengine - INFO - Iter(train) [ 11600/160000] lr: 9.3517e-03 eta: 22:46:07 time: 0.5527 data_time: 0.0062 memory: 7635 loss: 0.0222 decode.loss_ce: 0.0129 decode.acc_seg: 99.4855 aux.loss_ce: 0.0094 aux.acc_seg: 98.9380 +04/17 04:10:03 - mmengine - INFO - Iter(train) [ 11650/160000] lr: 9.3489e-03 eta: 22:45:40 time: 0.5530 data_time: 0.0064 memory: 7635 loss: 0.0225 decode.loss_ce: 0.0132 decode.acc_seg: 99.4992 aux.loss_ce: 0.0093 aux.acc_seg: 99.1189 +04/17 04:10:30 - mmengine - INFO - Iter(train) [ 11700/160000] lr: 9.3461e-03 eta: 22:45:13 time: 0.5544 data_time: 0.0057 memory: 7635 loss: 0.0256 decode.loss_ce: 0.0151 decode.acc_seg: 99.4854 aux.loss_ce: 0.0105 aux.acc_seg: 98.9978 +04/17 04:10:58 - mmengine - INFO - Iter(train) [ 11750/160000] lr: 9.3433e-03 eta: 22:44:46 time: 0.5533 data_time: 0.0058 memory: 7635 loss: 0.0225 decode.loss_ce: 0.0131 decode.acc_seg: 99.5871 aux.loss_ce: 0.0094 aux.acc_seg: 99.2082 +04/17 04:11:26 - mmengine - INFO - Iter(train) [ 11800/160000] lr: 9.3404e-03 eta: 22:44:19 time: 0.5526 data_time: 0.0060 memory: 7635 loss: 0.0250 decode.loss_ce: 0.0147 decode.acc_seg: 99.5094 aux.loss_ce: 0.0103 aux.acc_seg: 99.0664 +04/17 04:11:53 - mmengine - INFO - Iter(train) [ 11850/160000] lr: 9.3376e-03 eta: 22:43:52 time: 0.5531 data_time: 0.0064 memory: 7635 loss: 0.0266 decode.loss_ce: 0.0155 decode.acc_seg: 99.3142 aux.loss_ce: 0.0111 aux.acc_seg: 98.9658 +04/17 04:12:21 - mmengine - INFO - Iter(train) [ 11900/160000] lr: 9.3348e-03 eta: 22:43:25 time: 0.5512 data_time: 0.0060 memory: 7635 loss: 0.0260 decode.loss_ce: 0.0149 decode.acc_seg: 99.0739 aux.loss_ce: 0.0110 aux.acc_seg: 98.6077 +04/17 04:12:49 - mmengine - INFO - Iter(train) [ 11950/160000] lr: 9.3320e-03 eta: 22:42:57 time: 0.5521 data_time: 0.0062 memory: 7635 loss: 0.0237 decode.loss_ce: 0.0138 decode.acc_seg: 99.3942 aux.loss_ce: 0.0100 aux.acc_seg: 98.9405 +04/17 04:13:16 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 04:13:16 - mmengine - INFO - Iter(train) [ 12000/160000] lr: 9.3292e-03 eta: 22:42:31 time: 0.5550 data_time: 0.0068 memory: 7635 loss: 0.0235 decode.loss_ce: 0.0136 decode.acc_seg: 99.3548 aux.loss_ce: 0.0099 aux.acc_seg: 98.9933 +04/17 04:13:44 - mmengine - INFO - Iter(train) [ 12050/160000] lr: 9.3264e-03 eta: 22:42:03 time: 0.5519 data_time: 0.0059 memory: 7635 loss: 0.0238 decode.loss_ce: 0.0138 decode.acc_seg: 99.5077 aux.loss_ce: 0.0099 aux.acc_seg: 99.0158 +04/17 04:14:12 - mmengine - INFO - Iter(train) [ 12100/160000] lr: 9.3236e-03 eta: 22:41:36 time: 0.5543 data_time: 0.0069 memory: 7635 loss: 0.0262 decode.loss_ce: 0.0156 decode.acc_seg: 99.2000 aux.loss_ce: 0.0106 aux.acc_seg: 98.8114 +04/17 04:14:39 - mmengine - INFO - Iter(train) [ 12150/160000] lr: 9.3208e-03 eta: 22:41:09 time: 0.5535 data_time: 0.0055 memory: 7635 loss: 0.0266 decode.loss_ce: 0.0157 decode.acc_seg: 99.3168 aux.loss_ce: 0.0109 aux.acc_seg: 98.7861 +04/17 04:15:07 - mmengine - INFO - Iter(train) [ 12200/160000] lr: 9.3180e-03 eta: 22:40:42 time: 0.5557 data_time: 0.0064 memory: 7635 loss: 0.0275 decode.loss_ce: 0.0165 decode.acc_seg: 99.1772 aux.loss_ce: 0.0111 aux.acc_seg: 98.8053 +04/17 04:15:35 - mmengine - INFO - Iter(train) [ 12250/160000] lr: 9.3152e-03 eta: 22:40:16 time: 0.5527 data_time: 0.0065 memory: 7635 loss: 0.0251 decode.loss_ce: 0.0147 decode.acc_seg: 99.2440 aux.loss_ce: 0.0104 aux.acc_seg: 98.8281 +04/17 04:16:02 - mmengine - INFO - Iter(train) [ 12300/160000] lr: 9.3124e-03 eta: 22:39:49 time: 0.5536 data_time: 0.0062 memory: 7635 loss: 0.0269 decode.loss_ce: 0.0155 decode.acc_seg: 99.4492 aux.loss_ce: 0.0114 aux.acc_seg: 98.9661 +04/17 04:16:30 - mmengine - INFO - Iter(train) [ 12350/160000] lr: 9.3096e-03 eta: 22:39:22 time: 0.5533 data_time: 0.0063 memory: 7635 loss: 0.0266 decode.loss_ce: 0.0160 decode.acc_seg: 99.3697 aux.loss_ce: 0.0106 aux.acc_seg: 98.9136 +04/17 04:16:58 - mmengine - INFO - Iter(train) [ 12400/160000] lr: 9.3068e-03 eta: 22:38:56 time: 0.5531 data_time: 0.0059 memory: 7635 loss: 0.0240 decode.loss_ce: 0.0140 decode.acc_seg: 99.4109 aux.loss_ce: 0.0100 aux.acc_seg: 98.8147 +04/17 04:17:26 - mmengine - INFO - Iter(train) [ 12450/160000] lr: 9.3040e-03 eta: 22:38:29 time: 0.5546 data_time: 0.0068 memory: 7635 loss: 0.0238 decode.loss_ce: 0.0137 decode.acc_seg: 99.3942 aux.loss_ce: 0.0101 aux.acc_seg: 98.8612 +04/17 04:17:53 - mmengine - INFO - Iter(train) [ 12500/160000] lr: 9.3012e-03 eta: 22:38:02 time: 0.5536 data_time: 0.0064 memory: 7635 loss: 0.0238 decode.loss_ce: 0.0138 decode.acc_seg: 99.4101 aux.loss_ce: 0.0100 aux.acc_seg: 98.9651 +04/17 04:18:21 - mmengine - INFO - Iter(train) [ 12550/160000] lr: 9.2984e-03 eta: 22:37:34 time: 0.5534 data_time: 0.0063 memory: 7635 loss: 0.0273 decode.loss_ce: 0.0162 decode.acc_seg: 99.2426 aux.loss_ce: 0.0111 aux.acc_seg: 98.7826 +04/17 04:18:49 - mmengine - INFO - Iter(train) [ 12600/160000] lr: 9.2955e-03 eta: 22:37:08 time: 0.5533 data_time: 0.0061 memory: 7635 loss: 0.0241 decode.loss_ce: 0.0141 decode.acc_seg: 99.5333 aux.loss_ce: 0.0099 aux.acc_seg: 98.8665 +04/17 04:19:16 - mmengine - INFO - Iter(train) [ 12650/160000] lr: 9.2927e-03 eta: 22:36:41 time: 0.5535 data_time: 0.0060 memory: 7635 loss: 0.0245 decode.loss_ce: 0.0145 decode.acc_seg: 99.2301 aux.loss_ce: 0.0100 aux.acc_seg: 98.4837 +04/17 04:19:44 - mmengine - INFO - Iter(train) [ 12700/160000] lr: 9.2899e-03 eta: 22:36:13 time: 0.5533 data_time: 0.0061 memory: 7635 loss: 0.0223 decode.loss_ce: 0.0129 decode.acc_seg: 99.3186 aux.loss_ce: 0.0095 aux.acc_seg: 98.6949 +04/17 04:20:12 - mmengine - INFO - Iter(train) [ 12750/160000] lr: 9.2871e-03 eta: 22:35:46 time: 0.5537 data_time: 0.0074 memory: 7635 loss: 0.0223 decode.loss_ce: 0.0130 decode.acc_seg: 99.4762 aux.loss_ce: 0.0093 aux.acc_seg: 98.9079 +04/17 04:20:39 - mmengine - INFO - Iter(train) [ 12800/160000] lr: 9.2843e-03 eta: 22:35:20 time: 0.5530 data_time: 0.0058 memory: 7635 loss: 0.0225 decode.loss_ce: 0.0129 decode.acc_seg: 99.3520 aux.loss_ce: 0.0095 aux.acc_seg: 98.9733 +04/17 04:21:07 - mmengine - INFO - Iter(train) [ 12850/160000] lr: 9.2815e-03 eta: 22:34:53 time: 0.5536 data_time: 0.0064 memory: 7635 loss: 0.0233 decode.loss_ce: 0.0137 decode.acc_seg: 99.3434 aux.loss_ce: 0.0097 aux.acc_seg: 98.7396 +04/17 04:21:35 - mmengine - INFO - Iter(train) [ 12900/160000] lr: 9.2787e-03 eta: 22:34:26 time: 0.5535 data_time: 0.0068 memory: 7635 loss: 0.0224 decode.loss_ce: 0.0130 decode.acc_seg: 99.5496 aux.loss_ce: 0.0094 aux.acc_seg: 99.1914 +04/17 04:22:02 - mmengine - INFO - Iter(train) [ 12950/160000] lr: 9.2759e-03 eta: 22:33:58 time: 0.5540 data_time: 0.0056 memory: 7635 loss: 0.0242 decode.loss_ce: 0.0139 decode.acc_seg: 99.4299 aux.loss_ce: 0.0103 aux.acc_seg: 98.9958 +04/17 04:22:30 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 04:22:30 - mmengine - INFO - Iter(train) [ 13000/160000] lr: 9.2731e-03 eta: 22:33:31 time: 0.5526 data_time: 0.0057 memory: 7635 loss: 0.0260 decode.loss_ce: 0.0154 decode.acc_seg: 99.3240 aux.loss_ce: 0.0106 aux.acc_seg: 98.8968 +04/17 04:22:58 - mmengine - INFO - Iter(train) [ 13050/160000] lr: 9.2703e-03 eta: 22:33:04 time: 0.5518 data_time: 0.0060 memory: 7635 loss: 0.0254 decode.loss_ce: 0.0151 decode.acc_seg: 99.1981 aux.loss_ce: 0.0102 aux.acc_seg: 98.7233 +04/17 04:23:25 - mmengine - INFO - Iter(train) [ 13100/160000] lr: 9.2675e-03 eta: 22:32:37 time: 0.5540 data_time: 0.0065 memory: 7635 loss: 0.0234 decode.loss_ce: 0.0137 decode.acc_seg: 99.5710 aux.loss_ce: 0.0097 aux.acc_seg: 99.2008 +04/17 04:23:53 - mmengine - INFO - Iter(train) [ 13150/160000] lr: 9.2647e-03 eta: 22:32:10 time: 0.5532 data_time: 0.0056 memory: 7635 loss: 0.0224 decode.loss_ce: 0.0129 decode.acc_seg: 99.4144 aux.loss_ce: 0.0096 aux.acc_seg: 98.8643 +04/17 04:24:21 - mmengine - INFO - Iter(train) [ 13200/160000] lr: 9.2618e-03 eta: 22:31:43 time: 0.5535 data_time: 0.0062 memory: 7635 loss: 0.0233 decode.loss_ce: 0.0134 decode.acc_seg: 99.4789 aux.loss_ce: 0.0099 aux.acc_seg: 98.7773 +04/17 04:24:48 - mmengine - INFO - Iter(train) [ 13250/160000] lr: 9.2590e-03 eta: 22:31:16 time: 0.5537 data_time: 0.0060 memory: 7635 loss: 0.0213 decode.loss_ce: 0.0121 decode.acc_seg: 99.3909 aux.loss_ce: 0.0092 aux.acc_seg: 98.9341 +04/17 04:25:16 - mmengine - INFO - Iter(train) [ 13300/160000] lr: 9.2562e-03 eta: 22:30:49 time: 0.5623 data_time: 0.0060 memory: 7635 loss: 0.0240 decode.loss_ce: 0.0140 decode.acc_seg: 99.4434 aux.loss_ce: 0.0100 aux.acc_seg: 99.0723 +04/17 04:25:44 - mmengine - INFO - Iter(train) [ 13350/160000] lr: 9.2534e-03 eta: 22:30:22 time: 0.5538 data_time: 0.0059 memory: 7635 loss: 0.0234 decode.loss_ce: 0.0136 decode.acc_seg: 99.3601 aux.loss_ce: 0.0098 aux.acc_seg: 98.8899 +04/17 04:26:11 - mmengine - INFO - Iter(train) [ 13400/160000] lr: 9.2506e-03 eta: 22:29:55 time: 0.5526 data_time: 0.0059 memory: 7635 loss: 0.0216 decode.loss_ce: 0.0121 decode.acc_seg: 99.3878 aux.loss_ce: 0.0095 aux.acc_seg: 98.7974 +04/17 04:26:39 - mmengine - INFO - Iter(train) [ 13450/160000] lr: 9.2478e-03 eta: 22:29:29 time: 0.5622 data_time: 0.0067 memory: 7635 loss: 0.0247 decode.loss_ce: 0.0143 decode.acc_seg: 99.2961 aux.loss_ce: 0.0104 aux.acc_seg: 98.8160 +04/17 04:27:07 - mmengine - INFO - Iter(train) [ 13500/160000] lr: 9.2450e-03 eta: 22:29:02 time: 0.5531 data_time: 0.0057 memory: 7635 loss: 0.0239 decode.loss_ce: 0.0138 decode.acc_seg: 99.5115 aux.loss_ce: 0.0101 aux.acc_seg: 99.0416 +04/17 04:27:34 - mmengine - INFO - Iter(train) [ 13550/160000] lr: 9.2422e-03 eta: 22:28:34 time: 0.5529 data_time: 0.0067 memory: 7635 loss: 0.0281 decode.loss_ce: 0.0168 decode.acc_seg: 99.4470 aux.loss_ce: 0.0113 aux.acc_seg: 99.1710 +04/17 04:28:02 - mmengine - INFO - Iter(train) [ 13600/160000] lr: 9.2394e-03 eta: 22:28:07 time: 0.5526 data_time: 0.0067 memory: 7635 loss: 0.0215 decode.loss_ce: 0.0123 decode.acc_seg: 99.4963 aux.loss_ce: 0.0092 aux.acc_seg: 99.0114 +04/17 04:28:30 - mmengine - INFO - Iter(train) [ 13650/160000] lr: 9.2366e-03 eta: 22:27:40 time: 0.5525 data_time: 0.0061 memory: 7635 loss: 0.0200 decode.loss_ce: 0.0113 decode.acc_seg: 99.5402 aux.loss_ce: 0.0087 aux.acc_seg: 99.2253 +04/17 04:28:57 - mmengine - INFO - Iter(train) [ 13700/160000] lr: 9.2338e-03 eta: 22:27:12 time: 0.5516 data_time: 0.0058 memory: 7635 loss: 0.0236 decode.loss_ce: 0.0137 decode.acc_seg: 99.5404 aux.loss_ce: 0.0099 aux.acc_seg: 99.0393 +04/17 04:29:25 - mmengine - INFO - Iter(train) [ 13750/160000] lr: 9.2309e-03 eta: 22:26:45 time: 0.5539 data_time: 0.0062 memory: 7635 loss: 0.0214 decode.loss_ce: 0.0121 decode.acc_seg: 99.5745 aux.loss_ce: 0.0093 aux.acc_seg: 98.9949 +04/17 04:29:53 - mmengine - INFO - Iter(train) [ 13800/160000] lr: 9.2281e-03 eta: 22:26:17 time: 0.5522 data_time: 0.0057 memory: 7635 loss: 0.0228 decode.loss_ce: 0.0132 decode.acc_seg: 99.3312 aux.loss_ce: 0.0095 aux.acc_seg: 98.9589 +04/17 04:30:20 - mmengine - INFO - Iter(train) [ 13850/160000] lr: 9.2253e-03 eta: 22:25:49 time: 0.5518 data_time: 0.0063 memory: 7635 loss: 0.0226 decode.loss_ce: 0.0130 decode.acc_seg: 99.5625 aux.loss_ce: 0.0096 aux.acc_seg: 99.0304 +04/17 04:30:48 - mmengine - INFO - Iter(train) [ 13900/160000] lr: 9.2225e-03 eta: 22:25:21 time: 0.5518 data_time: 0.0057 memory: 7635 loss: 0.0250 decode.loss_ce: 0.0144 decode.acc_seg: 99.4550 aux.loss_ce: 0.0106 aux.acc_seg: 99.1648 +04/17 04:31:15 - mmengine - INFO - Iter(train) [ 13950/160000] lr: 9.2197e-03 eta: 22:24:54 time: 0.5525 data_time: 0.0054 memory: 7635 loss: 0.0210 decode.loss_ce: 0.0119 decode.acc_seg: 99.4488 aux.loss_ce: 0.0091 aux.acc_seg: 98.7908 +04/17 04:31:43 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 04:31:43 - mmengine - INFO - Iter(train) [ 14000/160000] lr: 9.2169e-03 eta: 22:24:26 time: 0.5523 data_time: 0.0056 memory: 7635 loss: 0.0217 decode.loss_ce: 0.0124 decode.acc_seg: 99.5605 aux.loss_ce: 0.0092 aux.acc_seg: 99.1129 +04/17 04:32:11 - mmengine - INFO - Iter(train) [ 14050/160000] lr: 9.2141e-03 eta: 22:23:58 time: 0.5515 data_time: 0.0059 memory: 7635 loss: 0.0229 decode.loss_ce: 0.0131 decode.acc_seg: 99.1985 aux.loss_ce: 0.0098 aux.acc_seg: 98.6768 +04/17 04:32:38 - mmengine - INFO - Iter(train) [ 14100/160000] lr: 9.2113e-03 eta: 22:23:30 time: 0.5543 data_time: 0.0073 memory: 7635 loss: 0.0261 decode.loss_ce: 0.0155 decode.acc_seg: 99.6572 aux.loss_ce: 0.0106 aux.acc_seg: 99.3086 +04/17 04:33:06 - mmengine - INFO - Iter(train) [ 14150/160000] lr: 9.2085e-03 eta: 22:23:03 time: 0.5526 data_time: 0.0064 memory: 7635 loss: 0.0225 decode.loss_ce: 0.0128 decode.acc_seg: 99.3685 aux.loss_ce: 0.0097 aux.acc_seg: 98.7271 +04/17 04:33:34 - mmengine - INFO - Iter(train) [ 14200/160000] lr: 9.2057e-03 eta: 22:22:35 time: 0.5519 data_time: 0.0061 memory: 7635 loss: 0.0242 decode.loss_ce: 0.0141 decode.acc_seg: 99.3918 aux.loss_ce: 0.0101 aux.acc_seg: 98.8184 +04/17 04:34:01 - mmengine - INFO - Iter(train) [ 14250/160000] lr: 9.2028e-03 eta: 22:22:07 time: 0.5522 data_time: 0.0065 memory: 7635 loss: 0.0235 decode.loss_ce: 0.0136 decode.acc_seg: 99.4950 aux.loss_ce: 0.0100 aux.acc_seg: 99.2186 +04/17 04:34:29 - mmengine - INFO - Iter(train) [ 14300/160000] lr: 9.2000e-03 eta: 22:21:40 time: 0.5525 data_time: 0.0067 memory: 7635 loss: 0.0218 decode.loss_ce: 0.0125 decode.acc_seg: 99.5231 aux.loss_ce: 0.0093 aux.acc_seg: 99.0948 +04/17 04:34:56 - mmengine - INFO - Iter(train) [ 14350/160000] lr: 9.1972e-03 eta: 22:21:12 time: 0.5535 data_time: 0.0075 memory: 7635 loss: 0.0260 decode.loss_ce: 0.0150 decode.acc_seg: 99.4243 aux.loss_ce: 0.0109 aux.acc_seg: 98.8042 +04/17 04:35:24 - mmengine - INFO - Iter(train) [ 14400/160000] lr: 9.1944e-03 eta: 22:20:45 time: 0.5533 data_time: 0.0066 memory: 7635 loss: 0.0219 decode.loss_ce: 0.0124 decode.acc_seg: 99.4529 aux.loss_ce: 0.0095 aux.acc_seg: 98.9574 +04/17 04:35:52 - mmengine - INFO - Iter(train) [ 14450/160000] lr: 9.1916e-03 eta: 22:20:18 time: 0.5523 data_time: 0.0061 memory: 7635 loss: 0.0237 decode.loss_ce: 0.0137 decode.acc_seg: 99.3893 aux.loss_ce: 0.0100 aux.acc_seg: 98.9051 +04/17 04:36:19 - mmengine - INFO - Iter(train) [ 14500/160000] lr: 9.1888e-03 eta: 22:19:50 time: 0.5514 data_time: 0.0062 memory: 7635 loss: 0.0235 decode.loss_ce: 0.0135 decode.acc_seg: 99.4831 aux.loss_ce: 0.0100 aux.acc_seg: 99.0737 +04/17 04:36:47 - mmengine - INFO - Iter(train) [ 14550/160000] lr: 9.1860e-03 eta: 22:19:24 time: 0.5534 data_time: 0.0068 memory: 7635 loss: 0.0228 decode.loss_ce: 0.0131 decode.acc_seg: 99.5494 aux.loss_ce: 0.0097 aux.acc_seg: 99.2378 +04/17 04:37:15 - mmengine - INFO - Iter(train) [ 14600/160000] lr: 9.1832e-03 eta: 22:18:56 time: 0.5527 data_time: 0.0057 memory: 7635 loss: 0.0245 decode.loss_ce: 0.0142 decode.acc_seg: 99.4138 aux.loss_ce: 0.0103 aux.acc_seg: 98.9269 +04/17 04:37:42 - mmengine - INFO - Iter(train) [ 14650/160000] lr: 9.1804e-03 eta: 22:18:28 time: 0.5526 data_time: 0.0066 memory: 7635 loss: 0.0226 decode.loss_ce: 0.0132 decode.acc_seg: 99.5334 aux.loss_ce: 0.0094 aux.acc_seg: 99.0604 +04/17 04:38:10 - mmengine - INFO - Iter(train) [ 14700/160000] lr: 9.1776e-03 eta: 22:18:01 time: 0.5539 data_time: 0.0065 memory: 7635 loss: 0.0239 decode.loss_ce: 0.0138 decode.acc_seg: 99.4104 aux.loss_ce: 0.0101 aux.acc_seg: 99.0147 +04/17 04:38:38 - mmengine - INFO - Iter(train) [ 14750/160000] lr: 9.1747e-03 eta: 22:17:33 time: 0.5523 data_time: 0.0061 memory: 7635 loss: 0.0257 decode.loss_ce: 0.0150 decode.acc_seg: 99.2423 aux.loss_ce: 0.0107 aux.acc_seg: 98.8490 +04/17 04:39:05 - mmengine - INFO - Iter(train) [ 14800/160000] lr: 9.1719e-03 eta: 22:17:06 time: 0.5521 data_time: 0.0064 memory: 7635 loss: 0.0226 decode.loss_ce: 0.0129 decode.acc_seg: 99.5416 aux.loss_ce: 0.0097 aux.acc_seg: 99.0267 +04/17 04:39:33 - mmengine - INFO - Iter(train) [ 14850/160000] lr: 9.1691e-03 eta: 22:16:38 time: 0.5527 data_time: 0.0062 memory: 7635 loss: 0.0248 decode.loss_ce: 0.0143 decode.acc_seg: 99.3867 aux.loss_ce: 0.0105 aux.acc_seg: 98.8303 +04/17 04:40:01 - mmengine - INFO - Iter(train) [ 14900/160000] lr: 9.1663e-03 eta: 22:16:11 time: 0.5536 data_time: 0.0064 memory: 7635 loss: 0.0261 decode.loss_ce: 0.0153 decode.acc_seg: 99.5447 aux.loss_ce: 0.0108 aux.acc_seg: 99.1027 +04/17 04:40:28 - mmengine - INFO - Iter(train) [ 14950/160000] lr: 9.1635e-03 eta: 22:15:43 time: 0.5531 data_time: 0.0057 memory: 7635 loss: 0.0226 decode.loss_ce: 0.0127 decode.acc_seg: 99.5181 aux.loss_ce: 0.0099 aux.acc_seg: 99.0941 +04/17 04:40:56 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 04:40:56 - mmengine - INFO - Iter(train) [ 15000/160000] lr: 9.1607e-03 eta: 22:15:16 time: 0.5532 data_time: 0.0055 memory: 7635 loss: 0.0220 decode.loss_ce: 0.0126 decode.acc_seg: 99.4672 aux.loss_ce: 0.0094 aux.acc_seg: 98.8308 +04/17 04:41:24 - mmengine - INFO - Iter(train) [ 15050/160000] lr: 9.1579e-03 eta: 22:14:49 time: 0.5530 data_time: 0.0060 memory: 7635 loss: 0.0232 decode.loss_ce: 0.0132 decode.acc_seg: 99.4109 aux.loss_ce: 0.0100 aux.acc_seg: 98.9364 +04/17 04:41:51 - mmengine - INFO - Iter(train) [ 15100/160000] lr: 9.1551e-03 eta: 22:14:21 time: 0.5527 data_time: 0.0062 memory: 7635 loss: 0.0194 decode.loss_ce: 0.0109 decode.acc_seg: 99.6185 aux.loss_ce: 0.0085 aux.acc_seg: 99.2871 +04/17 04:42:19 - mmengine - INFO - Iter(train) [ 15150/160000] lr: 9.1522e-03 eta: 22:13:54 time: 0.5525 data_time: 0.0058 memory: 7635 loss: 0.0214 decode.loss_ce: 0.0121 decode.acc_seg: 99.6285 aux.loss_ce: 0.0093 aux.acc_seg: 99.2195 +04/17 04:42:46 - mmengine - INFO - Iter(train) [ 15200/160000] lr: 9.1494e-03 eta: 22:13:26 time: 0.5535 data_time: 0.0069 memory: 7635 loss: 0.0212 decode.loss_ce: 0.0121 decode.acc_seg: 99.4781 aux.loss_ce: 0.0091 aux.acc_seg: 99.0991 +04/17 04:43:14 - mmengine - INFO - Iter(train) [ 15250/160000] lr: 9.1466e-03 eta: 22:12:59 time: 0.5535 data_time: 0.0059 memory: 7635 loss: 0.0223 decode.loss_ce: 0.0130 decode.acc_seg: 99.5077 aux.loss_ce: 0.0093 aux.acc_seg: 99.0988 +04/17 04:43:42 - mmengine - INFO - Iter(train) [ 15300/160000] lr: 9.1438e-03 eta: 22:12:31 time: 0.5524 data_time: 0.0065 memory: 7635 loss: 0.0234 decode.loss_ce: 0.0133 decode.acc_seg: 99.3849 aux.loss_ce: 0.0101 aux.acc_seg: 98.8997 +04/17 04:44:09 - mmengine - INFO - Iter(train) [ 15350/160000] lr: 9.1410e-03 eta: 22:12:04 time: 0.5533 data_time: 0.0059 memory: 7635 loss: 0.0218 decode.loss_ce: 0.0124 decode.acc_seg: 99.3773 aux.loss_ce: 0.0094 aux.acc_seg: 99.0790 +04/17 04:44:37 - mmengine - INFO - Iter(train) [ 15400/160000] lr: 9.1382e-03 eta: 22:11:36 time: 0.5523 data_time: 0.0063 memory: 7635 loss: 0.0233 decode.loss_ce: 0.0136 decode.acc_seg: 99.4472 aux.loss_ce: 0.0098 aux.acc_seg: 99.0348 +04/17 04:45:05 - mmengine - INFO - Iter(train) [ 15450/160000] lr: 9.1354e-03 eta: 22:11:09 time: 0.5526 data_time: 0.0058 memory: 7635 loss: 0.0233 decode.loss_ce: 0.0132 decode.acc_seg: 99.3195 aux.loss_ce: 0.0100 aux.acc_seg: 98.7923 +04/17 04:45:32 - mmengine - INFO - Iter(train) [ 15500/160000] lr: 9.1326e-03 eta: 22:10:42 time: 0.5528 data_time: 0.0067 memory: 7635 loss: 0.0217 decode.loss_ce: 0.0124 decode.acc_seg: 99.4955 aux.loss_ce: 0.0093 aux.acc_seg: 99.1011 +04/17 04:46:00 - mmengine - INFO - Iter(train) [ 15550/160000] lr: 9.1297e-03 eta: 22:10:15 time: 0.5527 data_time: 0.0058 memory: 7635 loss: 0.0201 decode.loss_ce: 0.0113 decode.acc_seg: 99.4737 aux.loss_ce: 0.0088 aux.acc_seg: 98.9878 +04/17 04:46:28 - mmengine - INFO - Iter(train) [ 15600/160000] lr: 9.1269e-03 eta: 22:09:48 time: 0.5610 data_time: 0.0068 memory: 7635 loss: 0.0238 decode.loss_ce: 0.0136 decode.acc_seg: 99.3832 aux.loss_ce: 0.0102 aux.acc_seg: 98.8857 +04/17 04:46:55 - mmengine - INFO - Iter(train) [ 15650/160000] lr: 9.1241e-03 eta: 22:09:20 time: 0.5533 data_time: 0.0059 memory: 7635 loss: 0.0200 decode.loss_ce: 0.0111 decode.acc_seg: 99.5057 aux.loss_ce: 0.0089 aux.acc_seg: 99.0979 +04/17 04:47:23 - mmengine - INFO - Iter(train) [ 15700/160000] lr: 9.1213e-03 eta: 22:08:53 time: 0.5538 data_time: 0.0070 memory: 7635 loss: 0.0214 decode.loss_ce: 0.0125 decode.acc_seg: 99.4911 aux.loss_ce: 0.0089 aux.acc_seg: 99.1747 +04/17 04:47:51 - mmengine - INFO - Iter(train) [ 15750/160000] lr: 9.1185e-03 eta: 22:08:26 time: 0.5527 data_time: 0.0061 memory: 7635 loss: 0.0224 decode.loss_ce: 0.0128 decode.acc_seg: 99.4130 aux.loss_ce: 0.0095 aux.acc_seg: 99.0441 +04/17 04:48:18 - mmengine - INFO - Iter(train) [ 15800/160000] lr: 9.1157e-03 eta: 22:07:58 time: 0.5525 data_time: 0.0062 memory: 7635 loss: 0.0214 decode.loss_ce: 0.0121 decode.acc_seg: 99.5008 aux.loss_ce: 0.0093 aux.acc_seg: 99.0557 +04/17 04:48:46 - mmengine - INFO - Iter(train) [ 15850/160000] lr: 9.1129e-03 eta: 22:07:31 time: 0.5522 data_time: 0.0057 memory: 7635 loss: 0.0218 decode.loss_ce: 0.0125 decode.acc_seg: 99.4353 aux.loss_ce: 0.0093 aux.acc_seg: 99.0086 +04/17 04:49:14 - mmengine - INFO - Iter(train) [ 15900/160000] lr: 9.1101e-03 eta: 22:07:04 time: 0.5535 data_time: 0.0064 memory: 7635 loss: 0.0222 decode.loss_ce: 0.0127 decode.acc_seg: 99.5936 aux.loss_ce: 0.0095 aux.acc_seg: 99.1505 +04/17 04:49:41 - mmengine - INFO - Iter(train) [ 15950/160000] lr: 9.1072e-03 eta: 22:06:37 time: 0.5549 data_time: 0.0077 memory: 7635 loss: 0.0204 decode.loss_ce: 0.0117 decode.acc_seg: 99.6064 aux.loss_ce: 0.0087 aux.acc_seg: 99.2549 +04/17 04:50:09 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 04:50:09 - mmengine - INFO - Iter(train) [ 16000/160000] lr: 9.1044e-03 eta: 22:06:10 time: 0.5529 data_time: 0.0060 memory: 7635 loss: 0.0261 decode.loss_ce: 0.0152 decode.acc_seg: 99.3852 aux.loss_ce: 0.0109 aux.acc_seg: 98.5144 +04/17 04:50:37 - mmengine - INFO - Iter(train) [ 16050/160000] lr: 9.1016e-03 eta: 22:05:43 time: 0.5534 data_time: 0.0059 memory: 7635 loss: 0.0229 decode.loss_ce: 0.0133 decode.acc_seg: 99.7216 aux.loss_ce: 0.0096 aux.acc_seg: 99.4108 +04/17 04:51:05 - mmengine - INFO - Iter(train) [ 16100/160000] lr: 9.0988e-03 eta: 22:05:15 time: 0.5529 data_time: 0.0060 memory: 7635 loss: 0.0236 decode.loss_ce: 0.0135 decode.acc_seg: 99.6138 aux.loss_ce: 0.0101 aux.acc_seg: 99.1704 +04/17 04:51:32 - mmengine - INFO - Iter(train) [ 16150/160000] lr: 9.0960e-03 eta: 22:04:48 time: 0.5531 data_time: 0.0062 memory: 7635 loss: 0.0233 decode.loss_ce: 0.0134 decode.acc_seg: 99.1458 aux.loss_ce: 0.0098 aux.acc_seg: 98.5909 +04/17 04:52:00 - mmengine - INFO - Iter(train) [ 16200/160000] lr: 9.0932e-03 eta: 22:04:21 time: 0.5528 data_time: 0.0070 memory: 7635 loss: 0.0229 decode.loss_ce: 0.0133 decode.acc_seg: 99.5446 aux.loss_ce: 0.0095 aux.acc_seg: 99.0372 +04/17 04:52:27 - mmengine - INFO - Iter(train) [ 16250/160000] lr: 9.0904e-03 eta: 22:03:53 time: 0.5528 data_time: 0.0061 memory: 7635 loss: 0.0261 decode.loss_ce: 0.0153 decode.acc_seg: 99.5235 aux.loss_ce: 0.0108 aux.acc_seg: 98.8697 +04/17 04:52:55 - mmengine - INFO - Iter(train) [ 16300/160000] lr: 9.0875e-03 eta: 22:03:26 time: 0.5541 data_time: 0.0056 memory: 7635 loss: 0.0222 decode.loss_ce: 0.0124 decode.acc_seg: 99.3850 aux.loss_ce: 0.0098 aux.acc_seg: 98.9579 +04/17 04:53:23 - mmengine - INFO - Iter(train) [ 16350/160000] lr: 9.0847e-03 eta: 22:02:58 time: 0.5550 data_time: 0.0056 memory: 7635 loss: 0.0235 decode.loss_ce: 0.0138 decode.acc_seg: 99.4367 aux.loss_ce: 0.0097 aux.acc_seg: 98.9577 +04/17 04:53:50 - mmengine - INFO - Iter(train) [ 16400/160000] lr: 9.0819e-03 eta: 22:02:31 time: 0.5536 data_time: 0.0066 memory: 7635 loss: 0.0215 decode.loss_ce: 0.0124 decode.acc_seg: 99.6416 aux.loss_ce: 0.0091 aux.acc_seg: 99.4053 +04/17 04:54:18 - mmengine - INFO - Iter(train) [ 16450/160000] lr: 9.0791e-03 eta: 22:02:04 time: 0.5535 data_time: 0.0065 memory: 7635 loss: 0.0228 decode.loss_ce: 0.0132 decode.acc_seg: 99.5540 aux.loss_ce: 0.0095 aux.acc_seg: 98.9800 +04/17 04:54:46 - mmengine - INFO - Iter(train) [ 16500/160000] lr: 9.0763e-03 eta: 22:01:36 time: 0.5531 data_time: 0.0063 memory: 7635 loss: 0.0261 decode.loss_ce: 0.0155 decode.acc_seg: 99.5005 aux.loss_ce: 0.0106 aux.acc_seg: 99.0659 +04/17 04:55:14 - mmengine - INFO - Iter(train) [ 16550/160000] lr: 9.0735e-03 eta: 22:01:10 time: 0.5529 data_time: 0.0067 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0116 decode.acc_seg: 99.6051 aux.loss_ce: 0.0093 aux.acc_seg: 99.1837 +04/17 04:55:41 - mmengine - INFO - Iter(train) [ 16600/160000] lr: 9.0706e-03 eta: 22:00:42 time: 0.5527 data_time: 0.0065 memory: 7635 loss: 0.0216 decode.loss_ce: 0.0125 decode.acc_seg: 99.5650 aux.loss_ce: 0.0092 aux.acc_seg: 99.2495 +04/17 04:56:09 - mmengine - INFO - Iter(train) [ 16650/160000] lr: 9.0678e-03 eta: 22:00:15 time: 0.5541 data_time: 0.0069 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0118 decode.acc_seg: 99.5330 aux.loss_ce: 0.0091 aux.acc_seg: 99.2336 +04/17 04:56:37 - mmengine - INFO - Iter(train) [ 16700/160000] lr: 9.0650e-03 eta: 21:59:48 time: 0.5529 data_time: 0.0069 memory: 7635 loss: 0.0233 decode.loss_ce: 0.0138 decode.acc_seg: 99.2662 aux.loss_ce: 0.0096 aux.acc_seg: 98.9785 +04/17 04:57:04 - mmengine - INFO - Iter(train) [ 16750/160000] lr: 9.0622e-03 eta: 21:59:21 time: 0.5536 data_time: 0.0069 memory: 7635 loss: 0.0252 decode.loss_ce: 0.0147 decode.acc_seg: 99.5746 aux.loss_ce: 0.0105 aux.acc_seg: 99.1149 +04/17 04:57:32 - mmengine - INFO - Iter(train) [ 16800/160000] lr: 9.0594e-03 eta: 21:58:53 time: 0.5535 data_time: 0.0058 memory: 7635 loss: 0.0226 decode.loss_ce: 0.0133 decode.acc_seg: 99.4496 aux.loss_ce: 0.0093 aux.acc_seg: 99.1562 +04/17 04:58:00 - mmengine - INFO - Iter(train) [ 16850/160000] lr: 9.0566e-03 eta: 21:58:26 time: 0.5536 data_time: 0.0057 memory: 7635 loss: 0.0228 decode.loss_ce: 0.0129 decode.acc_seg: 99.3956 aux.loss_ce: 0.0098 aux.acc_seg: 98.8151 +04/17 04:58:27 - mmengine - INFO - Iter(train) [ 16900/160000] lr: 9.0538e-03 eta: 21:57:58 time: 0.5533 data_time: 0.0060 memory: 7635 loss: 0.0223 decode.loss_ce: 0.0128 decode.acc_seg: 99.6158 aux.loss_ce: 0.0095 aux.acc_seg: 99.3298 +04/17 04:58:55 - mmengine - INFO - Iter(train) [ 16950/160000] lr: 9.0509e-03 eta: 21:57:31 time: 0.5533 data_time: 0.0060 memory: 7635 loss: 0.0249 decode.loss_ce: 0.0146 decode.acc_seg: 99.5616 aux.loss_ce: 0.0103 aux.acc_seg: 99.1063 +04/17 04:59:23 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 04:59:23 - mmengine - INFO - Iter(train) [ 17000/160000] lr: 9.0481e-03 eta: 21:57:04 time: 0.5523 data_time: 0.0061 memory: 7635 loss: 0.0218 decode.loss_ce: 0.0125 decode.acc_seg: 99.3622 aux.loss_ce: 0.0093 aux.acc_seg: 99.0306 +04/17 04:59:50 - mmengine - INFO - Iter(train) [ 17050/160000] lr: 9.0453e-03 eta: 21:56:36 time: 0.5529 data_time: 0.0073 memory: 7635 loss: 0.0211 decode.loss_ce: 0.0118 decode.acc_seg: 99.5401 aux.loss_ce: 0.0093 aux.acc_seg: 99.0092 +04/17 05:00:18 - mmengine - INFO - Iter(train) [ 17100/160000] lr: 9.0425e-03 eta: 21:56:09 time: 0.5518 data_time: 0.0062 memory: 7635 loss: 0.0215 decode.loss_ce: 0.0122 decode.acc_seg: 99.4996 aux.loss_ce: 0.0093 aux.acc_seg: 99.1595 +04/17 05:00:46 - mmengine - INFO - Iter(train) [ 17150/160000] lr: 9.0397e-03 eta: 21:55:41 time: 0.5529 data_time: 0.0058 memory: 7635 loss: 0.0225 decode.loss_ce: 0.0131 decode.acc_seg: 99.5560 aux.loss_ce: 0.0094 aux.acc_seg: 99.0885 +04/17 05:01:13 - mmengine - INFO - Iter(train) [ 17200/160000] lr: 9.0369e-03 eta: 21:55:14 time: 0.5536 data_time: 0.0063 memory: 7635 loss: 0.0215 decode.loss_ce: 0.0122 decode.acc_seg: 99.5878 aux.loss_ce: 0.0092 aux.acc_seg: 99.2035 +04/17 05:01:41 - mmengine - INFO - Iter(train) [ 17250/160000] lr: 9.0340e-03 eta: 21:54:46 time: 0.5522 data_time: 0.0064 memory: 7635 loss: 0.0203 decode.loss_ce: 0.0114 decode.acc_seg: 99.6227 aux.loss_ce: 0.0089 aux.acc_seg: 99.1842 +04/17 05:02:08 - mmengine - INFO - Iter(train) [ 17300/160000] lr: 9.0312e-03 eta: 21:54:19 time: 0.5531 data_time: 0.0064 memory: 7635 loss: 0.0214 decode.loss_ce: 0.0121 decode.acc_seg: 99.5550 aux.loss_ce: 0.0093 aux.acc_seg: 99.2855 +04/17 05:02:36 - mmengine - INFO - Iter(train) [ 17350/160000] lr: 9.0284e-03 eta: 21:53:51 time: 0.5523 data_time: 0.0066 memory: 7635 loss: 0.0230 decode.loss_ce: 0.0132 decode.acc_seg: 99.5240 aux.loss_ce: 0.0098 aux.acc_seg: 98.9947 +04/17 05:03:04 - mmengine - INFO - Iter(train) [ 17400/160000] lr: 9.0256e-03 eta: 21:53:23 time: 0.5515 data_time: 0.0066 memory: 7635 loss: 0.0246 decode.loss_ce: 0.0143 decode.acc_seg: 99.4363 aux.loss_ce: 0.0103 aux.acc_seg: 99.0086 +04/17 05:03:31 - mmengine - INFO - Iter(train) [ 17450/160000] lr: 9.0228e-03 eta: 21:52:56 time: 0.5523 data_time: 0.0055 memory: 7635 loss: 0.0239 decode.loss_ce: 0.0139 decode.acc_seg: 99.1714 aux.loss_ce: 0.0101 aux.acc_seg: 98.6675 +04/17 05:03:59 - mmengine - INFO - Iter(train) [ 17500/160000] lr: 9.0200e-03 eta: 21:52:28 time: 0.5528 data_time: 0.0071 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0116 decode.acc_seg: 99.6930 aux.loss_ce: 0.0091 aux.acc_seg: 99.3424 +04/17 05:04:27 - mmengine - INFO - Iter(train) [ 17550/160000] lr: 9.0171e-03 eta: 21:52:01 time: 0.5524 data_time: 0.0062 memory: 7635 loss: 0.0199 decode.loss_ce: 0.0110 decode.acc_seg: 99.5161 aux.loss_ce: 0.0090 aux.acc_seg: 98.9089 +04/17 05:04:54 - mmengine - INFO - Iter(train) [ 17600/160000] lr: 9.0143e-03 eta: 21:51:33 time: 0.5539 data_time: 0.0071 memory: 7635 loss: 0.0223 decode.loss_ce: 0.0125 decode.acc_seg: 99.5656 aux.loss_ce: 0.0098 aux.acc_seg: 98.9678 +04/17 05:05:22 - mmengine - INFO - Iter(train) [ 17650/160000] lr: 9.0115e-03 eta: 21:51:06 time: 0.5536 data_time: 0.0064 memory: 7635 loss: 0.0217 decode.loss_ce: 0.0121 decode.acc_seg: 99.5657 aux.loss_ce: 0.0095 aux.acc_seg: 99.1462 +04/17 05:05:50 - mmengine - INFO - Iter(train) [ 17700/160000] lr: 9.0087e-03 eta: 21:50:39 time: 0.5512 data_time: 0.0066 memory: 7635 loss: 0.0224 decode.loss_ce: 0.0129 decode.acc_seg: 99.3938 aux.loss_ce: 0.0095 aux.acc_seg: 98.9109 +04/17 05:06:17 - mmengine - INFO - Iter(train) [ 17750/160000] lr: 9.0059e-03 eta: 21:50:12 time: 0.5525 data_time: 0.0058 memory: 7635 loss: 0.0198 decode.loss_ce: 0.0112 decode.acc_seg: 99.6460 aux.loss_ce: 0.0086 aux.acc_seg: 99.1364 +04/17 05:06:45 - mmengine - INFO - Iter(train) [ 17800/160000] lr: 9.0031e-03 eta: 21:49:44 time: 0.5530 data_time: 0.0061 memory: 7635 loss: 0.0219 decode.loss_ce: 0.0125 decode.acc_seg: 99.6550 aux.loss_ce: 0.0094 aux.acc_seg: 99.1895 +04/17 05:07:13 - mmengine - INFO - Iter(train) [ 17850/160000] lr: 9.0002e-03 eta: 21:49:17 time: 0.5531 data_time: 0.0065 memory: 7635 loss: 0.0233 decode.loss_ce: 0.0136 decode.acc_seg: 99.5639 aux.loss_ce: 0.0097 aux.acc_seg: 99.1932 +04/17 05:07:40 - mmengine - INFO - Iter(train) [ 17900/160000] lr: 8.9974e-03 eta: 21:48:49 time: 0.5521 data_time: 0.0060 memory: 7635 loss: 0.0227 decode.loss_ce: 0.0129 decode.acc_seg: 99.4536 aux.loss_ce: 0.0098 aux.acc_seg: 99.0761 +04/17 05:08:08 - mmengine - INFO - Iter(train) [ 17950/160000] lr: 8.9946e-03 eta: 21:48:22 time: 0.5533 data_time: 0.0058 memory: 7635 loss: 0.0198 decode.loss_ce: 0.0111 decode.acc_seg: 99.6744 aux.loss_ce: 0.0087 aux.acc_seg: 99.3177 +04/17 05:08:36 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 05:08:36 - mmengine - INFO - Iter(train) [ 18000/160000] lr: 8.9918e-03 eta: 21:47:54 time: 0.5542 data_time: 0.0061 memory: 7635 loss: 0.0230 decode.loss_ce: 0.0135 decode.acc_seg: 99.3295 aux.loss_ce: 0.0095 aux.acc_seg: 98.8405 +04/17 05:09:03 - mmengine - INFO - Iter(train) [ 18050/160000] lr: 8.9890e-03 eta: 21:47:27 time: 0.5532 data_time: 0.0067 memory: 7635 loss: 0.0223 decode.loss_ce: 0.0126 decode.acc_seg: 99.3703 aux.loss_ce: 0.0097 aux.acc_seg: 98.7511 +04/17 05:09:31 - mmengine - INFO - Iter(train) [ 18100/160000] lr: 8.9862e-03 eta: 21:46:59 time: 0.5531 data_time: 0.0062 memory: 7635 loss: 0.0210 decode.loss_ce: 0.0118 decode.acc_seg: 99.4115 aux.loss_ce: 0.0092 aux.acc_seg: 99.0157 +04/17 05:09:59 - mmengine - INFO - Iter(train) [ 18150/160000] lr: 8.9833e-03 eta: 21:46:32 time: 0.5542 data_time: 0.0057 memory: 7635 loss: 0.0219 decode.loss_ce: 0.0123 decode.acc_seg: 99.4532 aux.loss_ce: 0.0096 aux.acc_seg: 99.0580 +04/17 05:10:26 - mmengine - INFO - Iter(train) [ 18200/160000] lr: 8.9805e-03 eta: 21:46:05 time: 0.5543 data_time: 0.0063 memory: 7635 loss: 0.0250 decode.loss_ce: 0.0146 decode.acc_seg: 99.4089 aux.loss_ce: 0.0104 aux.acc_seg: 99.1213 +04/17 05:10:54 - mmengine - INFO - Iter(train) [ 18250/160000] lr: 8.9777e-03 eta: 21:45:38 time: 0.5533 data_time: 0.0066 memory: 7635 loss: 0.0229 decode.loss_ce: 0.0134 decode.acc_seg: 99.4775 aux.loss_ce: 0.0096 aux.acc_seg: 98.9356 +04/17 05:11:22 - mmengine - INFO - Iter(train) [ 18300/160000] lr: 8.9749e-03 eta: 21:45:10 time: 0.5528 data_time: 0.0062 memory: 7635 loss: 0.0214 decode.loss_ce: 0.0119 decode.acc_seg: 99.4629 aux.loss_ce: 0.0094 aux.acc_seg: 98.8179 +04/17 05:11:49 - mmengine - INFO - Iter(train) [ 18350/160000] lr: 8.9721e-03 eta: 21:44:43 time: 0.5520 data_time: 0.0060 memory: 7635 loss: 0.0251 decode.loss_ce: 0.0148 decode.acc_seg: 99.3949 aux.loss_ce: 0.0103 aux.acc_seg: 98.8255 +04/17 05:12:17 - mmengine - INFO - Iter(train) [ 18400/160000] lr: 8.9692e-03 eta: 21:44:15 time: 0.5529 data_time: 0.0064 memory: 7635 loss: 0.0241 decode.loss_ce: 0.0139 decode.acc_seg: 99.2954 aux.loss_ce: 0.0102 aux.acc_seg: 99.0137 +04/17 05:12:45 - mmengine - INFO - Iter(train) [ 18450/160000] lr: 8.9664e-03 eta: 21:43:48 time: 0.5535 data_time: 0.0061 memory: 7635 loss: 0.0230 decode.loss_ce: 0.0133 decode.acc_seg: 99.6344 aux.loss_ce: 0.0097 aux.acc_seg: 99.0255 +04/17 05:13:12 - mmengine - INFO - Iter(train) [ 18500/160000] lr: 8.9636e-03 eta: 21:43:20 time: 0.5546 data_time: 0.0063 memory: 7635 loss: 0.0223 decode.loss_ce: 0.0130 decode.acc_seg: 99.5232 aux.loss_ce: 0.0094 aux.acc_seg: 99.1300 +04/17 05:13:40 - mmengine - INFO - Iter(train) [ 18550/160000] lr: 8.9608e-03 eta: 21:42:53 time: 0.5543 data_time: 0.0063 memory: 7635 loss: 0.0221 decode.loss_ce: 0.0126 decode.acc_seg: 99.5961 aux.loss_ce: 0.0095 aux.acc_seg: 99.1305 +04/17 05:14:08 - mmengine - INFO - Iter(train) [ 18600/160000] lr: 8.9580e-03 eta: 21:42:25 time: 0.5524 data_time: 0.0066 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0117 decode.acc_seg: 99.5456 aux.loss_ce: 0.0091 aux.acc_seg: 99.1299 +04/17 05:14:35 - mmengine - INFO - Iter(train) [ 18650/160000] lr: 8.9551e-03 eta: 21:41:58 time: 0.5537 data_time: 0.0069 memory: 7635 loss: 0.0223 decode.loss_ce: 0.0126 decode.acc_seg: 99.0963 aux.loss_ce: 0.0097 aux.acc_seg: 98.5669 +04/17 05:15:03 - mmengine - INFO - Iter(train) [ 18700/160000] lr: 8.9523e-03 eta: 21:41:31 time: 0.5532 data_time: 0.0060 memory: 7635 loss: 0.0226 decode.loss_ce: 0.0130 decode.acc_seg: 99.2403 aux.loss_ce: 0.0096 aux.acc_seg: 98.9312 +04/17 05:15:31 - mmengine - INFO - Iter(train) [ 18750/160000] lr: 8.9495e-03 eta: 21:41:04 time: 0.5528 data_time: 0.0056 memory: 7635 loss: 0.0199 decode.loss_ce: 0.0114 decode.acc_seg: 99.1571 aux.loss_ce: 0.0085 aux.acc_seg: 98.7487 +04/17 05:15:58 - mmengine - INFO - Iter(train) [ 18800/160000] lr: 8.9467e-03 eta: 21:40:37 time: 0.5532 data_time: 0.0059 memory: 7635 loss: 0.0215 decode.loss_ce: 0.0122 decode.acc_seg: 99.4122 aux.loss_ce: 0.0093 aux.acc_seg: 99.0150 +04/17 05:16:26 - mmengine - INFO - Iter(train) [ 18850/160000] lr: 8.9439e-03 eta: 21:40:10 time: 0.5552 data_time: 0.0060 memory: 7635 loss: 0.0227 decode.loss_ce: 0.0133 decode.acc_seg: 99.2135 aux.loss_ce: 0.0095 aux.acc_seg: 98.6745 +04/17 05:16:54 - mmengine - INFO - Iter(train) [ 18900/160000] lr: 8.9411e-03 eta: 21:39:42 time: 0.5523 data_time: 0.0063 memory: 7635 loss: 0.0220 decode.loss_ce: 0.0126 decode.acc_seg: 99.4956 aux.loss_ce: 0.0094 aux.acc_seg: 99.0646 +04/17 05:17:21 - mmengine - INFO - Iter(train) [ 18950/160000] lr: 8.9382e-03 eta: 21:39:15 time: 0.5538 data_time: 0.0062 memory: 7635 loss: 0.0202 decode.loss_ce: 0.0113 decode.acc_seg: 99.3465 aux.loss_ce: 0.0090 aux.acc_seg: 98.9604 +04/17 05:17:49 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 05:17:49 - mmengine - INFO - Iter(train) [ 19000/160000] lr: 8.9354e-03 eta: 21:38:47 time: 0.5536 data_time: 0.0056 memory: 7635 loss: 0.0213 decode.loss_ce: 0.0120 decode.acc_seg: 99.5575 aux.loss_ce: 0.0093 aux.acc_seg: 99.0992 +04/17 05:18:17 - mmengine - INFO - Iter(train) [ 19050/160000] lr: 8.9326e-03 eta: 21:38:20 time: 0.5552 data_time: 0.0063 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0119 decode.acc_seg: 99.3151 aux.loss_ce: 0.0090 aux.acc_seg: 98.9167 +04/17 05:18:44 - mmengine - INFO - Iter(train) [ 19100/160000] lr: 8.9298e-03 eta: 21:37:53 time: 0.5525 data_time: 0.0066 memory: 7635 loss: 0.0223 decode.loss_ce: 0.0128 decode.acc_seg: 99.2525 aux.loss_ce: 0.0095 aux.acc_seg: 98.7122 +04/17 05:19:12 - mmengine - INFO - Iter(train) [ 19150/160000] lr: 8.9270e-03 eta: 21:37:26 time: 0.5525 data_time: 0.0061 memory: 7635 loss: 0.0211 decode.loss_ce: 0.0119 decode.acc_seg: 99.6479 aux.loss_ce: 0.0092 aux.acc_seg: 99.2475 +04/17 05:19:40 - mmengine - INFO - Iter(train) [ 19200/160000] lr: 8.9241e-03 eta: 21:36:58 time: 0.5531 data_time: 0.0063 memory: 7635 loss: 0.0220 decode.loss_ce: 0.0125 decode.acc_seg: 99.4622 aux.loss_ce: 0.0095 aux.acc_seg: 98.9523 +04/17 05:20:07 - mmengine - INFO - Iter(train) [ 19250/160000] lr: 8.9213e-03 eta: 21:36:31 time: 0.5537 data_time: 0.0054 memory: 7635 loss: 0.0204 decode.loss_ce: 0.0117 decode.acc_seg: 99.7088 aux.loss_ce: 0.0086 aux.acc_seg: 99.4136 +04/17 05:20:35 - mmengine - INFO - Iter(train) [ 19300/160000] lr: 8.9185e-03 eta: 21:36:03 time: 0.5532 data_time: 0.0060 memory: 7635 loss: 0.0212 decode.loss_ce: 0.0118 decode.acc_seg: 99.5994 aux.loss_ce: 0.0093 aux.acc_seg: 99.1597 +04/17 05:21:03 - mmengine - INFO - Iter(train) [ 19350/160000] lr: 8.9157e-03 eta: 21:35:36 time: 0.5546 data_time: 0.0063 memory: 7635 loss: 0.0211 decode.loss_ce: 0.0118 decode.acc_seg: 99.5967 aux.loss_ce: 0.0093 aux.acc_seg: 98.9788 +04/17 05:21:30 - mmengine - INFO - Iter(train) [ 19400/160000] lr: 8.9129e-03 eta: 21:35:09 time: 0.5541 data_time: 0.0061 memory: 7635 loss: 0.0208 decode.loss_ce: 0.0117 decode.acc_seg: 99.4917 aux.loss_ce: 0.0090 aux.acc_seg: 98.9080 +04/17 05:21:58 - mmengine - INFO - Iter(train) [ 19450/160000] lr: 8.9100e-03 eta: 21:34:41 time: 0.5529 data_time: 0.0060 memory: 7635 loss: 0.0205 decode.loss_ce: 0.0115 decode.acc_seg: 99.5297 aux.loss_ce: 0.0090 aux.acc_seg: 98.9814 +04/17 05:22:26 - mmengine - INFO - Iter(train) [ 19500/160000] lr: 8.9072e-03 eta: 21:34:14 time: 0.5541 data_time: 0.0063 memory: 7635 loss: 0.0223 decode.loss_ce: 0.0124 decode.acc_seg: 99.0890 aux.loss_ce: 0.0098 aux.acc_seg: 98.7421 +04/17 05:22:54 - mmengine - INFO - Iter(train) [ 19550/160000] lr: 8.9044e-03 eta: 21:33:47 time: 0.5536 data_time: 0.0064 memory: 7635 loss: 0.0217 decode.loss_ce: 0.0127 decode.acc_seg: 99.6622 aux.loss_ce: 0.0090 aux.acc_seg: 99.3778 +04/17 05:23:21 - mmengine - INFO - Iter(train) [ 19600/160000] lr: 8.9016e-03 eta: 21:33:19 time: 0.5531 data_time: 0.0057 memory: 7635 loss: 0.0196 decode.loss_ce: 0.0110 decode.acc_seg: 99.5071 aux.loss_ce: 0.0087 aux.acc_seg: 98.8554 +04/17 05:23:49 - mmengine - INFO - Iter(train) [ 19650/160000] lr: 8.8987e-03 eta: 21:32:52 time: 0.5537 data_time: 0.0057 memory: 7635 loss: 0.0237 decode.loss_ce: 0.0137 decode.acc_seg: 99.3431 aux.loss_ce: 0.0101 aux.acc_seg: 98.9584 +04/17 05:24:17 - mmengine - INFO - Iter(train) [ 19700/160000] lr: 8.8959e-03 eta: 21:32:24 time: 0.5539 data_time: 0.0059 memory: 7635 loss: 0.0211 decode.loss_ce: 0.0116 decode.acc_seg: 99.5342 aux.loss_ce: 0.0095 aux.acc_seg: 98.7595 +04/17 05:24:44 - mmengine - INFO - Iter(train) [ 19750/160000] lr: 8.8931e-03 eta: 21:31:57 time: 0.5522 data_time: 0.0066 memory: 7635 loss: 0.0230 decode.loss_ce: 0.0130 decode.acc_seg: 99.5636 aux.loss_ce: 0.0100 aux.acc_seg: 99.1068 +04/17 05:25:12 - mmengine - INFO - Iter(train) [ 19800/160000] lr: 8.8903e-03 eta: 21:31:30 time: 0.5531 data_time: 0.0059 memory: 7635 loss: 0.0210 decode.loss_ce: 0.0120 decode.acc_seg: 99.3897 aux.loss_ce: 0.0090 aux.acc_seg: 98.8853 +04/17 05:25:40 - mmengine - INFO - Iter(train) [ 19850/160000] lr: 8.8875e-03 eta: 21:31:03 time: 0.5542 data_time: 0.0061 memory: 7635 loss: 0.0228 decode.loss_ce: 0.0129 decode.acc_seg: 99.5498 aux.loss_ce: 0.0099 aux.acc_seg: 99.0361 +04/17 05:26:07 - mmengine - INFO - Iter(train) [ 19900/160000] lr: 8.8846e-03 eta: 21:30:36 time: 0.5614 data_time: 0.0059 memory: 7635 loss: 0.0224 decode.loss_ce: 0.0127 decode.acc_seg: 99.6275 aux.loss_ce: 0.0097 aux.acc_seg: 99.1510 +04/17 05:26:35 - mmengine - INFO - Iter(train) [ 19950/160000] lr: 8.8818e-03 eta: 21:30:08 time: 0.5525 data_time: 0.0062 memory: 7635 loss: 0.0208 decode.loss_ce: 0.0118 decode.acc_seg: 99.4675 aux.loss_ce: 0.0090 aux.acc_seg: 98.8947 +04/17 05:27:03 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 05:27:03 - mmengine - INFO - Iter(train) [ 20000/160000] lr: 8.8790e-03 eta: 21:29:41 time: 0.5544 data_time: 0.0064 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0106 decode.acc_seg: 99.4899 aux.loss_ce: 0.0086 aux.acc_seg: 98.9120 +04/17 05:27:03 - mmengine - INFO - Saving checkpoint at 20000 iterations +04/17 05:27:07 - mmengine - INFO - Iter(val) [ 50/355] eta: 0:00:14 time: 0.0475 data_time: 0.0015 memory: 1657 +04/17 05:27:09 - mmengine - INFO - Iter(val) [100/355] eta: 0:00:12 time: 0.0466 data_time: 0.0015 memory: 1657 +04/17 05:27:11 - mmengine - INFO - Iter(val) [150/355] eta: 0:00:09 time: 0.0463 data_time: 0.0013 memory: 1657 +04/17 05:27:14 - mmengine - INFO - Iter(val) [200/355] eta: 0:00:07 time: 0.0462 data_time: 0.0014 memory: 1657 +04/17 05:27:16 - mmengine - INFO - Iter(val) [250/355] eta: 0:00:04 time: 0.0473 data_time: 0.0013 memory: 1657 +04/17 05:27:18 - mmengine - INFO - Iter(val) [300/355] eta: 0:00:02 time: 0.0478 data_time: 0.0014 memory: 1657 +04/17 05:27:21 - mmengine - INFO - Iter(val) [350/355] eta: 0:00:00 time: 0.0462 data_time: 0.0015 memory: 1657 +04/17 05:27:21 - mmengine - INFO - per class results: +04/17 05:27:21 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 99.12 | 99.64 | 99.56 | 99.48 | 99.64 | +| contrast | 80.59 | 87.49 | 89.25 | 91.08 | 87.49 | ++------------+-------+-------+--------+-----------+--------+ +04/17 05:27:21 - mmengine - INFO - Iter(val) [355/355] aAcc: 99.1500 mIoU: 89.8500 mAcc: 93.5700 mFscore: 94.4000 mPrecision: 95.2800 mRecall: 93.5700 data_time: 0.0015 time: 0.0467 +04/17 05:27:49 - mmengine - INFO - Iter(train) [ 20050/160000] lr: 8.8762e-03 eta: 21:29:17 time: 0.5527 data_time: 0.0057 memory: 7635 loss: 0.0255 decode.loss_ce: 0.0148 decode.acc_seg: 99.4461 aux.loss_ce: 0.0107 aux.acc_seg: 98.9660 +04/17 05:28:17 - mmengine - INFO - Iter(train) [ 20100/160000] lr: 8.8734e-03 eta: 21:28:49 time: 0.5534 data_time: 0.0060 memory: 7635 loss: 0.0222 decode.loss_ce: 0.0126 decode.acc_seg: 99.5745 aux.loss_ce: 0.0096 aux.acc_seg: 98.9091 +04/17 05:28:44 - mmengine - INFO - Iter(train) [ 20150/160000] lr: 8.8705e-03 eta: 21:28:22 time: 0.5543 data_time: 0.0061 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0109 decode.acc_seg: 99.4581 aux.loss_ce: 0.0086 aux.acc_seg: 99.0719 +04/17 05:29:12 - mmengine - INFO - Iter(train) [ 20200/160000] lr: 8.8677e-03 eta: 21:27:54 time: 0.5538 data_time: 0.0075 memory: 7635 loss: 0.0222 decode.loss_ce: 0.0126 decode.acc_seg: 99.5244 aux.loss_ce: 0.0096 aux.acc_seg: 98.9345 +04/17 05:29:40 - mmengine - INFO - Iter(train) [ 20250/160000] lr: 8.8649e-03 eta: 21:27:27 time: 0.5527 data_time: 0.0057 memory: 7635 loss: 0.0215 decode.loss_ce: 0.0123 decode.acc_seg: 99.5229 aux.loss_ce: 0.0092 aux.acc_seg: 99.0468 +04/17 05:30:07 - mmengine - INFO - Iter(train) [ 20300/160000] lr: 8.8621e-03 eta: 21:27:00 time: 0.5527 data_time: 0.0076 memory: 7635 loss: 0.0218 decode.loss_ce: 0.0121 decode.acc_seg: 99.6001 aux.loss_ce: 0.0097 aux.acc_seg: 99.0586 +04/17 05:30:35 - mmengine - INFO - Iter(train) [ 20350/160000] lr: 8.8592e-03 eta: 21:26:32 time: 0.5543 data_time: 0.0057 memory: 7635 loss: 0.0227 decode.loss_ce: 0.0129 decode.acc_seg: 99.4566 aux.loss_ce: 0.0097 aux.acc_seg: 98.9085 +04/17 05:31:03 - mmengine - INFO - Iter(train) [ 20400/160000] lr: 8.8564e-03 eta: 21:26:05 time: 0.5544 data_time: 0.0066 memory: 7635 loss: 0.0203 decode.loss_ce: 0.0113 decode.acc_seg: 99.6126 aux.loss_ce: 0.0090 aux.acc_seg: 99.3439 +04/17 05:31:30 - mmengine - INFO - Iter(train) [ 20450/160000] lr: 8.8536e-03 eta: 21:25:38 time: 0.5535 data_time: 0.0066 memory: 7635 loss: 0.0260 decode.loss_ce: 0.0152 decode.acc_seg: 99.1918 aux.loss_ce: 0.0108 aux.acc_seg: 98.7544 +04/17 05:31:58 - mmengine - INFO - Iter(train) [ 20500/160000] lr: 8.8508e-03 eta: 21:25:10 time: 0.5527 data_time: 0.0064 memory: 7635 loss: 0.0215 decode.loss_ce: 0.0123 decode.acc_seg: 99.6285 aux.loss_ce: 0.0092 aux.acc_seg: 99.0986 +04/17 05:32:26 - mmengine - INFO - Iter(train) [ 20550/160000] lr: 8.8480e-03 eta: 21:24:43 time: 0.5534 data_time: 0.0073 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0117 decode.acc_seg: 99.6758 aux.loss_ce: 0.0089 aux.acc_seg: 99.2632 +04/17 05:32:54 - mmengine - INFO - Iter(train) [ 20600/160000] lr: 8.8451e-03 eta: 21:24:16 time: 0.5535 data_time: 0.0067 memory: 7635 loss: 0.0214 decode.loss_ce: 0.0124 decode.acc_seg: 99.2835 aux.loss_ce: 0.0090 aux.acc_seg: 98.8653 +04/17 05:33:21 - mmengine - INFO - Iter(train) [ 20650/160000] lr: 8.8423e-03 eta: 21:23:48 time: 0.5521 data_time: 0.0063 memory: 7635 loss: 0.0203 decode.loss_ce: 0.0113 decode.acc_seg: 99.5019 aux.loss_ce: 0.0089 aux.acc_seg: 99.1848 +04/17 05:33:49 - mmengine - INFO - Iter(train) [ 20700/160000] lr: 8.8395e-03 eta: 21:23:21 time: 0.5525 data_time: 0.0060 memory: 7635 loss: 0.0200 decode.loss_ce: 0.0110 decode.acc_seg: 99.5225 aux.loss_ce: 0.0090 aux.acc_seg: 98.9757 +04/17 05:34:17 - mmengine - INFO - Iter(train) [ 20750/160000] lr: 8.8367e-03 eta: 21:22:54 time: 0.5647 data_time: 0.0066 memory: 7635 loss: 0.0216 decode.loss_ce: 0.0122 decode.acc_seg: 99.4334 aux.loss_ce: 0.0093 aux.acc_seg: 98.9375 +04/17 05:34:44 - mmengine - INFO - Iter(train) [ 20800/160000] lr: 8.8338e-03 eta: 21:22:27 time: 0.5537 data_time: 0.0068 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0120 decode.acc_seg: 99.4299 aux.loss_ce: 0.0090 aux.acc_seg: 98.9567 +04/17 05:35:12 - mmengine - INFO - Iter(train) [ 20850/160000] lr: 8.8310e-03 eta: 21:22:00 time: 0.5523 data_time: 0.0063 memory: 7635 loss: 0.0213 decode.loss_ce: 0.0122 decode.acc_seg: 99.6354 aux.loss_ce: 0.0091 aux.acc_seg: 99.2721 +04/17 05:35:40 - mmengine - INFO - Iter(train) [ 20900/160000] lr: 8.8282e-03 eta: 21:21:32 time: 0.5539 data_time: 0.0062 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0119 decode.acc_seg: 99.5650 aux.loss_ce: 0.0091 aux.acc_seg: 99.1774 +04/17 05:36:07 - mmengine - INFO - Iter(train) [ 20950/160000] lr: 8.8254e-03 eta: 21:21:05 time: 0.5537 data_time: 0.0059 memory: 7635 loss: 0.0196 decode.loss_ce: 0.0109 decode.acc_seg: 99.6063 aux.loss_ce: 0.0087 aux.acc_seg: 99.1423 +04/17 05:36:35 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 05:36:35 - mmengine - INFO - Iter(train) [ 21000/160000] lr: 8.8225e-03 eta: 21:20:38 time: 0.5542 data_time: 0.0067 memory: 7635 loss: 0.0216 decode.loss_ce: 0.0122 decode.acc_seg: 99.5399 aux.loss_ce: 0.0093 aux.acc_seg: 99.0988 +04/17 05:37:03 - mmengine - INFO - Iter(train) [ 21050/160000] lr: 8.8197e-03 eta: 21:20:11 time: 0.5540 data_time: 0.0061 memory: 7635 loss: 0.0228 decode.loss_ce: 0.0126 decode.acc_seg: 99.6324 aux.loss_ce: 0.0102 aux.acc_seg: 99.1165 +04/17 05:37:31 - mmengine - INFO - Iter(train) [ 21100/160000] lr: 8.8169e-03 eta: 21:19:44 time: 0.5532 data_time: 0.0063 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0107 decode.acc_seg: 99.5358 aux.loss_ce: 0.0085 aux.acc_seg: 99.0521 +04/17 05:37:58 - mmengine - INFO - Iter(train) [ 21150/160000] lr: 8.8141e-03 eta: 21:19:17 time: 0.5523 data_time: 0.0065 memory: 7635 loss: 0.0206 decode.loss_ce: 0.0117 decode.acc_seg: 99.4616 aux.loss_ce: 0.0090 aux.acc_seg: 98.9782 +04/17 05:38:26 - mmengine - INFO - Iter(train) [ 21200/160000] lr: 8.8112e-03 eta: 21:18:49 time: 0.5541 data_time: 0.0066 memory: 7635 loss: 0.0270 decode.loss_ce: 0.0159 decode.acc_seg: 99.1541 aux.loss_ce: 0.0110 aux.acc_seg: 98.4802 +04/17 05:38:54 - mmengine - INFO - Iter(train) [ 21250/160000] lr: 8.8084e-03 eta: 21:18:22 time: 0.5553 data_time: 0.0061 memory: 7635 loss: 0.0216 decode.loss_ce: 0.0121 decode.acc_seg: 99.5953 aux.loss_ce: 0.0095 aux.acc_seg: 99.0539 +04/17 05:39:21 - mmengine - INFO - Iter(train) [ 21300/160000] lr: 8.8056e-03 eta: 21:17:55 time: 0.5553 data_time: 0.0061 memory: 7635 loss: 0.0208 decode.loss_ce: 0.0118 decode.acc_seg: 99.6044 aux.loss_ce: 0.0090 aux.acc_seg: 99.2778 +04/17 05:39:49 - mmengine - INFO - Iter(train) [ 21350/160000] lr: 8.8028e-03 eta: 21:17:28 time: 0.5537 data_time: 0.0060 memory: 7635 loss: 0.0202 decode.loss_ce: 0.0113 decode.acc_seg: 99.4109 aux.loss_ce: 0.0089 aux.acc_seg: 98.8693 +04/17 05:40:17 - mmengine - INFO - Iter(train) [ 21400/160000] lr: 8.7999e-03 eta: 21:17:00 time: 0.5537 data_time: 0.0059 memory: 7635 loss: 0.0193 decode.loss_ce: 0.0110 decode.acc_seg: 99.3356 aux.loss_ce: 0.0083 aux.acc_seg: 99.0127 +04/17 05:40:44 - mmengine - INFO - Iter(train) [ 21450/160000] lr: 8.7971e-03 eta: 21:16:33 time: 0.5529 data_time: 0.0061 memory: 7635 loss: 0.0220 decode.loss_ce: 0.0123 decode.acc_seg: 99.6156 aux.loss_ce: 0.0097 aux.acc_seg: 99.1965 +04/17 05:41:12 - mmengine - INFO - Iter(train) [ 21500/160000] lr: 8.7943e-03 eta: 21:16:06 time: 0.5549 data_time: 0.0063 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0117 decode.acc_seg: 99.3603 aux.loss_ce: 0.0090 aux.acc_seg: 98.9705 +04/17 05:41:40 - mmengine - INFO - Iter(train) [ 21550/160000] lr: 8.7915e-03 eta: 21:15:38 time: 0.5531 data_time: 0.0071 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0107 decode.acc_seg: 99.6437 aux.loss_ce: 0.0082 aux.acc_seg: 99.3950 +04/17 05:42:08 - mmengine - INFO - Iter(train) [ 21600/160000] lr: 8.7886e-03 eta: 21:15:11 time: 0.5534 data_time: 0.0063 memory: 7635 loss: 0.0200 decode.loss_ce: 0.0112 decode.acc_seg: 99.5418 aux.loss_ce: 0.0087 aux.acc_seg: 98.9607 +04/17 05:42:35 - mmengine - INFO - Iter(train) [ 21650/160000] lr: 8.7858e-03 eta: 21:14:44 time: 0.5543 data_time: 0.0066 memory: 7635 loss: 0.0201 decode.loss_ce: 0.0113 decode.acc_seg: 99.5805 aux.loss_ce: 0.0088 aux.acc_seg: 99.2621 +04/17 05:43:03 - mmengine - INFO - Iter(train) [ 21700/160000] lr: 8.7830e-03 eta: 21:14:17 time: 0.5533 data_time: 0.0061 memory: 7635 loss: 0.0216 decode.loss_ce: 0.0123 decode.acc_seg: 99.3257 aux.loss_ce: 0.0092 aux.acc_seg: 99.0567 +04/17 05:43:31 - mmengine - INFO - Iter(train) [ 21750/160000] lr: 8.7802e-03 eta: 21:13:49 time: 0.5529 data_time: 0.0064 memory: 7635 loss: 0.0232 decode.loss_ce: 0.0135 decode.acc_seg: 99.5781 aux.loss_ce: 0.0098 aux.acc_seg: 99.1523 +04/17 05:43:58 - mmengine - INFO - Iter(train) [ 21800/160000] lr: 8.7773e-03 eta: 21:13:22 time: 0.5537 data_time: 0.0060 memory: 7635 loss: 0.0212 decode.loss_ce: 0.0123 decode.acc_seg: 99.3830 aux.loss_ce: 0.0089 aux.acc_seg: 99.0145 +04/17 05:44:26 - mmengine - INFO - Iter(train) [ 21850/160000] lr: 8.7745e-03 eta: 21:12:55 time: 0.5650 data_time: 0.0060 memory: 7635 loss: 0.0208 decode.loss_ce: 0.0118 decode.acc_seg: 99.4530 aux.loss_ce: 0.0090 aux.acc_seg: 98.9358 +04/17 05:44:54 - mmengine - INFO - Iter(train) [ 21900/160000] lr: 8.7717e-03 eta: 21:12:28 time: 0.5548 data_time: 0.0068 memory: 7635 loss: 0.0199 decode.loss_ce: 0.0112 decode.acc_seg: 99.6578 aux.loss_ce: 0.0088 aux.acc_seg: 99.1710 +04/17 05:45:22 - mmengine - INFO - Iter(train) [ 21950/160000] lr: 8.7689e-03 eta: 21:12:01 time: 0.5546 data_time: 0.0064 memory: 7635 loss: 0.0212 decode.loss_ce: 0.0121 decode.acc_seg: 99.4195 aux.loss_ce: 0.0092 aux.acc_seg: 99.0110 +04/17 05:45:49 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 05:45:49 - mmengine - INFO - Iter(train) [ 22000/160000] lr: 8.7660e-03 eta: 21:11:34 time: 0.5539 data_time: 0.0056 memory: 7635 loss: 0.0198 decode.loss_ce: 0.0111 decode.acc_seg: 99.5405 aux.loss_ce: 0.0087 aux.acc_seg: 99.0419 +04/17 05:46:17 - mmengine - INFO - Iter(train) [ 22050/160000] lr: 8.7632e-03 eta: 21:11:07 time: 0.5546 data_time: 0.0057 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0092 decode.acc_seg: 99.6042 aux.loss_ce: 0.0080 aux.acc_seg: 99.0553 +04/17 05:46:45 - mmengine - INFO - Iter(train) [ 22100/160000] lr: 8.7604e-03 eta: 21:10:40 time: 0.5547 data_time: 0.0064 memory: 7635 loss: 0.0218 decode.loss_ce: 0.0124 decode.acc_seg: 99.5714 aux.loss_ce: 0.0095 aux.acc_seg: 99.1062 +04/17 05:47:13 - mmengine - INFO - Iter(train) [ 22150/160000] lr: 8.7576e-03 eta: 21:10:14 time: 0.5547 data_time: 0.0056 memory: 7635 loss: 0.0203 decode.loss_ce: 0.0115 decode.acc_seg: 99.6792 aux.loss_ce: 0.0088 aux.acc_seg: 99.2495 +04/17 05:47:40 - mmengine - INFO - Iter(train) [ 22200/160000] lr: 8.7547e-03 eta: 21:09:46 time: 0.5549 data_time: 0.0063 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0103 decode.acc_seg: 99.4948 aux.loss_ce: 0.0088 aux.acc_seg: 98.9191 +04/17 05:48:08 - mmengine - INFO - Iter(train) [ 22250/160000] lr: 8.7519e-03 eta: 21:09:19 time: 0.5556 data_time: 0.0056 memory: 7635 loss: 0.0198 decode.loss_ce: 0.0112 decode.acc_seg: 99.6146 aux.loss_ce: 0.0086 aux.acc_seg: 99.2919 +04/17 05:48:36 - mmengine - INFO - Iter(train) [ 22300/160000] lr: 8.7491e-03 eta: 21:08:52 time: 0.5558 data_time: 0.0060 memory: 7635 loss: 0.0218 decode.loss_ce: 0.0122 decode.acc_seg: 99.5606 aux.loss_ce: 0.0096 aux.acc_seg: 99.0362 +04/17 05:49:04 - mmengine - INFO - Iter(train) [ 22350/160000] lr: 8.7463e-03 eta: 21:08:25 time: 0.5543 data_time: 0.0058 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0103 decode.acc_seg: 99.5120 aux.loss_ce: 0.0086 aux.acc_seg: 98.9896 +04/17 05:49:31 - mmengine - INFO - Iter(train) [ 22400/160000] lr: 8.7434e-03 eta: 21:07:58 time: 0.5532 data_time: 0.0065 memory: 7635 loss: 0.0224 decode.loss_ce: 0.0127 decode.acc_seg: 99.5862 aux.loss_ce: 0.0098 aux.acc_seg: 99.1762 +04/17 05:49:59 - mmengine - INFO - Iter(train) [ 22450/160000] lr: 8.7406e-03 eta: 21:07:31 time: 0.5553 data_time: 0.0066 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0102 decode.acc_seg: 99.6047 aux.loss_ce: 0.0083 aux.acc_seg: 99.0493 +04/17 05:50:27 - mmengine - INFO - Iter(train) [ 22500/160000] lr: 8.7378e-03 eta: 21:07:03 time: 0.5543 data_time: 0.0065 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0120 decode.acc_seg: 99.4661 aux.loss_ce: 0.0087 aux.acc_seg: 98.9499 +04/17 05:50:54 - mmengine - INFO - Iter(train) [ 22550/160000] lr: 8.7350e-03 eta: 21:06:36 time: 0.5536 data_time: 0.0059 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0099 decode.acc_seg: 99.5499 aux.loss_ce: 0.0080 aux.acc_seg: 99.1887 +04/17 05:51:22 - mmengine - INFO - Iter(train) [ 22600/160000] lr: 8.7321e-03 eta: 21:06:09 time: 0.5534 data_time: 0.0064 memory: 7635 loss: 0.0197 decode.loss_ce: 0.0111 decode.acc_seg: 99.5870 aux.loss_ce: 0.0086 aux.acc_seg: 99.2036 +04/17 05:51:50 - mmengine - INFO - Iter(train) [ 22650/160000] lr: 8.7293e-03 eta: 21:05:41 time: 0.5551 data_time: 0.0064 memory: 7635 loss: 0.0198 decode.loss_ce: 0.0111 decode.acc_seg: 99.6707 aux.loss_ce: 0.0086 aux.acc_seg: 99.2799 +04/17 05:52:18 - mmengine - INFO - Iter(train) [ 22700/160000] lr: 8.7265e-03 eta: 21:05:14 time: 0.5533 data_time: 0.0059 memory: 7635 loss: 0.0218 decode.loss_ce: 0.0124 decode.acc_seg: 99.7109 aux.loss_ce: 0.0094 aux.acc_seg: 99.4370 +04/17 05:52:45 - mmengine - INFO - Iter(train) [ 22750/160000] lr: 8.7236e-03 eta: 21:04:46 time: 0.5532 data_time: 0.0059 memory: 7635 loss: 0.0206 decode.loss_ce: 0.0112 decode.acc_seg: 99.5122 aux.loss_ce: 0.0094 aux.acc_seg: 99.0606 +04/17 05:53:13 - mmengine - INFO - Iter(train) [ 22800/160000] lr: 8.7208e-03 eta: 21:04:19 time: 0.5537 data_time: 0.0066 memory: 7635 loss: 0.0208 decode.loss_ce: 0.0115 decode.acc_seg: 99.4382 aux.loss_ce: 0.0092 aux.acc_seg: 99.0163 +04/17 05:53:41 - mmengine - INFO - Iter(train) [ 22850/160000] lr: 8.7180e-03 eta: 21:03:52 time: 0.5538 data_time: 0.0064 memory: 7635 loss: 0.0215 decode.loss_ce: 0.0120 decode.acc_seg: 99.6519 aux.loss_ce: 0.0095 aux.acc_seg: 99.1600 +04/17 05:54:08 - mmengine - INFO - Iter(train) [ 22900/160000] lr: 8.7152e-03 eta: 21:03:24 time: 0.5537 data_time: 0.0058 memory: 7635 loss: 0.0200 decode.loss_ce: 0.0110 decode.acc_seg: 99.4405 aux.loss_ce: 0.0090 aux.acc_seg: 98.8873 +04/17 05:54:36 - mmengine - INFO - Iter(train) [ 22950/160000] lr: 8.7123e-03 eta: 21:02:57 time: 0.5542 data_time: 0.0071 memory: 7635 loss: 0.0216 decode.loss_ce: 0.0121 decode.acc_seg: 99.5530 aux.loss_ce: 0.0095 aux.acc_seg: 98.8122 +04/17 05:55:04 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 05:55:04 - mmengine - INFO - Iter(train) [ 23000/160000] lr: 8.7095e-03 eta: 21:02:30 time: 0.5547 data_time: 0.0071 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0104 decode.acc_seg: 99.4958 aux.loss_ce: 0.0087 aux.acc_seg: 99.2250 +04/17 05:55:31 - mmengine - INFO - Iter(train) [ 23050/160000] lr: 8.7067e-03 eta: 21:02:02 time: 0.5531 data_time: 0.0063 memory: 7635 loss: 0.0197 decode.loss_ce: 0.0109 decode.acc_seg: 99.6135 aux.loss_ce: 0.0088 aux.acc_seg: 99.1367 +04/17 05:55:59 - mmengine - INFO - Iter(train) [ 23100/160000] lr: 8.7038e-03 eta: 21:01:35 time: 0.5545 data_time: 0.0063 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0101 decode.acc_seg: 99.5773 aux.loss_ce: 0.0084 aux.acc_seg: 99.1000 +04/17 05:56:27 - mmengine - INFO - Iter(train) [ 23150/160000] lr: 8.7010e-03 eta: 21:01:08 time: 0.5630 data_time: 0.0062 memory: 7635 loss: 0.0193 decode.loss_ce: 0.0106 decode.acc_seg: 99.6355 aux.loss_ce: 0.0087 aux.acc_seg: 99.1062 +04/17 05:56:55 - mmengine - INFO - Iter(train) [ 23200/160000] lr: 8.6982e-03 eta: 21:00:41 time: 0.5541 data_time: 0.0068 memory: 7635 loss: 0.0198 decode.loss_ce: 0.0109 decode.acc_seg: 99.6552 aux.loss_ce: 0.0089 aux.acc_seg: 99.1572 +04/17 05:57:22 - mmengine - INFO - Iter(train) [ 23250/160000] lr: 8.6954e-03 eta: 21:00:14 time: 0.5542 data_time: 0.0061 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0116 decode.acc_seg: 99.3690 aux.loss_ce: 0.0093 aux.acc_seg: 99.0586 +04/17 05:57:50 - mmengine - INFO - Iter(train) [ 23300/160000] lr: 8.6925e-03 eta: 20:59:46 time: 0.5526 data_time: 0.0058 memory: 7635 loss: 0.0206 decode.loss_ce: 0.0120 decode.acc_seg: 99.5623 aux.loss_ce: 0.0086 aux.acc_seg: 99.2438 +04/17 05:58:18 - mmengine - INFO - Iter(train) [ 23350/160000] lr: 8.6897e-03 eta: 20:59:19 time: 0.5544 data_time: 0.0068 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0101 decode.acc_seg: 99.4234 aux.loss_ce: 0.0083 aux.acc_seg: 98.9252 +04/17 05:58:45 - mmengine - INFO - Iter(train) [ 23400/160000] lr: 8.6869e-03 eta: 20:58:51 time: 0.5530 data_time: 0.0062 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0115 decode.acc_seg: 99.4738 aux.loss_ce: 0.0093 aux.acc_seg: 99.0345 +04/17 05:59:13 - mmengine - INFO - Iter(train) [ 23450/160000] lr: 8.6840e-03 eta: 20:58:24 time: 0.5542 data_time: 0.0066 memory: 7635 loss: 0.0198 decode.loss_ce: 0.0113 decode.acc_seg: 99.6345 aux.loss_ce: 0.0085 aux.acc_seg: 99.2324 +04/17 05:59:41 - mmengine - INFO - Iter(train) [ 23500/160000] lr: 8.6812e-03 eta: 20:57:56 time: 0.5520 data_time: 0.0062 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0103 decode.acc_seg: 99.5246 aux.loss_ce: 0.0084 aux.acc_seg: 99.1241 +04/17 06:00:08 - mmengine - INFO - Iter(train) [ 23550/160000] lr: 8.6784e-03 eta: 20:57:29 time: 0.5542 data_time: 0.0061 memory: 7635 loss: 0.0196 decode.loss_ce: 0.0108 decode.acc_seg: 99.6054 aux.loss_ce: 0.0088 aux.acc_seg: 99.2346 +04/17 06:00:36 - mmengine - INFO - Iter(train) [ 23600/160000] lr: 8.6756e-03 eta: 20:57:02 time: 0.5533 data_time: 0.0067 memory: 7635 loss: 0.0213 decode.loss_ce: 0.0121 decode.acc_seg: 99.5667 aux.loss_ce: 0.0092 aux.acc_seg: 99.2044 +04/17 06:01:04 - mmengine - INFO - Iter(train) [ 23650/160000] lr: 8.6727e-03 eta: 20:56:34 time: 0.5542 data_time: 0.0063 memory: 7635 loss: 0.0213 decode.loss_ce: 0.0121 decode.acc_seg: 99.3974 aux.loss_ce: 0.0092 aux.acc_seg: 98.9027 +04/17 06:01:32 - mmengine - INFO - Iter(train) [ 23700/160000] lr: 8.6699e-03 eta: 20:56:07 time: 0.5537 data_time: 0.0064 memory: 7635 loss: 0.0215 decode.loss_ce: 0.0122 decode.acc_seg: 99.4102 aux.loss_ce: 0.0094 aux.acc_seg: 98.7565 +04/17 06:01:59 - mmengine - INFO - Iter(train) [ 23750/160000] lr: 8.6671e-03 eta: 20:55:40 time: 0.5543 data_time: 0.0064 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0095 decode.acc_seg: 99.7561 aux.loss_ce: 0.0077 aux.acc_seg: 99.5364 +04/17 06:02:27 - mmengine - INFO - Iter(train) [ 23800/160000] lr: 8.6642e-03 eta: 20:55:13 time: 0.5564 data_time: 0.0062 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0094 decode.acc_seg: 99.7054 aux.loss_ce: 0.0081 aux.acc_seg: 99.1726 +04/17 06:02:55 - mmengine - INFO - Iter(train) [ 23850/160000] lr: 8.6614e-03 eta: 20:54:45 time: 0.5557 data_time: 0.0070 memory: 7635 loss: 0.0190 decode.loss_ce: 0.0106 decode.acc_seg: 99.5234 aux.loss_ce: 0.0084 aux.acc_seg: 99.0002 +04/17 06:03:22 - mmengine - INFO - Iter(train) [ 23900/160000] lr: 8.6586e-03 eta: 20:54:18 time: 0.5536 data_time: 0.0067 memory: 7635 loss: 0.0222 decode.loss_ce: 0.0127 decode.acc_seg: 99.5132 aux.loss_ce: 0.0095 aux.acc_seg: 98.9788 +04/17 06:03:50 - mmengine - INFO - Iter(train) [ 23950/160000] lr: 8.6558e-03 eta: 20:53:51 time: 0.5545 data_time: 0.0057 memory: 7635 loss: 0.0205 decode.loss_ce: 0.0115 decode.acc_seg: 99.4175 aux.loss_ce: 0.0090 aux.acc_seg: 98.9756 +04/17 06:04:18 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 06:04:18 - mmengine - INFO - Iter(train) [ 24000/160000] lr: 8.6529e-03 eta: 20:53:24 time: 0.5539 data_time: 0.0065 memory: 7635 loss: 0.0214 decode.loss_ce: 0.0121 decode.acc_seg: 99.4178 aux.loss_ce: 0.0094 aux.acc_seg: 98.9408 +04/17 06:04:46 - mmengine - INFO - Iter(train) [ 24050/160000] lr: 8.6501e-03 eta: 20:52:56 time: 0.5557 data_time: 0.0064 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0116 decode.acc_seg: 99.5782 aux.loss_ce: 0.0091 aux.acc_seg: 99.3015 +04/17 06:05:13 - mmengine - INFO - Iter(train) [ 24100/160000] lr: 8.6473e-03 eta: 20:52:29 time: 0.5527 data_time: 0.0062 memory: 7635 loss: 0.0204 decode.loss_ce: 0.0113 decode.acc_seg: 99.6398 aux.loss_ce: 0.0091 aux.acc_seg: 99.2044 +04/17 06:05:41 - mmengine - INFO - Iter(train) [ 24150/160000] lr: 8.6444e-03 eta: 20:52:02 time: 0.5551 data_time: 0.0065 memory: 7635 loss: 0.0199 decode.loss_ce: 0.0110 decode.acc_seg: 99.5223 aux.loss_ce: 0.0088 aux.acc_seg: 99.1008 +04/17 06:06:09 - mmengine - INFO - Iter(train) [ 24200/160000] lr: 8.6416e-03 eta: 20:51:34 time: 0.5539 data_time: 0.0060 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0109 decode.acc_seg: 99.4917 aux.loss_ce: 0.0086 aux.acc_seg: 99.1086 +04/17 06:06:36 - mmengine - INFO - Iter(train) [ 24250/160000] lr: 8.6388e-03 eta: 20:51:07 time: 0.5524 data_time: 0.0058 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0097 decode.acc_seg: 99.6735 aux.loss_ce: 0.0082 aux.acc_seg: 99.2729 +04/17 06:07:04 - mmengine - INFO - Iter(train) [ 24300/160000] lr: 8.6359e-03 eta: 20:50:40 time: 0.5532 data_time: 0.0060 memory: 7635 loss: 0.0223 decode.loss_ce: 0.0126 decode.acc_seg: 99.5812 aux.loss_ce: 0.0097 aux.acc_seg: 99.1949 +04/17 06:07:32 - mmengine - INFO - Iter(train) [ 24350/160000] lr: 8.6331e-03 eta: 20:50:12 time: 0.5541 data_time: 0.0060 memory: 7635 loss: 0.0203 decode.loss_ce: 0.0113 decode.acc_seg: 99.5622 aux.loss_ce: 0.0090 aux.acc_seg: 99.1330 +04/17 06:08:00 - mmengine - INFO - Iter(train) [ 24400/160000] lr: 8.6303e-03 eta: 20:49:45 time: 0.5535 data_time: 0.0059 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0115 decode.acc_seg: 99.6341 aux.loss_ce: 0.0093 aux.acc_seg: 99.1335 +04/17 06:08:27 - mmengine - INFO - Iter(train) [ 24450/160000] lr: 8.6275e-03 eta: 20:49:17 time: 0.5531 data_time: 0.0063 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0108 decode.acc_seg: 99.5450 aux.loss_ce: 0.0087 aux.acc_seg: 98.9058 +04/17 06:08:55 - mmengine - INFO - Iter(train) [ 24500/160000] lr: 8.6246e-03 eta: 20:48:50 time: 0.5521 data_time: 0.0063 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0105 decode.acc_seg: 99.5281 aux.loss_ce: 0.0087 aux.acc_seg: 99.0461 +04/17 06:09:23 - mmengine - INFO - Iter(train) [ 24550/160000] lr: 8.6218e-03 eta: 20:48:22 time: 0.5540 data_time: 0.0065 memory: 7635 loss: 0.0193 decode.loss_ce: 0.0106 decode.acc_seg: 99.5327 aux.loss_ce: 0.0087 aux.acc_seg: 98.9229 +04/17 06:09:50 - mmengine - INFO - Iter(train) [ 24600/160000] lr: 8.6190e-03 eta: 20:47:55 time: 0.5550 data_time: 0.0068 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0100 decode.acc_seg: 99.4426 aux.loss_ce: 0.0084 aux.acc_seg: 98.8665 +04/17 06:10:18 - mmengine - INFO - Iter(train) [ 24650/160000] lr: 8.6161e-03 eta: 20:47:27 time: 0.5545 data_time: 0.0057 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0107 decode.acc_seg: 99.5864 aux.loss_ce: 0.0085 aux.acc_seg: 99.0477 +04/17 06:10:46 - mmengine - INFO - Iter(train) [ 24700/160000] lr: 8.6133e-03 eta: 20:47:00 time: 0.5544 data_time: 0.0065 memory: 7635 loss: 0.0208 decode.loss_ce: 0.0118 decode.acc_seg: 99.6146 aux.loss_ce: 0.0090 aux.acc_seg: 99.0986 +04/17 06:11:13 - mmengine - INFO - Iter(train) [ 24750/160000] lr: 8.6105e-03 eta: 20:46:33 time: 0.5554 data_time: 0.0066 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0093 decode.acc_seg: 99.5455 aux.loss_ce: 0.0078 aux.acc_seg: 99.0612 +04/17 06:11:41 - mmengine - INFO - Iter(train) [ 24800/160000] lr: 8.6076e-03 eta: 20:46:06 time: 0.5547 data_time: 0.0059 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0116 decode.acc_seg: 99.6256 aux.loss_ce: 0.0093 aux.acc_seg: 99.1723 +04/17 06:12:09 - mmengine - INFO - Iter(train) [ 24850/160000] lr: 8.6048e-03 eta: 20:45:38 time: 0.5528 data_time: 0.0058 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0107 decode.acc_seg: 99.5779 aux.loss_ce: 0.0085 aux.acc_seg: 98.9428 +04/17 06:12:36 - mmengine - INFO - Iter(train) [ 24900/160000] lr: 8.6020e-03 eta: 20:45:11 time: 0.5531 data_time: 0.0060 memory: 7635 loss: 0.0194 decode.loss_ce: 0.0108 decode.acc_seg: 99.2304 aux.loss_ce: 0.0086 aux.acc_seg: 98.9576 +04/17 06:13:04 - mmengine - INFO - Iter(train) [ 24950/160000] lr: 8.5991e-03 eta: 20:44:43 time: 0.5525 data_time: 0.0065 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0117 decode.acc_seg: 99.5942 aux.loss_ce: 0.0092 aux.acc_seg: 99.1375 +04/17 06:13:32 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 06:13:32 - mmengine - INFO - Iter(train) [ 25000/160000] lr: 8.5963e-03 eta: 20:44:15 time: 0.5556 data_time: 0.0065 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0120 decode.acc_seg: 99.3312 aux.loss_ce: 0.0088 aux.acc_seg: 99.1397 +04/17 06:13:59 - mmengine - INFO - Iter(train) [ 25050/160000] lr: 8.5935e-03 eta: 20:43:48 time: 0.5528 data_time: 0.0061 memory: 7635 loss: 0.0197 decode.loss_ce: 0.0107 decode.acc_seg: 99.4274 aux.loss_ce: 0.0089 aux.acc_seg: 98.9287 +04/17 06:14:27 - mmengine - INFO - Iter(train) [ 25100/160000] lr: 8.5906e-03 eta: 20:43:21 time: 0.5545 data_time: 0.0065 memory: 7635 loss: 0.0197 decode.loss_ce: 0.0110 decode.acc_seg: 99.4225 aux.loss_ce: 0.0086 aux.acc_seg: 98.9243 +04/17 06:14:55 - mmengine - INFO - Iter(train) [ 25150/160000] lr: 8.5878e-03 eta: 20:42:53 time: 0.5540 data_time: 0.0063 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0101 decode.acc_seg: 99.5281 aux.loss_ce: 0.0083 aux.acc_seg: 99.1858 +04/17 06:15:23 - mmengine - INFO - Iter(train) [ 25200/160000] lr: 8.5850e-03 eta: 20:42:26 time: 0.5523 data_time: 0.0063 memory: 7635 loss: 0.0193 decode.loss_ce: 0.0106 decode.acc_seg: 99.6846 aux.loss_ce: 0.0087 aux.acc_seg: 99.2772 +04/17 06:15:50 - mmengine - INFO - Iter(train) [ 25250/160000] lr: 8.5821e-03 eta: 20:41:58 time: 0.5545 data_time: 0.0069 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0118 decode.acc_seg: 99.4268 aux.loss_ce: 0.0091 aux.acc_seg: 99.0958 +04/17 06:16:18 - mmengine - INFO - Iter(train) [ 25300/160000] lr: 8.5793e-03 eta: 20:41:31 time: 0.5526 data_time: 0.0061 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0115 decode.acc_seg: 99.4090 aux.loss_ce: 0.0092 aux.acc_seg: 98.9176 +04/17 06:16:46 - mmengine - INFO - Iter(train) [ 25350/160000] lr: 8.5765e-03 eta: 20:41:04 time: 0.5528 data_time: 0.0065 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0104 decode.acc_seg: 99.6112 aux.loss_ce: 0.0084 aux.acc_seg: 99.1398 +04/17 06:17:13 - mmengine - INFO - Iter(train) [ 25400/160000] lr: 8.5736e-03 eta: 20:40:36 time: 0.5527 data_time: 0.0061 memory: 7635 loss: 0.0190 decode.loss_ce: 0.0104 decode.acc_seg: 99.6025 aux.loss_ce: 0.0086 aux.acc_seg: 99.0537 +04/17 06:17:41 - mmengine - INFO - Iter(train) [ 25450/160000] lr: 8.5708e-03 eta: 20:40:09 time: 0.5544 data_time: 0.0061 memory: 7635 loss: 0.0206 decode.loss_ce: 0.0114 decode.acc_seg: 99.5256 aux.loss_ce: 0.0092 aux.acc_seg: 98.8937 +04/17 06:18:09 - mmengine - INFO - Iter(train) [ 25500/160000] lr: 8.5680e-03 eta: 20:39:41 time: 0.5536 data_time: 0.0063 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0103 decode.acc_seg: 99.5145 aux.loss_ce: 0.0088 aux.acc_seg: 98.8540 +04/17 06:18:36 - mmengine - INFO - Iter(train) [ 25550/160000] lr: 8.5651e-03 eta: 20:39:14 time: 0.5524 data_time: 0.0066 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0096 decode.acc_seg: 99.6950 aux.loss_ce: 0.0080 aux.acc_seg: 99.2313 +04/17 06:19:04 - mmengine - INFO - Iter(train) [ 25600/160000] lr: 8.5623e-03 eta: 20:38:46 time: 0.5530 data_time: 0.0057 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0101 decode.acc_seg: 99.5444 aux.loss_ce: 0.0082 aux.acc_seg: 99.2862 +04/17 06:19:32 - mmengine - INFO - Iter(train) [ 25650/160000] lr: 8.5595e-03 eta: 20:38:19 time: 0.5529 data_time: 0.0065 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0101 decode.acc_seg: 99.7309 aux.loss_ce: 0.0086 aux.acc_seg: 99.2523 +04/17 06:19:59 - mmengine - INFO - Iter(train) [ 25700/160000] lr: 8.5566e-03 eta: 20:37:51 time: 0.5544 data_time: 0.0076 memory: 7635 loss: 0.0254 decode.loss_ce: 0.0151 decode.acc_seg: 99.5668 aux.loss_ce: 0.0102 aux.acc_seg: 98.9284 +04/17 06:20:27 - mmengine - INFO - Iter(train) [ 25750/160000] lr: 8.5538e-03 eta: 20:37:24 time: 0.5540 data_time: 0.0059 memory: 7635 loss: 0.0198 decode.loss_ce: 0.0110 decode.acc_seg: 99.4843 aux.loss_ce: 0.0089 aux.acc_seg: 98.9662 +04/17 06:20:55 - mmengine - INFO - Iter(train) [ 25800/160000] lr: 8.5510e-03 eta: 20:36:56 time: 0.5533 data_time: 0.0060 memory: 7635 loss: 0.0206 decode.loss_ce: 0.0112 decode.acc_seg: 99.5192 aux.loss_ce: 0.0094 aux.acc_seg: 98.9622 +04/17 06:21:22 - mmengine - INFO - Iter(train) [ 25850/160000] lr: 8.5481e-03 eta: 20:36:28 time: 0.5540 data_time: 0.0062 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0118 decode.acc_seg: 99.5270 aux.loss_ce: 0.0091 aux.acc_seg: 99.0009 +04/17 06:21:50 - mmengine - INFO - Iter(train) [ 25900/160000] lr: 8.5453e-03 eta: 20:36:01 time: 0.5534 data_time: 0.0064 memory: 7635 loss: 0.0214 decode.loss_ce: 0.0118 decode.acc_seg: 99.5065 aux.loss_ce: 0.0096 aux.acc_seg: 99.0421 +04/17 06:22:18 - mmengine - INFO - Iter(train) [ 25950/160000] lr: 8.5425e-03 eta: 20:35:33 time: 0.5534 data_time: 0.0062 memory: 7635 loss: 0.0212 decode.loss_ce: 0.0119 decode.acc_seg: 99.4792 aux.loss_ce: 0.0093 aux.acc_seg: 98.8772 +04/17 06:22:45 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 06:22:45 - mmengine - INFO - Iter(train) [ 26000/160000] lr: 8.5396e-03 eta: 20:35:05 time: 0.5525 data_time: 0.0064 memory: 7635 loss: 0.0202 decode.loss_ce: 0.0112 decode.acc_seg: 99.6405 aux.loss_ce: 0.0090 aux.acc_seg: 99.1919 +04/17 06:23:13 - mmengine - INFO - Iter(train) [ 26050/160000] lr: 8.5368e-03 eta: 20:34:38 time: 0.5534 data_time: 0.0057 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0111 decode.acc_seg: 99.5835 aux.loss_ce: 0.0084 aux.acc_seg: 99.1900 +04/17 06:23:41 - mmengine - INFO - Iter(train) [ 26100/160000] lr: 8.5340e-03 eta: 20:34:10 time: 0.5538 data_time: 0.0060 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0116 decode.acc_seg: 99.5837 aux.loss_ce: 0.0091 aux.acc_seg: 99.0800 +04/17 06:24:09 - mmengine - INFO - Iter(train) [ 26150/160000] lr: 8.5311e-03 eta: 20:33:43 time: 0.5633 data_time: 0.0062 memory: 7635 loss: 0.0217 decode.loss_ce: 0.0124 decode.acc_seg: 99.5199 aux.loss_ce: 0.0094 aux.acc_seg: 98.9968 +04/17 06:24:36 - mmengine - INFO - Iter(train) [ 26200/160000] lr: 8.5283e-03 eta: 20:33:16 time: 0.5547 data_time: 0.0062 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0117 decode.acc_seg: 99.6347 aux.loss_ce: 0.0090 aux.acc_seg: 99.1317 +04/17 06:25:04 - mmengine - INFO - Iter(train) [ 26250/160000] lr: 8.5255e-03 eta: 20:32:48 time: 0.5541 data_time: 0.0064 memory: 7635 loss: 0.0217 decode.loss_ce: 0.0119 decode.acc_seg: 99.5460 aux.loss_ce: 0.0098 aux.acc_seg: 99.1327 +04/17 06:25:32 - mmengine - INFO - Iter(train) [ 26300/160000] lr: 8.5226e-03 eta: 20:32:21 time: 0.5540 data_time: 0.0070 memory: 7635 loss: 0.0194 decode.loss_ce: 0.0107 decode.acc_seg: 99.5099 aux.loss_ce: 0.0088 aux.acc_seg: 98.8997 +04/17 06:25:59 - mmengine - INFO - Iter(train) [ 26350/160000] lr: 8.5198e-03 eta: 20:31:53 time: 0.5527 data_time: 0.0067 memory: 7635 loss: 0.0211 decode.loss_ce: 0.0117 decode.acc_seg: 99.4177 aux.loss_ce: 0.0094 aux.acc_seg: 99.0552 +04/17 06:26:27 - mmengine - INFO - Iter(train) [ 26400/160000] lr: 8.5170e-03 eta: 20:31:26 time: 0.5544 data_time: 0.0063 memory: 7635 loss: 0.0196 decode.loss_ce: 0.0107 decode.acc_seg: 99.5645 aux.loss_ce: 0.0089 aux.acc_seg: 99.2006 +04/17 06:26:55 - mmengine - INFO - Iter(train) [ 26450/160000] lr: 8.5141e-03 eta: 20:30:59 time: 0.5534 data_time: 0.0060 memory: 7635 loss: 0.0202 decode.loss_ce: 0.0111 decode.acc_seg: 99.5743 aux.loss_ce: 0.0091 aux.acc_seg: 99.2047 +04/17 06:27:22 - mmengine - INFO - Iter(train) [ 26500/160000] lr: 8.5113e-03 eta: 20:30:31 time: 0.5532 data_time: 0.0062 memory: 7635 loss: 0.0198 decode.loss_ce: 0.0113 decode.acc_seg: 99.6210 aux.loss_ce: 0.0085 aux.acc_seg: 99.0860 +04/17 06:27:50 - mmengine - INFO - Iter(train) [ 26550/160000] lr: 8.5085e-03 eta: 20:30:04 time: 0.5527 data_time: 0.0059 memory: 7635 loss: 0.0204 decode.loss_ce: 0.0113 decode.acc_seg: 99.4643 aux.loss_ce: 0.0091 aux.acc_seg: 98.9594 +04/17 06:28:18 - mmengine - INFO - Iter(train) [ 26600/160000] lr: 8.5056e-03 eta: 20:29:36 time: 0.5533 data_time: 0.0065 memory: 7635 loss: 0.0206 decode.loss_ce: 0.0116 decode.acc_seg: 99.3349 aux.loss_ce: 0.0090 aux.acc_seg: 99.0543 +04/17 06:28:45 - mmengine - INFO - Iter(train) [ 26650/160000] lr: 8.5028e-03 eta: 20:29:08 time: 0.5529 data_time: 0.0061 memory: 7635 loss: 0.0197 decode.loss_ce: 0.0110 decode.acc_seg: 99.2491 aux.loss_ce: 0.0087 aux.acc_seg: 98.7731 +04/17 06:29:13 - mmengine - INFO - Iter(train) [ 26700/160000] lr: 8.5000e-03 eta: 20:28:41 time: 0.5536 data_time: 0.0070 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0102 decode.acc_seg: 99.4468 aux.loss_ce: 0.0081 aux.acc_seg: 99.0105 +04/17 06:29:41 - mmengine - INFO - Iter(train) [ 26750/160000] lr: 8.4971e-03 eta: 20:28:13 time: 0.5538 data_time: 0.0060 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0115 decode.acc_seg: 99.6280 aux.loss_ce: 0.0092 aux.acc_seg: 99.2716 +04/17 06:30:08 - mmengine - INFO - Iter(train) [ 26800/160000] lr: 8.4943e-03 eta: 20:27:46 time: 0.5540 data_time: 0.0070 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0100 decode.acc_seg: 99.6223 aux.loss_ce: 0.0087 aux.acc_seg: 99.2036 +04/17 06:30:36 - mmengine - INFO - Iter(train) [ 26850/160000] lr: 8.4914e-03 eta: 20:27:18 time: 0.5525 data_time: 0.0062 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0111 decode.acc_seg: 99.6864 aux.loss_ce: 0.0085 aux.acc_seg: 99.2263 +04/17 06:31:04 - mmengine - INFO - Iter(train) [ 26900/160000] lr: 8.4886e-03 eta: 20:26:51 time: 0.5534 data_time: 0.0067 memory: 7635 loss: 0.0204 decode.loss_ce: 0.0113 decode.acc_seg: 99.5808 aux.loss_ce: 0.0092 aux.acc_seg: 99.0973 +04/17 06:31:31 - mmengine - INFO - Iter(train) [ 26950/160000] lr: 8.4858e-03 eta: 20:26:23 time: 0.5537 data_time: 0.0063 memory: 7635 loss: 0.0201 decode.loss_ce: 0.0110 decode.acc_seg: 99.6390 aux.loss_ce: 0.0091 aux.acc_seg: 99.2087 +04/17 06:31:59 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 06:31:59 - mmengine - INFO - Iter(train) [ 27000/160000] lr: 8.4829e-03 eta: 20:25:55 time: 0.5525 data_time: 0.0058 memory: 7635 loss: 0.0181 decode.loss_ce: 0.0101 decode.acc_seg: 99.5855 aux.loss_ce: 0.0080 aux.acc_seg: 99.1312 +04/17 06:32:27 - mmengine - INFO - Iter(train) [ 27050/160000] lr: 8.4801e-03 eta: 20:25:28 time: 0.5531 data_time: 0.0059 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0095 decode.acc_seg: 99.6612 aux.loss_ce: 0.0082 aux.acc_seg: 99.3042 +04/17 06:32:54 - mmengine - INFO - Iter(train) [ 27100/160000] lr: 8.4773e-03 eta: 20:25:00 time: 0.5542 data_time: 0.0060 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0117 decode.acc_seg: 99.5104 aux.loss_ce: 0.0090 aux.acc_seg: 99.0363 +04/17 06:33:22 - mmengine - INFO - Iter(train) [ 27150/160000] lr: 8.4744e-03 eta: 20:24:32 time: 0.5541 data_time: 0.0066 memory: 7635 loss: 0.0194 decode.loss_ce: 0.0106 decode.acc_seg: 99.5104 aux.loss_ce: 0.0087 aux.acc_seg: 98.9560 +04/17 06:33:50 - mmengine - INFO - Iter(train) [ 27200/160000] lr: 8.4716e-03 eta: 20:24:06 time: 0.5650 data_time: 0.0064 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0111 decode.acc_seg: 99.5119 aux.loss_ce: 0.0084 aux.acc_seg: 99.2408 +04/17 06:34:18 - mmengine - INFO - Iter(train) [ 27250/160000] lr: 8.4688e-03 eta: 20:23:38 time: 0.5514 data_time: 0.0064 memory: 7635 loss: 0.0201 decode.loss_ce: 0.0113 decode.acc_seg: 99.5870 aux.loss_ce: 0.0088 aux.acc_seg: 99.2472 +04/17 06:34:45 - mmengine - INFO - Iter(train) [ 27300/160000] lr: 8.4659e-03 eta: 20:23:10 time: 0.5536 data_time: 0.0063 memory: 7635 loss: 0.0243 decode.loss_ce: 0.0145 decode.acc_seg: 99.5888 aux.loss_ce: 0.0098 aux.acc_seg: 99.1717 +04/17 06:35:13 - mmengine - INFO - Iter(train) [ 27350/160000] lr: 8.4631e-03 eta: 20:22:43 time: 0.5542 data_time: 0.0064 memory: 7635 loss: 0.0202 decode.loss_ce: 0.0113 decode.acc_seg: 99.4240 aux.loss_ce: 0.0089 aux.acc_seg: 98.7696 +04/17 06:35:41 - mmengine - INFO - Iter(train) [ 27400/160000] lr: 8.4602e-03 eta: 20:22:16 time: 0.5525 data_time: 0.0064 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0096 decode.acc_seg: 99.5619 aux.loss_ce: 0.0087 aux.acc_seg: 99.0899 +04/17 06:36:08 - mmengine - INFO - Iter(train) [ 27450/160000] lr: 8.4574e-03 eta: 20:21:48 time: 0.5526 data_time: 0.0062 memory: 7635 loss: 0.0198 decode.loss_ce: 0.0111 decode.acc_seg: 99.4605 aux.loss_ce: 0.0087 aux.acc_seg: 99.0443 +04/17 06:36:36 - mmengine - INFO - Iter(train) [ 27500/160000] lr: 8.4546e-03 eta: 20:21:21 time: 0.5548 data_time: 0.0059 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0103 decode.acc_seg: 99.5278 aux.loss_ce: 0.0082 aux.acc_seg: 99.1844 +04/17 06:37:04 - mmengine - INFO - Iter(train) [ 27550/160000] lr: 8.4517e-03 eta: 20:20:54 time: 0.5533 data_time: 0.0062 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0107 decode.acc_seg: 99.6335 aux.loss_ce: 0.0086 aux.acc_seg: 99.1250 +04/17 06:37:32 - mmengine - INFO - Iter(train) [ 27600/160000] lr: 8.4489e-03 eta: 20:20:26 time: 0.5542 data_time: 0.0056 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0106 decode.acc_seg: 99.6546 aux.loss_ce: 0.0090 aux.acc_seg: 99.0662 +04/17 06:37:59 - mmengine - INFO - Iter(train) [ 27650/160000] lr: 8.4461e-03 eta: 20:19:58 time: 0.5531 data_time: 0.0063 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0106 decode.acc_seg: 99.5378 aux.loss_ce: 0.0085 aux.acc_seg: 99.1428 +04/17 06:38:27 - mmengine - INFO - Iter(train) [ 27700/160000] lr: 8.4432e-03 eta: 20:19:31 time: 0.5529 data_time: 0.0064 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0095 decode.acc_seg: 99.5130 aux.loss_ce: 0.0080 aux.acc_seg: 99.1068 +04/17 06:38:54 - mmengine - INFO - Iter(train) [ 27750/160000] lr: 8.4404e-03 eta: 20:19:03 time: 0.5516 data_time: 0.0058 memory: 7635 loss: 0.0199 decode.loss_ce: 0.0110 decode.acc_seg: 99.6272 aux.loss_ce: 0.0089 aux.acc_seg: 99.2136 +04/17 06:39:22 - mmengine - INFO - Iter(train) [ 27800/160000] lr: 8.4375e-03 eta: 20:18:36 time: 0.5553 data_time: 0.0067 memory: 7635 loss: 0.0193 decode.loss_ce: 0.0109 decode.acc_seg: 99.6243 aux.loss_ce: 0.0084 aux.acc_seg: 99.2695 +04/17 06:39:50 - mmengine - INFO - Iter(train) [ 27850/160000] lr: 8.4347e-03 eta: 20:18:08 time: 0.5550 data_time: 0.0058 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0099 decode.acc_seg: 99.5437 aux.loss_ce: 0.0082 aux.acc_seg: 98.9361 +04/17 06:40:18 - mmengine - INFO - Iter(train) [ 27900/160000] lr: 8.4319e-03 eta: 20:17:41 time: 0.5543 data_time: 0.0062 memory: 7635 loss: 0.0231 decode.loss_ce: 0.0128 decode.acc_seg: 99.5472 aux.loss_ce: 0.0103 aux.acc_seg: 98.9493 +04/17 06:40:45 - mmengine - INFO - Iter(train) [ 27950/160000] lr: 8.4290e-03 eta: 20:17:13 time: 0.5528 data_time: 0.0062 memory: 7635 loss: 0.0213 decode.loss_ce: 0.0119 decode.acc_seg: 99.5675 aux.loss_ce: 0.0094 aux.acc_seg: 99.0839 +04/17 06:41:13 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 06:41:13 - mmengine - INFO - Iter(train) [ 28000/160000] lr: 8.4262e-03 eta: 20:16:46 time: 0.5543 data_time: 0.0064 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0100 decode.acc_seg: 99.6709 aux.loss_ce: 0.0083 aux.acc_seg: 99.2907 +04/17 06:41:41 - mmengine - INFO - Iter(train) [ 28050/160000] lr: 8.4233e-03 eta: 20:16:18 time: 0.5533 data_time: 0.0057 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0109 decode.acc_seg: 99.4734 aux.loss_ce: 0.0087 aux.acc_seg: 99.1693 +04/17 06:42:08 - mmengine - INFO - Iter(train) [ 28100/160000] lr: 8.4205e-03 eta: 20:15:50 time: 0.5533 data_time: 0.0066 memory: 7635 loss: 0.0201 decode.loss_ce: 0.0112 decode.acc_seg: 99.4831 aux.loss_ce: 0.0089 aux.acc_seg: 99.2082 +04/17 06:42:36 - mmengine - INFO - Iter(train) [ 28150/160000] lr: 8.4177e-03 eta: 20:15:23 time: 0.5535 data_time: 0.0071 memory: 7635 loss: 0.0197 decode.loss_ce: 0.0109 decode.acc_seg: 99.6026 aux.loss_ce: 0.0087 aux.acc_seg: 99.0334 +04/17 06:43:04 - mmengine - INFO - Iter(train) [ 28200/160000] lr: 8.4148e-03 eta: 20:14:55 time: 0.5538 data_time: 0.0064 memory: 7635 loss: 0.0199 decode.loss_ce: 0.0110 decode.acc_seg: 99.6615 aux.loss_ce: 0.0088 aux.acc_seg: 99.2807 +04/17 06:43:31 - mmengine - INFO - Iter(train) [ 28250/160000] lr: 8.4120e-03 eta: 20:14:28 time: 0.5542 data_time: 0.0056 memory: 7635 loss: 0.0208 decode.loss_ce: 0.0118 decode.acc_seg: 99.4784 aux.loss_ce: 0.0091 aux.acc_seg: 98.9578 +04/17 06:43:59 - mmengine - INFO - Iter(train) [ 28300/160000] lr: 8.4092e-03 eta: 20:14:00 time: 0.5630 data_time: 0.0059 memory: 7635 loss: 0.0218 decode.loss_ce: 0.0122 decode.acc_seg: 99.3261 aux.loss_ce: 0.0095 aux.acc_seg: 98.7772 +04/17 06:44:27 - mmengine - INFO - Iter(train) [ 28350/160000] lr: 8.4063e-03 eta: 20:13:33 time: 0.5533 data_time: 0.0063 memory: 7635 loss: 0.0226 decode.loss_ce: 0.0128 decode.acc_seg: 99.5937 aux.loss_ce: 0.0098 aux.acc_seg: 99.0811 +04/17 06:44:54 - mmengine - INFO - Iter(train) [ 28400/160000] lr: 8.4035e-03 eta: 20:13:05 time: 0.5534 data_time: 0.0063 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0105 decode.acc_seg: 99.6407 aux.loss_ce: 0.0086 aux.acc_seg: 99.3244 +04/17 06:45:22 - mmengine - INFO - Iter(train) [ 28450/160000] lr: 8.4006e-03 eta: 20:12:38 time: 0.5531 data_time: 0.0057 memory: 7635 loss: 0.0206 decode.loss_ce: 0.0114 decode.acc_seg: 99.5809 aux.loss_ce: 0.0092 aux.acc_seg: 99.1844 +04/17 06:45:50 - mmengine - INFO - Iter(train) [ 28500/160000] lr: 8.3978e-03 eta: 20:12:10 time: 0.5526 data_time: 0.0062 memory: 7635 loss: 0.0198 decode.loss_ce: 0.0111 decode.acc_seg: 99.4272 aux.loss_ce: 0.0088 aux.acc_seg: 98.8027 +04/17 06:46:17 - mmengine - INFO - Iter(train) [ 28550/160000] lr: 8.3950e-03 eta: 20:11:43 time: 0.5533 data_time: 0.0069 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0105 decode.acc_seg: 99.6221 aux.loss_ce: 0.0087 aux.acc_seg: 99.0142 +04/17 06:46:45 - mmengine - INFO - Iter(train) [ 28600/160000] lr: 8.3921e-03 eta: 20:11:15 time: 0.5537 data_time: 0.0064 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0098 decode.acc_seg: 99.6454 aux.loss_ce: 0.0081 aux.acc_seg: 99.0096 +04/17 06:47:13 - mmengine - INFO - Iter(train) [ 28650/160000] lr: 8.3893e-03 eta: 20:10:47 time: 0.5518 data_time: 0.0060 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0104 decode.acc_seg: 99.5618 aux.loss_ce: 0.0084 aux.acc_seg: 99.2650 +04/17 06:47:40 - mmengine - INFO - Iter(train) [ 28700/160000] lr: 8.3864e-03 eta: 20:10:20 time: 0.5548 data_time: 0.0058 memory: 7635 loss: 0.0222 decode.loss_ce: 0.0126 decode.acc_seg: 99.3545 aux.loss_ce: 0.0096 aux.acc_seg: 98.7711 +04/17 06:48:08 - mmengine - INFO - Iter(train) [ 28750/160000] lr: 8.3836e-03 eta: 20:09:52 time: 0.5525 data_time: 0.0063 memory: 7635 loss: 0.0213 decode.loss_ce: 0.0120 decode.acc_seg: 99.5036 aux.loss_ce: 0.0092 aux.acc_seg: 99.0538 +04/17 06:48:36 - mmengine - INFO - Iter(train) [ 28800/160000] lr: 8.3808e-03 eta: 20:09:25 time: 0.5541 data_time: 0.0058 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0101 decode.acc_seg: 99.5505 aux.loss_ce: 0.0087 aux.acc_seg: 98.8218 +04/17 06:49:04 - mmengine - INFO - Iter(train) [ 28850/160000] lr: 8.3779e-03 eta: 20:08:57 time: 0.5548 data_time: 0.0064 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0098 decode.acc_seg: 99.6283 aux.loss_ce: 0.0080 aux.acc_seg: 99.1926 +04/17 06:49:31 - mmengine - INFO - Iter(train) [ 28900/160000] lr: 8.3751e-03 eta: 20:08:30 time: 0.5540 data_time: 0.0060 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0101 decode.acc_seg: 99.4685 aux.loss_ce: 0.0083 aux.acc_seg: 99.0213 +04/17 06:49:59 - mmengine - INFO - Iter(train) [ 28950/160000] lr: 8.3722e-03 eta: 20:08:03 time: 0.5536 data_time: 0.0064 memory: 7635 loss: 0.0202 decode.loss_ce: 0.0112 decode.acc_seg: 99.6550 aux.loss_ce: 0.0090 aux.acc_seg: 99.1830 +04/17 06:50:27 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 06:50:27 - mmengine - INFO - Iter(train) [ 29000/160000] lr: 8.3694e-03 eta: 20:07:35 time: 0.5523 data_time: 0.0064 memory: 7635 loss: 0.0214 decode.loss_ce: 0.0122 decode.acc_seg: 99.6282 aux.loss_ce: 0.0092 aux.acc_seg: 99.1642 +04/17 06:50:54 - mmengine - INFO - Iter(train) [ 29050/160000] lr: 8.3666e-03 eta: 20:07:07 time: 0.5521 data_time: 0.0058 memory: 7635 loss: 0.0219 decode.loss_ce: 0.0120 decode.acc_seg: 99.3945 aux.loss_ce: 0.0099 aux.acc_seg: 98.7250 +04/17 06:51:22 - mmengine - INFO - Iter(train) [ 29100/160000] lr: 8.3637e-03 eta: 20:06:40 time: 0.5542 data_time: 0.0065 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0116 decode.acc_seg: 99.4136 aux.loss_ce: 0.0091 aux.acc_seg: 98.8048 +04/17 06:51:50 - mmengine - INFO - Iter(train) [ 29150/160000] lr: 8.3609e-03 eta: 20:06:12 time: 0.5545 data_time: 0.0061 memory: 7635 loss: 0.0198 decode.loss_ce: 0.0108 decode.acc_seg: 99.6570 aux.loss_ce: 0.0089 aux.acc_seg: 99.1039 +04/17 06:52:17 - mmengine - INFO - Iter(train) [ 29200/160000] lr: 8.3580e-03 eta: 20:05:45 time: 0.5534 data_time: 0.0057 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0103 decode.acc_seg: 99.5222 aux.loss_ce: 0.0080 aux.acc_seg: 99.0696 +04/17 06:52:45 - mmengine - INFO - Iter(train) [ 29250/160000] lr: 8.3552e-03 eta: 20:05:17 time: 0.5525 data_time: 0.0063 memory: 7635 loss: 0.0203 decode.loss_ce: 0.0113 decode.acc_seg: 99.5973 aux.loss_ce: 0.0090 aux.acc_seg: 99.3248 +04/17 06:53:13 - mmengine - INFO - Iter(train) [ 29300/160000] lr: 8.3524e-03 eta: 20:04:50 time: 0.5530 data_time: 0.0065 memory: 7635 loss: 0.0220 decode.loss_ce: 0.0125 decode.acc_seg: 99.4563 aux.loss_ce: 0.0094 aux.acc_seg: 98.9904 +04/17 06:53:40 - mmengine - INFO - Iter(train) [ 29350/160000] lr: 8.3495e-03 eta: 20:04:22 time: 0.5541 data_time: 0.0065 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0107 decode.acc_seg: 99.5584 aux.loss_ce: 0.0088 aux.acc_seg: 98.9595 +04/17 06:54:08 - mmengine - INFO - Iter(train) [ 29400/160000] lr: 8.3467e-03 eta: 20:03:55 time: 0.5521 data_time: 0.0064 memory: 7635 loss: 0.0206 decode.loss_ce: 0.0111 decode.acc_seg: 99.5743 aux.loss_ce: 0.0095 aux.acc_seg: 99.1711 +04/17 06:54:36 - mmengine - INFO - Iter(train) [ 29450/160000] lr: 8.3438e-03 eta: 20:03:27 time: 0.5526 data_time: 0.0074 memory: 7635 loss: 0.0201 decode.loss_ce: 0.0108 decode.acc_seg: 99.5565 aux.loss_ce: 0.0092 aux.acc_seg: 99.0637 +04/17 06:55:03 - mmengine - INFO - Iter(train) [ 29500/160000] lr: 8.3410e-03 eta: 20:02:59 time: 0.5532 data_time: 0.0066 memory: 7635 loss: 0.0210 decode.loss_ce: 0.0116 decode.acc_seg: 99.6478 aux.loss_ce: 0.0094 aux.acc_seg: 99.1285 +04/17 06:55:31 - mmengine - INFO - Iter(train) [ 29550/160000] lr: 8.3381e-03 eta: 20:02:32 time: 0.5534 data_time: 0.0063 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0102 decode.acc_seg: 99.7054 aux.loss_ce: 0.0083 aux.acc_seg: 99.2275 +04/17 06:55:59 - mmengine - INFO - Iter(train) [ 29600/160000] lr: 8.3353e-03 eta: 20:02:05 time: 0.5619 data_time: 0.0062 memory: 7635 loss: 0.0198 decode.loss_ce: 0.0108 decode.acc_seg: 99.5362 aux.loss_ce: 0.0090 aux.acc_seg: 99.1015 +04/17 06:56:27 - mmengine - INFO - Iter(train) [ 29650/160000] lr: 8.3325e-03 eta: 20:01:37 time: 0.5533 data_time: 0.0063 memory: 7635 loss: 0.0212 decode.loss_ce: 0.0120 decode.acc_seg: 99.3860 aux.loss_ce: 0.0092 aux.acc_seg: 98.9589 +04/17 06:56:54 - mmengine - INFO - Iter(train) [ 29700/160000] lr: 8.3296e-03 eta: 20:01:10 time: 0.5535 data_time: 0.0058 memory: 7635 loss: 0.0193 decode.loss_ce: 0.0104 decode.acc_seg: 99.6583 aux.loss_ce: 0.0088 aux.acc_seg: 99.1363 +04/17 06:57:22 - mmengine - INFO - Iter(train) [ 29750/160000] lr: 8.3268e-03 eta: 20:00:42 time: 0.5533 data_time: 0.0061 memory: 7635 loss: 0.0218 decode.loss_ce: 0.0123 decode.acc_seg: 99.5695 aux.loss_ce: 0.0094 aux.acc_seg: 98.8542 +04/17 06:57:50 - mmengine - INFO - Iter(train) [ 29800/160000] lr: 8.3239e-03 eta: 20:00:14 time: 0.5519 data_time: 0.0063 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0113 decode.acc_seg: 99.6489 aux.loss_ce: 0.0094 aux.acc_seg: 99.1776 +04/17 06:58:17 - mmengine - INFO - Iter(train) [ 29850/160000] lr: 8.3211e-03 eta: 19:59:47 time: 0.5542 data_time: 0.0071 memory: 7635 loss: 0.0205 decode.loss_ce: 0.0114 decode.acc_seg: 99.2705 aux.loss_ce: 0.0092 aux.acc_seg: 98.8076 +04/17 06:58:45 - mmengine - INFO - Iter(train) [ 29900/160000] lr: 8.3182e-03 eta: 19:59:19 time: 0.5512 data_time: 0.0062 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0100 decode.acc_seg: 99.5226 aux.loss_ce: 0.0088 aux.acc_seg: 98.6992 +04/17 06:59:12 - mmengine - INFO - Iter(train) [ 29950/160000] lr: 8.3154e-03 eta: 19:58:51 time: 0.5522 data_time: 0.0062 memory: 7635 loss: 0.0214 decode.loss_ce: 0.0119 decode.acc_seg: 99.4985 aux.loss_ce: 0.0095 aux.acc_seg: 98.7783 +04/17 06:59:40 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 06:59:40 - mmengine - INFO - Iter(train) [ 30000/160000] lr: 8.3126e-03 eta: 19:58:24 time: 0.5540 data_time: 0.0062 memory: 7635 loss: 0.0194 decode.loss_ce: 0.0106 decode.acc_seg: 99.5883 aux.loss_ce: 0.0087 aux.acc_seg: 99.2011 +04/17 06:59:40 - mmengine - INFO - Saving checkpoint at 30000 iterations +04/17 06:59:44 - mmengine - INFO - Iter(val) [ 50/355] eta: 0:00:14 time: 0.0467 data_time: 0.0015 memory: 1657 +04/17 06:59:46 - mmengine - INFO - Iter(val) [100/355] eta: 0:00:11 time: 0.0464 data_time: 0.0014 memory: 1657 +04/17 06:59:49 - mmengine - INFO - Iter(val) [150/355] eta: 0:00:09 time: 0.0462 data_time: 0.0013 memory: 1657 +04/17 06:59:51 - mmengine - INFO - Iter(val) [200/355] eta: 0:00:07 time: 0.0463 data_time: 0.0015 memory: 1657 +04/17 06:59:53 - mmengine - INFO - Iter(val) [250/355] eta: 0:00:04 time: 0.0461 data_time: 0.0014 memory: 1657 +04/17 06:59:56 - mmengine - INFO - Iter(val) [300/355] eta: 0:00:02 time: 0.0466 data_time: 0.0015 memory: 1657 +04/17 06:59:58 - mmengine - INFO - Iter(val) [350/355] eta: 0:00:00 time: 0.0458 data_time: 0.0013 memory: 1657 +04/17 06:59:59 - mmengine - INFO - per class results: +04/17 06:59:59 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 99.14 | 99.64 | 99.57 | 99.49 | 99.64 | +| contrast | 80.93 | 87.83 | 89.46 | 91.15 | 87.83 | ++------------+-------+-------+--------+-----------+--------+ +04/17 06:59:59 - mmengine - INFO - Iter(val) [355/355] aAcc: 99.1700 mIoU: 90.0300 mAcc: 93.7400 mFscore: 94.5100 mPrecision: 95.3200 mRecall: 93.7400 data_time: 0.0015 time: 0.0465 +04/17 07:00:26 - mmengine - INFO - Iter(train) [ 30050/160000] lr: 8.3097e-03 eta: 19:57:58 time: 0.5519 data_time: 0.0058 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0104 decode.acc_seg: 99.5041 aux.loss_ce: 0.0085 aux.acc_seg: 99.1146 +04/17 07:00:54 - mmengine - INFO - Iter(train) [ 30100/160000] lr: 8.3069e-03 eta: 19:57:30 time: 0.5518 data_time: 0.0061 memory: 7635 loss: 0.0197 decode.loss_ce: 0.0109 decode.acc_seg: 99.6054 aux.loss_ce: 0.0088 aux.acc_seg: 99.0529 +04/17 07:01:21 - mmengine - INFO - Iter(train) [ 30150/160000] lr: 8.3040e-03 eta: 19:57:02 time: 0.5528 data_time: 0.0059 memory: 7635 loss: 0.0190 decode.loss_ce: 0.0104 decode.acc_seg: 99.6139 aux.loss_ce: 0.0085 aux.acc_seg: 99.1058 +04/17 07:01:49 - mmengine - INFO - Iter(train) [ 30200/160000] lr: 8.3012e-03 eta: 19:56:34 time: 0.5533 data_time: 0.0064 memory: 7635 loss: 0.0212 decode.loss_ce: 0.0116 decode.acc_seg: 99.4350 aux.loss_ce: 0.0096 aux.acc_seg: 98.9349 +04/17 07:02:17 - mmengine - INFO - Iter(train) [ 30250/160000] lr: 8.2983e-03 eta: 19:56:06 time: 0.5543 data_time: 0.0066 memory: 7635 loss: 0.0208 decode.loss_ce: 0.0114 decode.acc_seg: 99.5555 aux.loss_ce: 0.0094 aux.acc_seg: 99.0211 +04/17 07:02:44 - mmengine - INFO - Iter(train) [ 30300/160000] lr: 8.2955e-03 eta: 19:55:39 time: 0.5531 data_time: 0.0065 memory: 7635 loss: 0.0210 decode.loss_ce: 0.0118 decode.acc_seg: 99.6118 aux.loss_ce: 0.0093 aux.acc_seg: 99.1990 +04/17 07:03:12 - mmengine - INFO - Iter(train) [ 30350/160000] lr: 8.2927e-03 eta: 19:55:11 time: 0.5535 data_time: 0.0064 memory: 7635 loss: 0.0203 decode.loss_ce: 0.0114 decode.acc_seg: 99.6575 aux.loss_ce: 0.0089 aux.acc_seg: 99.1538 +04/17 07:03:40 - mmengine - INFO - Iter(train) [ 30400/160000] lr: 8.2898e-03 eta: 19:54:44 time: 0.5511 data_time: 0.0057 memory: 7635 loss: 0.0198 decode.loss_ce: 0.0109 decode.acc_seg: 99.5982 aux.loss_ce: 0.0089 aux.acc_seg: 99.0856 +04/17 07:04:07 - mmengine - INFO - Iter(train) [ 30450/160000] lr: 8.2870e-03 eta: 19:54:16 time: 0.5541 data_time: 0.0065 memory: 7635 loss: 0.0193 decode.loss_ce: 0.0106 decode.acc_seg: 99.3167 aux.loss_ce: 0.0087 aux.acc_seg: 98.8218 +04/17 07:04:35 - mmengine - INFO - Iter(train) [ 30500/160000] lr: 8.2841e-03 eta: 19:53:48 time: 0.5524 data_time: 0.0063 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0105 decode.acc_seg: 99.5469 aux.loss_ce: 0.0084 aux.acc_seg: 99.2051 +04/17 07:05:03 - mmengine - INFO - Iter(train) [ 30550/160000] lr: 8.2813e-03 eta: 19:53:20 time: 0.5527 data_time: 0.0061 memory: 7635 loss: 0.0203 decode.loss_ce: 0.0114 decode.acc_seg: 99.5904 aux.loss_ce: 0.0089 aux.acc_seg: 99.2641 +04/17 07:05:30 - mmengine - INFO - Iter(train) [ 30600/160000] lr: 8.2784e-03 eta: 19:52:53 time: 0.5529 data_time: 0.0063 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0120 decode.acc_seg: 99.6280 aux.loss_ce: 0.0089 aux.acc_seg: 99.2269 +04/17 07:05:58 - mmengine - INFO - Iter(train) [ 30650/160000] lr: 8.2756e-03 eta: 19:52:25 time: 0.5528 data_time: 0.0073 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0105 decode.acc_seg: 99.6181 aux.loss_ce: 0.0086 aux.acc_seg: 99.2119 +04/17 07:06:26 - mmengine - INFO - Iter(train) [ 30700/160000] lr: 8.2728e-03 eta: 19:51:57 time: 0.5528 data_time: 0.0062 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0103 decode.acc_seg: 99.6109 aux.loss_ce: 0.0086 aux.acc_seg: 99.1459 +04/17 07:06:53 - mmengine - INFO - Iter(train) [ 30750/160000] lr: 8.2699e-03 eta: 19:51:30 time: 0.5534 data_time: 0.0059 memory: 7635 loss: 0.0190 decode.loss_ce: 0.0105 decode.acc_seg: 99.6828 aux.loss_ce: 0.0085 aux.acc_seg: 99.3529 +04/17 07:07:21 - mmengine - INFO - Iter(train) [ 30800/160000] lr: 8.2671e-03 eta: 19:51:02 time: 0.5528 data_time: 0.0066 memory: 7635 loss: 0.0197 decode.loss_ce: 0.0110 decode.acc_seg: 99.5640 aux.loss_ce: 0.0087 aux.acc_seg: 99.0981 +04/17 07:07:49 - mmengine - INFO - Iter(train) [ 30850/160000] lr: 8.2642e-03 eta: 19:50:35 time: 0.5522 data_time: 0.0063 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0117 decode.acc_seg: 99.4967 aux.loss_ce: 0.0092 aux.acc_seg: 99.1055 +04/17 07:08:16 - mmengine - INFO - Iter(train) [ 30900/160000] lr: 8.2614e-03 eta: 19:50:07 time: 0.5533 data_time: 0.0061 memory: 7635 loss: 0.0202 decode.loss_ce: 0.0113 decode.acc_seg: 99.5605 aux.loss_ce: 0.0089 aux.acc_seg: 99.1894 +04/17 07:08:44 - mmengine - INFO - Iter(train) [ 30950/160000] lr: 8.2585e-03 eta: 19:49:39 time: 0.5524 data_time: 0.0056 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0119 decode.acc_seg: 99.4685 aux.loss_ce: 0.0088 aux.acc_seg: 99.0562 +04/17 07:09:12 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240417_022222 +04/17 07:09:12 - mmengine - INFO - Iter(train) [ 31000/160000] lr: 8.2557e-03 eta: 19:49:11 time: 0.5543 data_time: 0.0067 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0093 decode.acc_seg: 99.5527 aux.loss_ce: 0.0081 aux.acc_seg: 99.0878 +04/17 07:09:39 - mmengine - INFO - Iter(train) [ 31050/160000] lr: 8.2528e-03 eta: 19:48:44 time: 0.5524 data_time: 0.0064 memory: 7635 loss: 0.0224 decode.loss_ce: 0.0126 decode.acc_seg: 99.5718 aux.loss_ce: 0.0097 aux.acc_seg: 98.9946 +04/17 07:10:07 - mmengine - INFO - Iter(train) [ 31100/160000] lr: 8.2500e-03 eta: 19:48:16 time: 0.5521 data_time: 0.0060 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0112 decode.acc_seg: 99.7195 aux.loss_ce: 0.0096 aux.acc_seg: 99.1817 +WARNING:torch.distributed.elastic.agent.server.api:Received 2 death signal, shutting down workers +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 1259 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 1260 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 1261 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 1262 closing signal SIGINT +Traceback (most recent call last): +Traceback (most recent call last): +Traceback (most recent call last): + File "tools/train.py", line 104, in + File "tools/train.py", line 104, in +Traceback (most recent call last): + File "tools/train.py", line 104, in + File "tools/train.py", line 104, in + main()main()main()main() + + + + File "tools/train.py", line 100, in main + File "tools/train.py", line 100, in main + File "tools/train.py", line 100, in main + File "tools/train.py", line 100, in main + runner.train() + runner.train()runner.train() File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train +runner.train() + + + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + model = self.train_loop.run() # type: ignoremodel = self.train_loop.run() # type: ignore + + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + model = self.train_loop.run() # type: ignoremodel = self.train_loop.run() # type: ignore + + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + self.run_iter(data_batch) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + self.run_iter(data_batch) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + self.run_iter(data_batch)self.run_iter(data_batch) + + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + outputs = self.runner.model.train_step( + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 121, in train_step + outputs = self.runner.model.train_step( + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 121, in train_step + outputs = self.runner.model.train_step(outputs = self.runner.model.train_step( + + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 121, in train_step + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 121, in train_step +losses = self._run_forward(data, mode='loss') +losses = self._run_forward(data, mode='loss') File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 161, in _run_forward + + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 161, in _run_forward + losses = self._run_forward(data, mode='loss') + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 161, in _run_forward + losses = self._run_forward(data, mode='loss') + results = self(**data, mode=mode) File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 161, in _run_forward +results = self(**data, mode=mode) + + File "/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl + File "/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl + results = self(**data, mode=mode) + File "/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl + results = self(**data, mode=mode) + File "/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl + return forward_call(*input, **kwargs)return forward_call(*input, **kwargs) + + File "/opt/conda/lib/python3.8/site-packages/torch/nn/parallel/distributed.py", line 963, in forward + File "/opt/conda/lib/python3.8/site-packages/torch/nn/parallel/distributed.py", line 963, in forward +return forward_call(*input, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/nn/parallel/distributed.py", line 963, in forward +return forward_call(*input, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/nn/parallel/distributed.py", line 963, in forward + output = self.module(*inputs[0], **kwargs[0])output = self.module(*inputs[0], **kwargs[0]) + +output = self.module(*inputs[0], **kwargs[0]) File "/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl + File "/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl + + File "/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl +output = self.module(*inputs[0], **kwargs[0]) + File "/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl + return forward_call(*input, **kwargs) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/base.py", line 94, in forward +return forward_call(*input, **kwargs) +return forward_call(*input, **kwargs) File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/base.py", line 94, in forward + + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/base.py", line 94, in forward +return forward_call(*input, **kwargs) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/base.py", line 94, in forward + return self.loss(inputs, data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/encoder_decoder.py", line 178, in loss + return self.loss(inputs, data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/encoder_decoder.py", line 178, in loss +return self.loss(inputs, data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/encoder_decoder.py", line 178, in loss +return self.loss(inputs, data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/encoder_decoder.py", line 178, in loss + loss_decode = self._decode_head_forward_train(x, data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/encoder_decoder.py", line 139, in _decode_head_forward_train + loss_decode = self._decode_head_forward_train(x, data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/encoder_decoder.py", line 139, in _decode_head_forward_train + loss_decode = self._decode_head_forward_train(x, data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/encoder_decoder.py", line 139, in _decode_head_forward_train +loss_decode = self._decode_head_forward_train(x, data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/encoder_decoder.py", line 139, in _decode_head_forward_train + loss_decode = self.decode_head.loss(inputs, data_samples, + File "/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py", line 262, in loss +loss_decode = self.decode_head.loss(inputs, data_samples, + File "/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py", line 262, in loss + loss_decode = self.decode_head.loss(inputs, data_samples, + File "/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py", line 262, in loss + loss_decode = self.decode_head.loss(inputs, data_samples, + File "/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py", line 262, in loss + losses = self.loss_by_feat(seg_logits, batch_data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py", line 336, in loss_by_feat + losses = self.loss_by_feat(seg_logits, batch_data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py", line 336, in loss_by_feat + losses = self.loss_by_feat(seg_logits, batch_data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py", line 336, in loss_by_feat + losses = self.loss_by_feat(seg_logits, batch_data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py", line 336, in loss_by_feat + loss['acc_seg'] = accuracy( + File "/workspaces/mmsegmentation-1/mmseg/models/losses/accuracy.py", line 49, in accuracy + loss['acc_seg'] = accuracy( + File "/workspaces/mmsegmentation-1/mmseg/models/losses/accuracy.py", line 49, in accuracy +loss['acc_seg'] = accuracy( + File "/workspaces/mmsegmentation-1/mmseg/models/losses/accuracy.py", line 49, in accuracy + loss['acc_seg'] = accuracy(correct = correct[:, target != ignore_index] + + File "/workspaces/mmsegmentation-1/mmseg/models/losses/accuracy.py", line 49, in accuracy + KeyboardInterruptcorrect = correct[:, target != ignore_index] + +correct = correct[:, target != ignore_index]KeyboardInterrupt + +KeyboardInterrupt correct = correct[:, target != ignore_index] + +KeyboardInterrupt +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 1259 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 1260 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 1261 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 1262 closing signal SIGTERM +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 709, in run + result = self._invoke_run(role) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 850, in _invoke_run + time.sleep(monitor_interval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 1218 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 716, in run + self._shutdown(e.sigval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 1218 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 236, in launch_agent + result = agent.run() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/metrics/api.py", line 125, in wrapper + result = f(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 721, in run + self._shutdown() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 1218 got signal: 2 +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/17 08:31:33 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 1692964102 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 1692964102 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/17 08:31:33 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +fp16 = dict(loss_scale='dynamic') +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = None +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=768, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + num_convs=1, + type='FCNHead'), + backbone=dict( + act_cfg=dict(type='GELU'), + attn_drop_rate=0.0, + drop_path_rate=0.1, + embed_dims=768, + img_size=( + 512, + 512, + ), + in_channels=3, + init_values=1.0, + mlp_ratio=4, + norm_cfg=dict(eps=1e-06, type='LN'), + norm_eval=False, + num_heads=12, + num_layers=12, + out_indices=[ + 3, + 5, + 7, + 11, + ], + patch_size=16, + type='MAE'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=768, + dropout_ratio=0.1, + in_channels=[ + 768, + 768, + 768, + 768, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + neck=dict( + embed_dim=768, rescales=[ + 4, + 2, + 1, + 0.5, + ], type='Feature2Pyramid'), + pretrained='/workspaces/mmsegmentation-1/converted_model.pth', + test_cfg=dict(crop_size=( + 512, + 512, + ), mode='slide', stride=( + 341, + 341, + )), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + constructor='LayerDecayOptimizerConstructor', + optimizer=dict( + betas=( + 0.9, + 0.999, + ), lr=0.0001, type='AdamW', weight_decay=0.05), + paramwise_cfg=dict(layer_decay_rate=0.65, num_layers=12), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, by_epoch=False, end=1500, start_factor=1e-06, + type='LinearLR'), + dict( + begin=1500, + by_epoch=False, + end=160000, + eta_min=0.0, + power=1.0, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=2, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = '/workspaces/mmsegmentation-1/work_dirs/MAE4000' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/17 08:31:36 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65}set param backbone.pos_embed as id 0 + +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1set param backbone.cls_token as id 0 + +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.pos_embed as id 0 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1set param backbone.layers.0.ln1.bias as id 1 + +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.0.attn.proj.bias as id 1set param backbone.layers.1.attn.qkv.bias as id 2 + +set param backbone.layers.1.attn.proj.weight as id 2set param backbone.layers.0.ln2.weight as id 1 + +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1set param backbone.layers.1.ffn.layers.0.0.weight as id 2 + +set param backbone.layers.0.ffn.layers.0.0.bias as id 1set param backbone.layers.1.ffn.layers.0.0.bias as id 2 + +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.1.gamma_2 as id 2set param backbone.layers.2.attn.relative_position_bias_table as id 3 + +set param backbone.layers.1.ln1.weight as id 2set param backbone.layers.2.attn.qkv.weight as id 3 + +set param backbone.layers.1.ln1.bias as id 2set param backbone.layers.2.attn.qkv.bias as id 3 + +set param backbone.layers.1.attn.relative_position_bias_table as id 2set param backbone.layers.2.attn.proj.weight as id 3 + +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.1.attn.proj.bias as id 2set param backbone.layers.2.ffn.layers.0.0.bias as id 3 + +set param backbone.layers.2.ffn.layers.1.weight as id 3set param backbone.layers.1.ln2.weight as id 2 + +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2set param backbone.layers.3.gamma_2 as id 4 + +set param backbone.layers.1.ffn.layers.0.0.bias as id 2set param backbone.layers.3.ln1.weight as id 4 + +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.2.ln1.bias as id 3set param backbone.layers.3.ln2.bias as id 4 + +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.2.attn.qkv.weight as id 3set param backbone.layers.3.ffn.layers.0.0.bias as id 4 + +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3set param backbone.layers.4.gamma_1 as id 5 + +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.4.ln1.weight as id 5set param backbone.layers.2.ln2.bias as id 3 + +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.2.ffn.layers.1.bias as id 3set param backbone.layers.4.attn.proj.bias as id 5 + +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5set param backbone.layers.3.gamma_2 as id 4 + +set param backbone.layers.4.ffn.layers.0.0.bias as id 5set param backbone.layers.3.ln1.weight as id 4 + +set param backbone.layers.3.ln1.bias as id 4set param backbone.layers.4.ffn.layers.1.weight as id 5 + +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.5.gamma_1 as id 6set param backbone.layers.3.attn.qkv.weight as id 4 + +set param backbone.layers.5.gamma_2 as id 6set param backbone.layers.3.attn.qkv.bias as id 4 + +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.3.attn.proj.weight as id 4set param backbone.layers.5.ln1.bias as id 6 + +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.3.ln2.weight as id 4set param backbone.layers.5.attn.qkv.weight as id 6 + +set param backbone.layers.3.ln2.bias as id 4set param backbone.layers.5.attn.qkv.bias as id 6 + +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4set param backbone.layers.5.attn.proj.bias as id 6 + +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4set param backbone.layers.5.ffn.layers.0.0.weight as id 6 + +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6set param backbone.layers.4.gamma_1 as id 5 + +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.4.attn.relative_position_bias_table as id 5set param backbone.layers.6.ln1.bias as id 7 + +set param backbone.layers.6.attn.relative_position_bias_table as id 7set param backbone.layers.4.attn.qkv.weight as id 5 + +set param backbone.layers.6.attn.qkv.weight as id 7set param backbone.layers.4.attn.qkv.bias as id 5 + +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.4.attn.proj.bias as id 5set param backbone.layers.6.attn.proj.bias as id 7 + +set param backbone.layers.6.ln2.weight as id 7set param backbone.layers.4.ln2.weight as id 5 + +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7set param backbone.layers.5.ln1.bias as id 6 + +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.5.attn.qkv.weight as id 6set param backbone.layers.6.ffn.layers.1.bias as id 7 + +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.5.ln2.weight as id 6set param backbone.layers.7.ln1.bias as id 8 + +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8set param backbone.layers.5.ffn.layers.1.weight as id 6 + +set param backbone.layers.5.ffn.layers.1.bias as id 6set param backbone.layers.7.ln2.weight as id 8 + +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.6.gamma_2 as id 7set param backbone.layers.7.ffn.layers.0.0.bias as id 8 + +set param backbone.layers.6.ln1.weight as id 7set param backbone.layers.7.ffn.layers.1.weight as id 8 + +set param backbone.layers.6.ln1.bias as id 7set param backbone.layers.7.ffn.layers.1.bias as id 8 + +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.6.attn.qkv.weight as id 7set param backbone.layers.8.gamma_2 as id 9 + +set param backbone.layers.6.attn.qkv.bias as id 7set param backbone.layers.8.ln1.weight as id 9 + +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9set param backbone.layers.6.ln2.weight as id 7 + +set param backbone.layers.6.ln2.bias as id 7set param backbone.layers.8.attn.proj.weight as id 9 + +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.6.ffn.layers.1.weight as id 7set param backbone.layers.9.ffn.layers.0.0.bias as id 10 + +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.7.gamma_1 as id 8set param backbone.layers.10.gamma_1 as id 11 + +set param backbone.layers.10.gamma_2 as id 11set param backbone.layers.7.gamma_2 as id 8 + +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.7.ln1.weight as id 8set param backbone.layers.10.ln1.bias as id 11 + +set param backbone.layers.7.ln1.bias as id 8set param backbone.layers.10.attn.relative_position_bias_table as id 11 + +set param backbone.layers.10.attn.qkv.weight as id 11set param backbone.layers.7.attn.relative_position_bias_table as id 8 + +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.7.attn.qkv.weight as id 8set param backbone.layers.10.attn.proj.weight as id 11 + +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.7.attn.proj.weight as id 8set param backbone.layers.10.ln2.bias as id 11 + +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.7.ln2.weight as id 8set param backbone.layers.10.ffn.layers.0.0.bias as id 11 + +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.8.gamma_1 as id 9set param backbone.layers.11.attn.qkv.bias as id 12 + +set param backbone.layers.8.gamma_2 as id 9set param backbone.layers.11.attn.proj.weight as id 12 + +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.11.ln2.weight as id 12set param backbone.layers.8.ln1.bias as id 9 + +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.8.attn.qkv.weight as id 9set param backbone.layers.11.ffn.layers.0.0.bias as id 12 + +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12set param backbone.layers.8.attn.proj.weight as id 9 + +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9set param neck.upsample_4x.0.weight as id 13 + +set param neck.upsample_4x.0.bias as id 13 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param neck.upsample_4x.1.weight as id 13set param backbone.layers.8.ffn.layers.0.0.bias as id 9 + +set param neck.upsample_4x.1.bias as id 13 +set param backbone.layers.8.ffn.layers.1.weight as id 9set param neck.upsample_4x.3.weight as id 13 + +set param neck.upsample_4x.3.bias as id 13set param backbone.layers.8.ffn.layers.1.bias as id 9 + +set param neck.upsample_2x.0.weight as id 13 +set param backbone.layers.9.gamma_1 as id 10set param neck.upsample_2x.0.bias as id 13 + +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10set param decode_head.conv_seg.weight as id 13 + +set param decode_head.conv_seg.bias as id 13 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param backbone.layers.9.ln2.weight as id 10set param decode_head.psp_modules.0.1.bn.bias as id 13 + +set param backbone.layers.9.ln2.bias as id 10 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10set param decode_head.psp_modules.1.1.bn.bias as id 13 + +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param backbone.layers.10.gamma_1 as id 11 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param backbone.layers.10.gamma_2 as id 11 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13set param backbone.layers.10.ln1.weight as id 11 + +set param backbone.layers.10.ln1.bias as id 11set param decode_head.bottleneck.conv.weight as id 13 + +set param decode_head.bottleneck.bn.weight as id 13set param backbone.layers.10.attn.relative_position_bias_table as id 11 + +set param decode_head.bottleneck.bn.bias as id 13 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param decode_head.lateral_convs.0.conv.weight as id 13set param backbone.layers.10.attn.qkv.bias as id 11 + +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param backbone.layers.10.attn.proj.weight as id 11set param decode_head.lateral_convs.0.bn.bias as id 13 + +set param backbone.layers.10.attn.proj.bias as id 11 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13set param backbone.layers.10.ln2.weight as id 11 + +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param backbone.layers.10.ln2.bias as id 11 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param backbone.layers.10.ffn.layers.1.weight as id 11set param decode_head.fpn_convs.0.bn.weight as id 13 + +set param decode_head.fpn_convs.0.bn.bias as id 13set param backbone.layers.10.ffn.layers.1.bias as id 11 + +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13set param backbone.layers.11.gamma_1 as id 12 + +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param backbone.layers.11.gamma_2 as id 12 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param backbone.layers.11.ln1.weight as id 12 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param backbone.layers.11.ln1.bias as id 12set param decode_head.fpn_convs.2.bn.bias as id 13 + +set param backbone.layers.11.attn.relative_position_bias_table as id 12set param decode_head.fpn_bottleneck.conv.weight as id 13 + +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12set param auxiliary_head.conv_seg.weight as id 13 + +set param auxiliary_head.conv_seg.bias as id 13 +set param backbone.layers.11.ln2.weight as id 12 +set param auxiliary_head.convs.0.conv.weight as id 13set param backbone.layers.11.ln2.bias as id 12 + +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +Param groups = { + "layer_0_decay": { + "param_names": [ + "backbone.cls_token", + "backbone.pos_embed", + "backbone.patch_embed.projection.weight" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.05 + }, + "layer_0_no_decay": { + "param_names": [ + "backbone.patch_embed.projection.bias" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.0 + }, + "layer_1_no_decay": { + "param_names": [ + "backbone.layers.0.gamma_1", + "backbone.layers.0.gamma_2", + "backbone.layers.0.ln1.weight", + "backbone.layers.0.ln1.bias", + "backbone.layers.0.attn.qkv.bias", + "backbone.layers.0.attn.proj.bias", + "backbone.layers.0.ln2.weight", + "backbone.layers.0.ln2.bias", + "backbone.layers.0.ffn.layers.0.0.bias", + "backbone.layers.0.ffn.layers.1.bias" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.0 + }, + "layer_1_decay": { + "param_names": [ + "backbone.layers.0.attn.relative_position_bias_table", + "backbone.layers.0.attn.qkv.weight", + "backbone.layers.0.attn.proj.weight", + "backbone.layers.0.ffn.layers.0.0.weight", + "backbone.layers.0.ffn.layers.1.weight" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.05 + }, + "layer_2_no_decay": { + "param_names": [ + "backbone.layers.1.gamma_1", + "backbone.layers.1.gamma_2", + "backbone.layers.1.ln1.weight", + "backbone.layers.1.ln1.bias", + "backbone.layers.1.attn.qkv.bias", + "backbone.layers.1.attn.proj.bias", + "backbone.layers.1.ln2.weight", + "backbone.layers.1.ln2.bias", + "backbone.layers.1.ffn.layers.0.0.bias", + "backbone.layers.1.ffn.layers.1.bias" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.0 + }, + "layer_2_decay": { + "param_names": [ + "backbone.layers.1.attn.relative_position_bias_table", + "backbone.layers.1.attn.qkv.weight", + "backbone.layers.1.attn.proj.weight", + "backbone.layers.1.ffn.layers.0.0.weight", + "backbone.layers.1.ffn.layers.1.weight" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.05 + }, + "layer_3_no_decay": { + "param_names": [ + "backbone.layers.2.gamma_1", + "backbone.layers.2.gamma_2", + "backbone.layers.2.ln1.weight", + "backbone.layers.2.ln1.bias", + "backbone.layers.2.attn.qkv.bias", + "backbone.layers.2.attn.proj.bias", + "backbone.layers.2.ln2.weight", + "backbone.layers.2.ln2.bias", + "backbone.layers.2.ffn.layers.0.0.bias", + "backbone.layers.2.ffn.layers.1.bias" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.0 + }, + "layer_3_decay": { + "param_names": [ + "backbone.layers.2.attn.relative_position_bias_table", + "backbone.layers.2.attn.qkv.weight", + "backbone.layers.2.attn.proj.weight", + "backbone.layers.2.ffn.layers.0.0.weight", + "backbone.layers.2.ffn.layers.1.weight" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.05 + }, + "layer_4_no_decay": { + "param_names": [ + "backbone.layers.3.gamma_1", + "backbone.layers.3.gamma_2", + "backbone.layers.3.ln1.weight", + "backbone.layers.3.ln1.bias", + "backbone.layers.3.attn.qkv.bias", + "backbone.layers.3.attn.proj.bias", + "backbone.layers.3.ln2.weight", + "backbone.layers.3.ln2.bias", + "backbone.layers.3.ffn.layers.0.0.bias", + "backbone.layers.3.ffn.layers.1.bias" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.0 + }, + "layer_4_decay": { + "param_names": [ + "backbone.layers.3.attn.relative_position_bias_table", + "backbone.layers.3.attn.qkv.weight", + "backbone.layers.3.attn.proj.weight", + "backbone.layers.3.ffn.layers.0.0.weight", + "backbone.layers.3.ffn.layers.1.weight" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.05 + }, + "layer_5_no_decay": { + "param_names": [ + "backbone.layers.4.gamma_1", + "backbone.layers.4.gamma_2", + "backbone.layers.4.ln1.weight", + "backbone.layers.4.ln1.bias", + "backbone.layers.4.attn.qkv.bias", + "backbone.layers.4.attn.proj.bias", + "backbone.layers.4.ln2.weight", + "backbone.layers.4.ln2.bias", + "backbone.layers.4.ffn.layers.0.0.bias", + "backbone.layers.4.ffn.layers.1.bias" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.0 + }, + "layer_5_decay": { + "param_names": [ + "backbone.layers.4.attn.relative_position_bias_table", + "backbone.layers.4.attn.qkv.weight", + "backbone.layers.4.attn.proj.weight", + "backbone.layers.4.ffn.layers.0.0.weight", + "backbone.layers.4.ffn.layers.1.weight" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.05 + }, + "layer_6_no_decay": { + "param_names": [ + "backbone.layers.5.gamma_1", + "backbone.layers.5.gamma_2", + "backbone.layers.5.ln1.weight", + "backbone.layers.5.ln1.bias", + "backbone.layers.5.attn.qkv.bias", + "backbone.layers.5.attn.proj.bias", + "backbone.layers.5.ln2.weight", + "backbone.layers.5.ln2.bias", + "backbone.layers.5.ffn.layers.0.0.bias", + "backbone.layers.5.ffn.layers.1.bias" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.0 + }, + "layer_6_decay": { + "param_names": [ + "backbone.layers.5.attn.relative_position_bias_table", + "backbone.layers.5.attn.qkv.weight", + "backbone.layers.5.attn.proj.weight", + "backbone.layers.5.ffn.layers.0.0.weight", + "backbone.layers.5.ffn.layers.1.weight" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.05 + }, + "layer_7_no_decay": { + "param_names": [ + "backbone.layers.6.gamma_1", + "backbone.layers.6.gamma_2", + "backbone.layers.6.ln1.weight", + "backbone.layers.6.ln1.bias", + "backbone.layers.6.attn.qkv.bias", + "backbone.layers.6.attn.proj.bias", + "backbone.layers.6.ln2.weight", + "backbone.layers.6.ln2.bias", + "backbone.layers.6.ffn.layers.0.0.bias", + "backbone.layers.6.ffn.layers.1.bias" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.0 + }, + "layer_7_decay": { + "param_names": [ + "backbone.layers.6.attn.relative_position_bias_table", + "backbone.layers.6.attn.qkv.weight", + "backbone.layers.6.attn.proj.weight", + "backbone.layers.6.ffn.layers.0.0.weight", + "backbone.layers.6.ffn.layers.1.weight" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.05 + }, + "layer_8_no_decay": { + "param_names": [ + "backbone.layers.7.gamma_1", + "backbone.layers.7.gamma_2", + "backbone.layers.7.ln1.weight", + "backbone.layers.7.ln1.bias", + "backbone.layers.7.attn.qkv.bias", + "backbone.layers.7.attn.proj.bias", + "backbone.layers.7.ln2.weight", + "backbone.layers.7.ln2.bias", + "backbone.layers.7.ffn.layers.0.0.bias", + "backbone.layers.7.ffn.layers.1.bias" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.0 + }, + "layer_8_decay": { + "param_names": [ + "backbone.layers.7.attn.relative_position_bias_table", + "backbone.layers.7.attn.qkv.weight", + "backbone.layers.7.attn.proj.weight", + "backbone.layers.7.ffn.layers.0.0.weight", + "backbone.layers.7.ffn.layers.1.weight" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.05 + }, + "layer_9_no_decay": { + "param_names": [ + "backbone.layers.8.gamma_1", + "backbone.layers.8.gamma_2", + "backbone.layers.8.ln1.weight", + "backbone.layers.8.ln1.bias", + "backbone.layers.8.attn.qkv.bias", + "backbone.layers.8.attn.proj.bias", + "backbone.layers.8.ln2.weight", + "backbone.layers.8.ln2.bias", + "backbone.layers.8.ffn.layers.0.0.bias", + "backbone.layers.8.ffn.layers.1.bias" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.0 + }, + "layer_9_decay": { + "param_names": [ + "backbone.layers.8.attn.relative_position_bias_table", + "backbone.layers.8.attn.qkv.weight", + "backbone.layers.8.attn.proj.weight", + "backbone.layers.8.ffn.layers.0.0.weight", + "backbone.layers.8.ffn.layers.1.weight" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.05 + }, + "layer_10_no_decay": { + "param_names": [ + "backbone.layers.9.gamma_1", + "backbone.layers.9.gamma_2", + "backbone.layers.9.ln1.weight", + "backbone.layers.9.ln1.bias", + "backbone.layers.9.attn.qkv.bias", + "backbone.layers.9.attn.proj.bias", + "backbone.layers.9.ln2.weight", + "backbone.layers.9.ln2.bias", + "backbone.layers.9.ffn.layers.0.0.bias", + "backbone.layers.9.ffn.layers.1.bias" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.0 + }, + "layer_10_decay": { + "param_names": [ + "backbone.layers.9.attn.relative_position_bias_table", + "backbone.layers.9.attn.qkv.weight", + "backbone.layers.9.attn.proj.weight", + "backbone.layers.9.ffn.layers.0.0.weight", + "backbone.layers.9.ffn.layers.1.weight" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.05 + }, + "layer_11_no_decay": { + "param_names": [ + "backbone.layers.10.gamma_1", + "backbone.layers.10.gamma_2", + "backbone.layers.10.ln1.weight", + "backbone.layers.10.ln1.bias", + "backbone.layers.10.attn.qkv.bias", + "backbone.layers.10.attn.proj.bias", + "backbone.layers.10.ln2.weight", + "backbone.layers.10.ln2.bias", + "backbone.layers.10.ffn.layers.0.0.bias", + "backbone.layers.10.ffn.layers.1.bias" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.0 + }, + "layer_11_decay": { + "param_names": [ + "backbone.layers.10.attn.relative_position_bias_table", + "backbone.layers.10.attn.qkv.weight", + "backbone.layers.10.attn.proj.weight", + "backbone.layers.10.ffn.layers.0.0.weight", + "backbone.layers.10.ffn.layers.1.weight" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.05 + }, + "layer_12_no_decay": { + "param_names": [ + "backbone.layers.11.gamma_1", + "backbone.layers.11.gamma_2", + "backbone.layers.11.ln1.weight", + "backbone.layers.11.ln1.bias", + "backbone.layers.11.attn.qkv.bias", + "backbone.layers.11.attn.proj.bias", + "backbone.layers.11.ln2.weight", + "backbone.layers.11.ln2.bias", + "backbone.layers.11.ffn.layers.0.0.bias", + "backbone.layers.11.ffn.layers.1.bias" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.0 + }, + "layer_12_decay": { + "param_names": [ + "backbone.layers.11.attn.relative_position_bias_table", + "backbone.layers.11.attn.qkv.weight", + "backbone.layers.11.attn.proj.weight", + "backbone.layers.11.ffn.layers.0.0.weight", + "backbone.layers.11.ffn.layers.1.weight" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.05 + }, + "layer_13_decay": { + "param_names": [ + "neck.upsample_4x.0.weight", + "neck.upsample_4x.3.weight", + "neck.upsample_2x.0.weight", + "decode_head.conv_seg.weight", + "decode_head.psp_modules.0.1.conv.weight", + "decode_head.psp_modules.1.1.conv.weight", + "decode_head.psp_modules.2.1.conv.weight", + "decode_head.psp_modules.3.1.conv.weight", + "decode_head.bottleneck.conv.weight", + "decode_head.lateral_convs.0.conv.weight", + "decode_head.lateral_convs.1.conv.weight", + "decode_head.lateral_convs.2.conv.weight", + "decode_head.fpn_convs.0.conv.weight", + "decode_head.fpn_convs.1.conv.weight", + "decode_head.fpn_convs.2.conv.weight", + "decode_head.fpn_bottleneck.conv.weight", + "auxiliary_head.conv_seg.weight", + "auxiliary_head.convs.0.conv.weight" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.05 + }, + "layer_13_no_decay": { + "param_names": [ + "neck.upsample_4x.0.bias", + "neck.upsample_4x.1.weight", + "neck.upsample_4x.1.bias", + "neck.upsample_4x.3.bias", + "neck.upsample_2x.0.bias", + "decode_head.conv_seg.bias", + "decode_head.psp_modules.0.1.bn.weight", + "decode_head.psp_modules.0.1.bn.bias", + "decode_head.psp_modules.1.1.bn.weight", + "decode_head.psp_modules.1.1.bn.bias", + "decode_head.psp_modules.2.1.bn.weight", + "decode_head.psp_modules.2.1.bn.bias", + "decode_head.psp_modules.3.1.bn.weight", + "decode_head.psp_modules.3.1.bn.bias", + "decode_head.bottleneck.bn.weight", + "decode_head.bottleneck.bn.bias", + "decode_head.lateral_convs.0.bn.weight", + "decode_head.lateral_convs.0.bn.bias", + "decode_head.lateral_convs.1.bn.weight", + "decode_head.lateral_convs.1.bn.bias", + "decode_head.lateral_convs.2.bn.weight", + "decode_head.lateral_convs.2.bn.bias", + "decode_head.fpn_convs.0.bn.weight", + "decode_head.fpn_convs.0.bn.bias", + "decode_head.fpn_convs.1.bn.weight", + "decode_head.fpn_convs.1.bn.bias", + "decode_head.fpn_convs.2.bn.weight", + "decode_head.fpn_convs.2.bn.bias", + "decode_head.fpn_bottleneck.bn.weight", + "decode_head.fpn_bottleneck.bn.bias", + "auxiliary_head.conv_seg.bias", + "auxiliary_head.convs.0.bn.weight", + "auxiliary_head.convs.0.bn.bias" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.0 + } +} +04/17 08:31:37 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +04/17 08:31:37 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +04/17 08:31:37 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +04/17 08:31:37 - mmengine - INFO - Checkpoints will be saved to /workspaces/mmsegmentation-1/work_dirs/MAE4000. +04/17 08:32:34 - mmengine - INFO - Iter(train) [ 50/160000] base_lr: 3.2689e-06 lr: 1.2086e-08 eta: 2 days, 2:06:51 time: 1.0239 data_time: 0.0042 memory: 8935 loss: 7.0101 decode.loss_ce: 5.0132 decode.acc_seg: 1.2255 aux.loss_ce: 1.9970 aux.acc_seg: 0.1034 +04/17 08:33:25 - mmengine - INFO - Iter(train) [ 100/160000] base_lr: 6.6045e-06 lr: 2.4418e-08 eta: 1 day, 23:47:20 time: 1.0232 data_time: 0.0044 memory: 8462 loss: 6.7600 decode.loss_ce: 4.7872 decode.acc_seg: 22.1193 aux.loss_ce: 1.9728 aux.acc_seg: 13.3171 +04/17 08:34:16 - mmengine - INFO - Iter(train) [ 150/160000] base_lr: 9.9401e-06 lr: 3.6750e-08 eta: 1 day, 22:59:29 time: 1.0228 data_time: 0.0041 memory: 8462 loss: 6.2894 decode.loss_ce: 4.3502 decode.acc_seg: 42.1999 aux.loss_ce: 1.9392 aux.acc_seg: 23.7051 +04/17 08:35:07 - mmengine - INFO - Iter(train) [ 200/160000] base_lr: 1.3276e-05 lr: 4.9083e-08 eta: 1 day, 22:33:58 time: 1.0184 data_time: 0.0040 memory: 8462 loss: 5.7614 decode.loss_ce: 3.8749 decode.acc_seg: 58.2272 aux.loss_ce: 1.8865 aux.acc_seg: 34.2005 +04/17 08:35:58 - mmengine - INFO - Iter(train) [ 250/160000] base_lr: 1.6611e-05 lr: 6.1415e-08 eta: 1 day, 22:15:08 time: 1.0145 data_time: 0.0039 memory: 8462 loss: 5.0487 decode.loss_ce: 3.2347 decode.acc_seg: 78.0340 aux.loss_ce: 1.8141 aux.acc_seg: 43.1890 +04/17 08:36:49 - mmengine - INFO - Iter(train) [ 300/160000] base_lr: 1.9947e-05 lr: 7.3747e-08 eta: 1 day, 22:00:43 time: 1.0089 data_time: 0.0040 memory: 8462 loss: 4.3936 decode.loss_ce: 2.6770 decode.acc_seg: 85.6838 aux.loss_ce: 1.7167 aux.acc_seg: 64.1836 +04/17 08:37:39 - mmengine - INFO - Iter(train) [ 350/160000] base_lr: 2.3282e-05 lr: 8.6079e-08 eta: 1 day, 21:48:21 time: 1.0050 data_time: 0.0044 memory: 8462 loss: 3.7508 decode.loss_ce: 2.1300 decode.acc_seg: 87.5259 aux.loss_ce: 1.6208 aux.acc_seg: 62.4052 +04/17 08:38:29 - mmengine - INFO - Iter(train) [ 400/160000] base_lr: 2.6618e-05 lr: 9.8412e-08 eta: 1 day, 21:38:10 time: 1.0046 data_time: 0.0045 memory: 8462 loss: 3.1693 decode.loss_ce: 1.6545 decode.acc_seg: 95.9000 aux.loss_ce: 1.5148 aux.acc_seg: 74.7307 +04/17 08:39:19 - mmengine - INFO - Iter(train) [ 450/160000] base_lr: 2.9953e-05 lr: 1.1074e-07 eta: 1 day, 21:29:51 time: 1.0036 data_time: 0.0041 memory: 8462 loss: 2.5232 decode.loss_ce: 1.1461 decode.acc_seg: 96.0243 aux.loss_ce: 1.3771 aux.acc_seg: 65.2546 +04/17 08:40:10 - mmengine - INFO - Iter(train) [ 500/160000] base_lr: 3.3289e-05 lr: 1.2308e-07 eta: 1 day, 21:23:02 time: 1.0035 data_time: 0.0040 memory: 8462 loss: 2.0180 decode.loss_ce: 0.7713 decode.acc_seg: 95.5585 aux.loss_ce: 1.2467 aux.acc_seg: 72.2454 +04/17 08:41:00 - mmengine - INFO - Iter(train) [ 550/160000] base_lr: 3.6624e-05 lr: 1.3541e-07 eta: 1 day, 21:17:08 time: 1.0037 data_time: 0.0041 memory: 8462 loss: 1.6391 decode.loss_ce: 0.5415 decode.acc_seg: 97.0821 aux.loss_ce: 1.0976 aux.acc_seg: 81.3814 +04/17 08:41:50 - mmengine - INFO - Iter(train) [ 600/160000] base_lr: 3.9960e-05 lr: 1.4774e-07 eta: 1 day, 21:11:56 time: 1.0025 data_time: 0.0044 memory: 8462 loss: 1.2957 decode.loss_ce: 0.3839 decode.acc_seg: 96.9400 aux.loss_ce: 0.9119 aux.acc_seg: 88.7630 +04/17 08:42:40 - mmengine - INFO - Iter(train) [ 650/160000] base_lr: 4.3296e-05 lr: 1.6007e-07 eta: 1 day, 21:07:17 time: 1.0012 data_time: 0.0042 memory: 8462 loss: 1.0201 decode.loss_ce: 0.2598 decode.acc_seg: 99.0055 aux.loss_ce: 0.7604 aux.acc_seg: 94.6356 +04/17 08:43:30 - mmengine - INFO - Iter(train) [ 700/160000] base_lr: 4.6631e-05 lr: 1.7240e-07 eta: 1 day, 21:02:57 time: 1.0004 data_time: 0.0046 memory: 8462 loss: 0.8337 decode.loss_ce: 0.2176 decode.acc_seg: 96.4949 aux.loss_ce: 0.6161 aux.acc_seg: 95.5870 +04/17 08:44:20 - mmengine - INFO - Iter(train) [ 750/160000] base_lr: 4.9967e-05 lr: 1.8474e-07 eta: 1 day, 20:58:43 time: 0.9966 data_time: 0.0043 memory: 8462 loss: 0.6015 decode.loss_ce: 0.1586 decode.acc_seg: 97.6515 aux.loss_ce: 0.4429 aux.acc_seg: 97.1811 +04/17 08:45:10 - mmengine - INFO - Iter(train) [ 800/160000] base_lr: 5.3302e-05 lr: 1.9707e-07 eta: 1 day, 20:54:39 time: 0.9973 data_time: 0.0043 memory: 8462 loss: 0.4766 decode.loss_ce: 0.1337 decode.acc_seg: 98.6265 aux.loss_ce: 0.3429 aux.acc_seg: 97.3650 +04/17 08:46:00 - mmengine - INFO - Iter(train) [ 850/160000] base_lr: 5.6638e-05 lr: 2.0940e-07 eta: 1 day, 20:50:56 time: 0.9974 data_time: 0.0039 memory: 8462 loss: 0.3757 decode.loss_ce: 0.1175 decode.acc_seg: 98.1531 aux.loss_ce: 0.2582 aux.acc_seg: 96.8962 +04/17 08:46:50 - mmengine - INFO - Iter(train) [ 900/160000] base_lr: 5.9973e-05 lr: 2.2173e-07 eta: 1 day, 20:47:32 time: 0.9970 data_time: 0.0040 memory: 8462 loss: 0.3494 decode.loss_ce: 0.1149 decode.acc_seg: 97.5227 aux.loss_ce: 0.2345 aux.acc_seg: 96.7276 +04/17 08:47:39 - mmengine - INFO - Iter(train) [ 950/160000] base_lr: 6.3309e-05 lr: 2.3407e-07 eta: 1 day, 20:44:24 time: 0.9968 data_time: 0.0040 memory: 8462 loss: 0.2647 decode.loss_ce: 0.0933 decode.acc_seg: 98.1697 aux.loss_ce: 0.1714 aux.acc_seg: 97.2790 +04/17 08:48:29 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 08:48:29 - mmengine - INFO - Iter(train) [ 1000/160000] base_lr: 6.6644e-05 lr: 2.4640e-07 eta: 1 day, 20:41:36 time: 0.9979 data_time: 0.0043 memory: 8462 loss: 0.2273 decode.loss_ce: 0.0908 decode.acc_seg: 98.6879 aux.loss_ce: 0.1365 aux.acc_seg: 97.6452 +04/17 08:49:19 - mmengine - INFO - Iter(train) [ 1050/160000] base_lr: 6.9980e-05 lr: 2.5873e-07 eta: 1 day, 20:38:59 time: 0.9989 data_time: 0.0049 memory: 8462 loss: 0.2224 decode.loss_ce: 0.0969 decode.acc_seg: 97.7600 aux.loss_ce: 0.1255 aux.acc_seg: 96.2608 +04/17 08:50:09 - mmengine - INFO - Iter(train) [ 1100/160000] base_lr: 7.3316e-05 lr: 2.7106e-07 eta: 1 day, 20:36:31 time: 0.9975 data_time: 0.0044 memory: 8462 loss: 0.1755 decode.loss_ce: 0.0823 decode.acc_seg: 97.8956 aux.loss_ce: 0.0933 aux.acc_seg: 96.8731 +04/17 08:50:59 - mmengine - INFO - Iter(train) [ 1150/160000] base_lr: 7.6651e-05 lr: 2.8339e-07 eta: 1 day, 20:34:13 time: 0.9981 data_time: 0.0045 memory: 8462 loss: 0.1492 decode.loss_ce: 0.0711 decode.acc_seg: 97.3215 aux.loss_ce: 0.0781 aux.acc_seg: 95.2385 +04/17 08:51:49 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 08:51:49 - mmengine - INFO - Iter(train) [ 1200/160000] base_lr: 7.9987e-05 lr: 2.9573e-07 eta: 1 day, 20:32:04 time: 0.9979 data_time: 0.0046 memory: 8462 loss: 0.1411 decode.loss_ce: 0.0710 decode.acc_seg: 98.7997 aux.loss_ce: 0.0700 aux.acc_seg: 97.9790 +04/17 08:52:39 - mmengine - INFO - Iter(train) [ 1250/160000] base_lr: 8.3322e-05 lr: 3.0806e-07 eta: 1 day, 20:30:02 time: 0.9994 data_time: 0.0047 memory: 8462 loss: 0.1245 decode.loss_ce: 0.0608 decode.acc_seg: 98.9174 aux.loss_ce: 0.0637 aux.acc_seg: 98.4625 +04/17 08:53:29 - mmengine - INFO - Iter(train) [ 1300/160000] base_lr: 8.6658e-05 lr: 3.2039e-07 eta: 1 day, 20:28:03 time: 0.9979 data_time: 0.0041 memory: 8462 loss: 0.1184 decode.loss_ce: 0.0597 decode.acc_seg: 97.2700 aux.loss_ce: 0.0587 aux.acc_seg: 96.9965 +04/17 08:54:19 - mmengine - INFO - Iter(train) [ 1350/160000] base_lr: 8.9993e-05 lr: 3.3272e-07 eta: 1 day, 20:26:10 time: 0.9970 data_time: 0.0039 memory: 8462 loss: 0.1053 decode.loss_ce: 0.0539 decode.acc_seg: 98.1895 aux.loss_ce: 0.0514 aux.acc_seg: 97.5515 +04/17 08:55:09 - mmengine - INFO - Iter(train) [ 1400/160000] base_lr: 9.3329e-05 lr: 3.4506e-07 eta: 1 day, 20:24:20 time: 0.9986 data_time: 0.0044 memory: 8462 loss: 0.0978 decode.loss_ce: 0.0511 decode.acc_seg: 98.8268 aux.loss_ce: 0.0467 aux.acc_seg: 98.3084 +04/17 08:55:58 - mmengine - INFO - Iter(train) [ 1450/160000] base_lr: 9.6664e-05 lr: 3.5739e-07 eta: 1 day, 20:22:34 time: 0.9980 data_time: 0.0045 memory: 8462 loss: 0.1061 decode.loss_ce: 0.0579 decode.acc_seg: 97.1462 aux.loss_ce: 0.0482 aux.acc_seg: 95.9591 +04/17 08:56:48 - mmengine - INFO - Iter(train) [ 1500/160000] base_lr: 1.0000e-04 lr: 3.6972e-07 eta: 1 day, 20:20:56 time: 0.9979 data_time: 0.0038 memory: 8462 loss: 0.1053 decode.loss_ce: 0.0602 decode.acc_seg: 97.4928 aux.loss_ce: 0.0451 aux.acc_seg: 97.0434 +04/17 08:57:38 - mmengine - INFO - Iter(train) [ 1550/160000] base_lr: 9.9969e-05 lr: 3.6961e-07 eta: 1 day, 20:19:24 time: 0.9988 data_time: 0.0041 memory: 8462 loss: 0.0964 decode.loss_ce: 0.0526 decode.acc_seg: 98.3452 aux.loss_ce: 0.0438 aux.acc_seg: 97.5035 +04/17 08:58:28 - mmengine - INFO - Iter(train) [ 1600/160000] base_lr: 9.9938e-05 lr: 3.6949e-07 eta: 1 day, 20:17:50 time: 0.9984 data_time: 0.0041 memory: 8462 loss: 0.0870 decode.loss_ce: 0.0482 decode.acc_seg: 97.9443 aux.loss_ce: 0.0388 aux.acc_seg: 97.1024 +04/17 08:59:18 - mmengine - INFO - Iter(train) [ 1650/160000] base_lr: 9.9906e-05 lr: 3.6937e-07 eta: 1 day, 20:16:19 time: 0.9976 data_time: 0.0046 memory: 8462 loss: 0.0931 decode.loss_ce: 0.0529 decode.acc_seg: 98.5298 aux.loss_ce: 0.0402 aux.acc_seg: 98.1544 +04/17 09:00:08 - mmengine - INFO - Iter(train) [ 1700/160000] base_lr: 9.9874e-05 lr: 3.6926e-07 eta: 1 day, 20:14:52 time: 0.9971 data_time: 0.0042 memory: 8462 loss: 0.0831 decode.loss_ce: 0.0484 decode.acc_seg: 97.0835 aux.loss_ce: 0.0347 aux.acc_seg: 96.7350 +04/17 09:00:58 - mmengine - INFO - Iter(train) [ 1750/160000] base_lr: 9.9843e-05 lr: 3.6914e-07 eta: 1 day, 20:13:28 time: 0.9986 data_time: 0.0043 memory: 8462 loss: 0.0656 decode.loss_ce: 0.0356 decode.acc_seg: 98.7150 aux.loss_ce: 0.0301 aux.acc_seg: 97.9824 +04/17 09:01:48 - mmengine - INFO - Iter(train) [ 1800/160000] base_lr: 9.9811e-05 lr: 3.6902e-07 eta: 1 day, 20:12:06 time: 0.9982 data_time: 0.0041 memory: 8462 loss: 0.0693 decode.loss_ce: 0.0407 decode.acc_seg: 99.0469 aux.loss_ce: 0.0285 aux.acc_seg: 98.2542 +04/17 09:02:38 - mmengine - INFO - Iter(train) [ 1850/160000] base_lr: 9.9780e-05 lr: 3.6891e-07 eta: 1 day, 20:10:46 time: 0.9991 data_time: 0.0042 memory: 8462 loss: 0.0593 decode.loss_ce: 0.0350 decode.acc_seg: 98.9889 aux.loss_ce: 0.0243 aux.acc_seg: 98.6946 +04/17 09:03:28 - mmengine - INFO - Iter(train) [ 1900/160000] base_lr: 9.9748e-05 lr: 3.6879e-07 eta: 1 day, 20:09:26 time: 0.9981 data_time: 0.0044 memory: 8462 loss: 0.0596 decode.loss_ce: 0.0332 decode.acc_seg: 98.8941 aux.loss_ce: 0.0264 aux.acc_seg: 98.0751 +04/17 09:04:18 - mmengine - INFO - Iter(train) [ 1950/160000] base_lr: 9.9717e-05 lr: 3.6867e-07 eta: 1 day, 20:08:11 time: 0.9993 data_time: 0.0047 memory: 8462 loss: 0.0573 decode.loss_ce: 0.0316 decode.acc_seg: 99.2884 aux.loss_ce: 0.0257 aux.acc_seg: 98.5992 +04/17 09:05:08 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 09:05:08 - mmengine - INFO - Iter(train) [ 2000/160000] base_lr: 9.9685e-05 lr: 3.6856e-07 eta: 1 day, 20:06:58 time: 0.9992 data_time: 0.0044 memory: 8462 loss: 0.0635 decode.loss_ce: 0.0379 decode.acc_seg: 99.3744 aux.loss_ce: 0.0256 aux.acc_seg: 98.4859 +04/17 09:05:58 - mmengine - INFO - Iter(train) [ 2050/160000] base_lr: 9.9654e-05 lr: 3.6844e-07 eta: 1 day, 20:05:42 time: 0.9965 data_time: 0.0040 memory: 8462 loss: 0.0711 decode.loss_ce: 0.0426 decode.acc_seg: 98.8665 aux.loss_ce: 0.0285 aux.acc_seg: 98.4669 +04/17 09:06:48 - mmengine - INFO - Iter(train) [ 2100/160000] base_lr: 9.9622e-05 lr: 3.6832e-07 eta: 1 day, 20:04:29 time: 0.9983 data_time: 0.0042 memory: 8462 loss: 0.0604 decode.loss_ce: 0.0348 decode.acc_seg: 98.1258 aux.loss_ce: 0.0257 aux.acc_seg: 97.6543 +04/17 09:07:38 - mmengine - INFO - Iter(train) [ 2150/160000] base_lr: 9.9591e-05 lr: 3.6821e-07 eta: 1 day, 20:03:15 time: 0.9988 data_time: 0.0042 memory: 8462 loss: 0.0644 decode.loss_ce: 0.0382 decode.acc_seg: 97.8350 aux.loss_ce: 0.0261 aux.acc_seg: 97.0947 +04/17 09:08:28 - mmengine - INFO - Iter(train) [ 2200/160000] base_lr: 9.9559e-05 lr: 3.6809e-07 eta: 1 day, 20:02:05 time: 0.9998 data_time: 0.0046 memory: 8462 loss: 0.0556 decode.loss_ce: 0.0321 decode.acc_seg: 98.8342 aux.loss_ce: 0.0235 aux.acc_seg: 98.3250 +04/17 09:09:18 - mmengine - INFO - Iter(train) [ 2250/160000] base_lr: 9.9527e-05 lr: 3.6797e-07 eta: 1 day, 20:00:56 time: 0.9994 data_time: 0.0040 memory: 8462 loss: 0.0595 decode.loss_ce: 0.0345 decode.acc_seg: 99.4347 aux.loss_ce: 0.0250 aux.acc_seg: 98.9706 +04/17 09:10:07 - mmengine - INFO - Iter(train) [ 2300/160000] base_lr: 9.9496e-05 lr: 3.6786e-07 eta: 1 day, 19:59:48 time: 0.9995 data_time: 0.0045 memory: 8462 loss: 0.0528 decode.loss_ce: 0.0305 decode.acc_seg: 99.5253 aux.loss_ce: 0.0223 aux.acc_seg: 98.7959 +04/17 09:10:57 - mmengine - INFO - Iter(train) [ 2350/160000] base_lr: 9.9464e-05 lr: 3.6774e-07 eta: 1 day, 19:58:40 time: 0.9990 data_time: 0.0043 memory: 8462 loss: 0.0578 decode.loss_ce: 0.0343 decode.acc_seg: 97.5035 aux.loss_ce: 0.0235 aux.acc_seg: 97.0406 +04/17 09:11:47 - mmengine - INFO - Iter(train) [ 2400/160000] base_lr: 9.9433e-05 lr: 3.6762e-07 eta: 1 day, 19:57:34 time: 0.9997 data_time: 0.0043 memory: 8462 loss: 0.0488 decode.loss_ce: 0.0280 decode.acc_seg: 98.8338 aux.loss_ce: 0.0208 aux.acc_seg: 98.2491 +04/17 09:12:37 - mmengine - INFO - Iter(train) [ 2450/160000] base_lr: 9.9401e-05 lr: 3.6751e-07 eta: 1 day, 19:56:30 time: 0.9997 data_time: 0.0042 memory: 8462 loss: 0.0573 decode.loss_ce: 0.0339 decode.acc_seg: 98.7181 aux.loss_ce: 0.0234 aux.acc_seg: 98.1361 +04/17 09:13:27 - mmengine - INFO - Iter(train) [ 2500/160000] base_lr: 9.9370e-05 lr: 3.6739e-07 eta: 1 day, 19:55:24 time: 0.9984 data_time: 0.0040 memory: 8462 loss: 0.0536 decode.loss_ce: 0.0317 decode.acc_seg: 98.9445 aux.loss_ce: 0.0219 aux.acc_seg: 98.4241 +04/17 09:14:17 - mmengine - INFO - Iter(train) [ 2550/160000] base_lr: 9.9338e-05 lr: 3.6727e-07 eta: 1 day, 19:54:20 time: 0.9994 data_time: 0.0041 memory: 8462 loss: 0.0553 decode.loss_ce: 0.0330 decode.acc_seg: 98.1611 aux.loss_ce: 0.0222 aux.acc_seg: 97.6421 +04/17 09:15:07 - mmengine - INFO - Iter(train) [ 2600/160000] base_lr: 9.9307e-05 lr: 3.6716e-07 eta: 1 day, 19:53:18 time: 0.9994 data_time: 0.0044 memory: 8462 loss: 0.0635 decode.loss_ce: 0.0380 decode.acc_seg: 99.0538 aux.loss_ce: 0.0255 aux.acc_seg: 98.2327 +04/17 09:15:57 - mmengine - INFO - Iter(train) [ 2650/160000] base_lr: 9.9275e-05 lr: 3.6704e-07 eta: 1 day, 19:52:16 time: 0.9996 data_time: 0.0042 memory: 8462 loss: 0.0500 decode.loss_ce: 0.0291 decode.acc_seg: 98.7988 aux.loss_ce: 0.0209 aux.acc_seg: 98.3414 +04/17 09:16:47 - mmengine - INFO - Iter(train) [ 2700/160000] base_lr: 9.9244e-05 lr: 3.6692e-07 eta: 1 day, 19:51:14 time: 0.9999 data_time: 0.0050 memory: 8462 loss: 0.0618 decode.loss_ce: 0.0368 decode.acc_seg: 98.3723 aux.loss_ce: 0.0249 aux.acc_seg: 97.3526 +04/17 09:17:37 - mmengine - INFO - Iter(train) [ 2750/160000] base_lr: 9.9212e-05 lr: 3.6681e-07 eta: 1 day, 19:50:11 time: 0.9998 data_time: 0.0045 memory: 8462 loss: 0.0538 decode.loss_ce: 0.0319 decode.acc_seg: 98.5094 aux.loss_ce: 0.0220 aux.acc_seg: 98.0539 +04/17 09:18:27 - mmengine - INFO - Iter(train) [ 2800/160000] base_lr: 9.9180e-05 lr: 3.6669e-07 eta: 1 day, 19:49:10 time: 1.0008 data_time: 0.0046 memory: 8462 loss: 0.0552 decode.loss_ce: 0.0334 decode.acc_seg: 98.9201 aux.loss_ce: 0.0217 aux.acc_seg: 98.0875 +04/17 09:19:17 - mmengine - INFO - Iter(train) [ 2850/160000] base_lr: 9.9149e-05 lr: 3.6657e-07 eta: 1 day, 19:48:11 time: 1.0006 data_time: 0.0046 memory: 8462 loss: 0.0417 decode.loss_ce: 0.0238 decode.acc_seg: 99.2218 aux.loss_ce: 0.0179 aux.acc_seg: 99.0133 +04/17 09:20:07 - mmengine - INFO - Iter(train) [ 2900/160000] base_lr: 9.9117e-05 lr: 3.6646e-07 eta: 1 day, 19:47:10 time: 0.9996 data_time: 0.0044 memory: 8462 loss: 0.0500 decode.loss_ce: 0.0298 decode.acc_seg: 99.1209 aux.loss_ce: 0.0201 aux.acc_seg: 98.5458 +04/17 09:20:57 - mmengine - INFO - Iter(train) [ 2950/160000] base_lr: 9.9086e-05 lr: 3.6634e-07 eta: 1 day, 19:46:10 time: 0.9993 data_time: 0.0043 memory: 8462 loss: 0.0529 decode.loss_ce: 0.0313 decode.acc_seg: 99.5939 aux.loss_ce: 0.0216 aux.acc_seg: 99.2641 +04/17 09:21:47 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 09:21:47 - mmengine - INFO - Iter(train) [ 3000/160000] base_lr: 9.9054e-05 lr: 3.6622e-07 eta: 1 day, 19:45:09 time: 0.9983 data_time: 0.0041 memory: 8462 loss: 0.0457 decode.loss_ce: 0.0268 decode.acc_seg: 98.7343 aux.loss_ce: 0.0190 aux.acc_seg: 98.1049 +04/17 09:22:37 - mmengine - INFO - Iter(train) [ 3050/160000] base_lr: 9.9023e-05 lr: 3.6611e-07 eta: 1 day, 19:44:11 time: 1.0002 data_time: 0.0042 memory: 8462 loss: 0.0476 decode.loss_ce: 0.0275 decode.acc_seg: 99.1837 aux.loss_ce: 0.0200 aux.acc_seg: 98.4442 +04/17 09:23:27 - mmengine - INFO - Iter(train) [ 3100/160000] base_lr: 9.8991e-05 lr: 3.6599e-07 eta: 1 day, 19:43:12 time: 0.9995 data_time: 0.0047 memory: 8462 loss: 0.0473 decode.loss_ce: 0.0283 decode.acc_seg: 98.4861 aux.loss_ce: 0.0190 aux.acc_seg: 97.4447 +04/17 09:24:17 - mmengine - INFO - Iter(train) [ 3150/160000] base_lr: 9.8960e-05 lr: 3.6587e-07 eta: 1 day, 19:42:14 time: 1.0001 data_time: 0.0043 memory: 8462 loss: 0.0475 decode.loss_ce: 0.0282 decode.acc_seg: 98.9435 aux.loss_ce: 0.0193 aux.acc_seg: 98.5840 +04/17 09:25:07 - mmengine - INFO - Iter(train) [ 3200/160000] base_lr: 9.8928e-05 lr: 3.6576e-07 eta: 1 day, 19:41:16 time: 0.9992 data_time: 0.0039 memory: 8462 loss: 0.0488 decode.loss_ce: 0.0294 decode.acc_seg: 98.9069 aux.loss_ce: 0.0194 aux.acc_seg: 98.5401 +04/17 09:25:57 - mmengine - INFO - Iter(train) [ 3250/160000] base_lr: 9.8897e-05 lr: 3.6564e-07 eta: 1 day, 19:40:18 time: 0.9995 data_time: 0.0045 memory: 8462 loss: 0.0468 decode.loss_ce: 0.0273 decode.acc_seg: 99.3895 aux.loss_ce: 0.0195 aux.acc_seg: 99.0913 +04/17 09:26:47 - mmengine - INFO - Iter(train) [ 3300/160000] base_lr: 9.8865e-05 lr: 3.6552e-07 eta: 1 day, 19:39:20 time: 0.9995 data_time: 0.0044 memory: 8462 loss: 0.0460 decode.loss_ce: 0.0271 decode.acc_seg: 97.9965 aux.loss_ce: 0.0189 aux.acc_seg: 97.4051 +04/17 09:27:37 - mmengine - INFO - Iter(train) [ 3350/160000] base_lr: 9.8833e-05 lr: 3.6541e-07 eta: 1 day, 19:38:21 time: 0.9993 data_time: 0.0042 memory: 8462 loss: 0.0462 decode.loss_ce: 0.0276 decode.acc_seg: 99.5575 aux.loss_ce: 0.0186 aux.acc_seg: 99.0801 +04/17 09:28:27 - mmengine - INFO - Iter(train) [ 3400/160000] base_lr: 9.8802e-05 lr: 3.6529e-07 eta: 1 day, 19:37:25 time: 1.0004 data_time: 0.0042 memory: 8462 loss: 0.0479 decode.loss_ce: 0.0294 decode.acc_seg: 99.4705 aux.loss_ce: 0.0185 aux.acc_seg: 98.8924 +04/17 09:29:17 - mmengine - INFO - Iter(train) [ 3450/160000] base_lr: 9.8770e-05 lr: 3.6517e-07 eta: 1 day, 19:36:28 time: 0.9993 data_time: 0.0047 memory: 8462 loss: 0.0443 decode.loss_ce: 0.0267 decode.acc_seg: 99.0513 aux.loss_ce: 0.0176 aux.acc_seg: 98.5983 +04/17 09:30:07 - mmengine - INFO - Iter(train) [ 3500/160000] base_lr: 9.8739e-05 lr: 3.6506e-07 eta: 1 day, 19:35:31 time: 1.0001 data_time: 0.0043 memory: 8462 loss: 0.0527 decode.loss_ce: 0.0326 decode.acc_seg: 99.3343 aux.loss_ce: 0.0202 aux.acc_seg: 98.6254 +04/17 09:30:57 - mmengine - INFO - Iter(train) [ 3550/160000] base_lr: 9.8707e-05 lr: 3.6494e-07 eta: 1 day, 19:34:35 time: 0.9992 data_time: 0.0042 memory: 8462 loss: 0.0407 decode.loss_ce: 0.0241 decode.acc_seg: 99.0374 aux.loss_ce: 0.0166 aux.acc_seg: 98.3089 +04/17 09:31:47 - mmengine - INFO - Iter(train) [ 3600/160000] base_lr: 9.8676e-05 lr: 3.6482e-07 eta: 1 day, 19:33:38 time: 0.9999 data_time: 0.0042 memory: 8462 loss: 0.0448 decode.loss_ce: 0.0264 decode.acc_seg: 98.7450 aux.loss_ce: 0.0184 aux.acc_seg: 98.2246 +04/17 09:32:37 - mmengine - INFO - Iter(train) [ 3650/160000] base_lr: 9.8644e-05 lr: 3.6471e-07 eta: 1 day, 19:32:43 time: 1.0006 data_time: 0.0043 memory: 8462 loss: 0.0418 decode.loss_ce: 0.0246 decode.acc_seg: 99.4370 aux.loss_ce: 0.0172 aux.acc_seg: 98.7982 +04/17 09:33:27 - mmengine - INFO - Iter(train) [ 3700/160000] base_lr: 9.8613e-05 lr: 3.6459e-07 eta: 1 day, 19:31:47 time: 1.0014 data_time: 0.0041 memory: 8462 loss: 0.0480 decode.loss_ce: 0.0291 decode.acc_seg: 98.6782 aux.loss_ce: 0.0189 aux.acc_seg: 98.5292 +04/17 09:34:17 - mmengine - INFO - Iter(train) [ 3750/160000] base_lr: 9.8581e-05 lr: 3.6447e-07 eta: 1 day, 19:30:50 time: 0.9994 data_time: 0.0042 memory: 8462 loss: 0.0477 decode.loss_ce: 0.0289 decode.acc_seg: 99.2256 aux.loss_ce: 0.0188 aux.acc_seg: 98.6662 +04/17 09:35:07 - mmengine - INFO - Iter(train) [ 3800/160000] base_lr: 9.8550e-05 lr: 3.6436e-07 eta: 1 day, 19:29:53 time: 0.9986 data_time: 0.0039 memory: 8462 loss: 0.0446 decode.loss_ce: 0.0276 decode.acc_seg: 99.1703 aux.loss_ce: 0.0170 aux.acc_seg: 98.7638 +04/17 09:35:57 - mmengine - INFO - Iter(train) [ 3850/160000] base_lr: 9.8518e-05 lr: 3.6424e-07 eta: 1 day, 19:28:56 time: 0.9994 data_time: 0.0042 memory: 8462 loss: 0.0389 decode.loss_ce: 0.0235 decode.acc_seg: 99.4406 aux.loss_ce: 0.0154 aux.acc_seg: 98.9601 +04/17 09:36:47 - mmengine - INFO - Iter(train) [ 3900/160000] base_lr: 9.8486e-05 lr: 3.6412e-07 eta: 1 day, 19:28:01 time: 0.9997 data_time: 0.0045 memory: 8462 loss: 0.0435 decode.loss_ce: 0.0262 decode.acc_seg: 98.1987 aux.loss_ce: 0.0173 aux.acc_seg: 97.4186 +04/17 09:37:37 - mmengine - INFO - Iter(train) [ 3950/160000] base_lr: 9.8455e-05 lr: 3.6401e-07 eta: 1 day, 19:27:06 time: 1.0012 data_time: 0.0041 memory: 8462 loss: 0.0486 decode.loss_ce: 0.0304 decode.acc_seg: 99.3372 aux.loss_ce: 0.0182 aux.acc_seg: 99.0168 +04/17 09:38:27 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 09:38:27 - mmengine - INFO - Iter(train) [ 4000/160000] base_lr: 9.8423e-05 lr: 3.6389e-07 eta: 1 day, 19:26:10 time: 0.9985 data_time: 0.0041 memory: 8462 loss: 0.0411 decode.loss_ce: 0.0260 decode.acc_seg: 99.3818 aux.loss_ce: 0.0151 aux.acc_seg: 98.6591 +04/17 09:39:17 - mmengine - INFO - Iter(train) [ 4050/160000] base_lr: 9.8392e-05 lr: 3.6377e-07 eta: 1 day, 19:25:15 time: 0.9987 data_time: 0.0040 memory: 8462 loss: 0.0447 decode.loss_ce: 0.0270 decode.acc_seg: 99.2233 aux.loss_ce: 0.0177 aux.acc_seg: 98.2870 +04/17 09:40:07 - mmengine - INFO - Iter(train) [ 4100/160000] base_lr: 9.8360e-05 lr: 3.6366e-07 eta: 1 day, 19:24:21 time: 1.0004 data_time: 0.0040 memory: 8462 loss: 0.0434 decode.loss_ce: 0.0265 decode.acc_seg: 99.3273 aux.loss_ce: 0.0169 aux.acc_seg: 98.5081 +04/17 09:40:57 - mmengine - INFO - Iter(train) [ 4150/160000] base_lr: 9.8329e-05 lr: 3.6354e-07 eta: 1 day, 19:23:25 time: 0.9991 data_time: 0.0042 memory: 8462 loss: 0.0374 decode.loss_ce: 0.0228 decode.acc_seg: 98.9891 aux.loss_ce: 0.0146 aux.acc_seg: 98.5933 +04/17 09:41:47 - mmengine - INFO - Iter(train) [ 4200/160000] base_lr: 9.8297e-05 lr: 3.6342e-07 eta: 1 day, 19:22:31 time: 0.9992 data_time: 0.0044 memory: 8462 loss: 0.0402 decode.loss_ce: 0.0240 decode.acc_seg: 98.2637 aux.loss_ce: 0.0162 aux.acc_seg: 98.2533 +04/17 09:42:37 - mmengine - INFO - Iter(train) [ 4250/160000] base_lr: 9.8266e-05 lr: 3.6331e-07 eta: 1 day, 19:21:36 time: 0.9989 data_time: 0.0041 memory: 8462 loss: 0.0347 decode.loss_ce: 0.0202 decode.acc_seg: 99.0572 aux.loss_ce: 0.0145 aux.acc_seg: 98.3082 +04/17 09:43:27 - mmengine - INFO - Iter(train) [ 4300/160000] base_lr: 9.8234e-05 lr: 3.6319e-07 eta: 1 day, 19:20:43 time: 1.0002 data_time: 0.0045 memory: 8462 loss: 0.0399 decode.loss_ce: 0.0239 decode.acc_seg: 99.0496 aux.loss_ce: 0.0160 aux.acc_seg: 98.3311 +04/17 09:44:17 - mmengine - INFO - Iter(train) [ 4350/160000] base_lr: 9.8203e-05 lr: 3.6307e-07 eta: 1 day, 19:19:49 time: 1.0009 data_time: 0.0042 memory: 8462 loss: 0.0424 decode.loss_ce: 0.0249 decode.acc_seg: 98.7101 aux.loss_ce: 0.0175 aux.acc_seg: 97.8025 +04/17 09:45:07 - mmengine - INFO - Iter(train) [ 4400/160000] base_lr: 9.8171e-05 lr: 3.6296e-07 eta: 1 day, 19:18:55 time: 0.9994 data_time: 0.0047 memory: 8462 loss: 0.0416 decode.loss_ce: 0.0249 decode.acc_seg: 99.3635 aux.loss_ce: 0.0167 aux.acc_seg: 98.9010 +04/17 09:45:57 - mmengine - INFO - Iter(train) [ 4450/160000] base_lr: 9.8139e-05 lr: 3.6284e-07 eta: 1 day, 19:18:00 time: 1.0001 data_time: 0.0048 memory: 8462 loss: 0.0336 decode.loss_ce: 0.0196 decode.acc_seg: 99.1730 aux.loss_ce: 0.0139 aux.acc_seg: 98.1127 +04/17 09:46:47 - mmengine - INFO - Iter(train) [ 4500/160000] base_lr: 9.8108e-05 lr: 3.6273e-07 eta: 1 day, 19:17:06 time: 1.0009 data_time: 0.0043 memory: 8462 loss: 0.0471 decode.loss_ce: 0.0296 decode.acc_seg: 99.0349 aux.loss_ce: 0.0175 aux.acc_seg: 98.5691 +04/17 09:47:37 - mmengine - INFO - Iter(train) [ 4550/160000] base_lr: 9.8076e-05 lr: 3.6261e-07 eta: 1 day, 19:16:11 time: 0.9984 data_time: 0.0042 memory: 8462 loss: 0.0390 decode.loss_ce: 0.0237 decode.acc_seg: 99.3019 aux.loss_ce: 0.0153 aux.acc_seg: 98.6286 +04/17 09:48:27 - mmengine - INFO - Iter(train) [ 4600/160000] base_lr: 9.8045e-05 lr: 3.6249e-07 eta: 1 day, 19:15:18 time: 1.0005 data_time: 0.0040 memory: 8462 loss: 0.0366 decode.loss_ce: 0.0219 decode.acc_seg: 99.6012 aux.loss_ce: 0.0147 aux.acc_seg: 99.2052 +04/17 09:49:17 - mmengine - INFO - Iter(train) [ 4650/160000] base_lr: 9.8013e-05 lr: 3.6238e-07 eta: 1 day, 19:14:23 time: 0.9996 data_time: 0.0043 memory: 8462 loss: 0.0447 decode.loss_ce: 0.0275 decode.acc_seg: 99.0816 aux.loss_ce: 0.0172 aux.acc_seg: 98.5561 +04/17 09:50:07 - mmengine - INFO - Iter(train) [ 4700/160000] base_lr: 9.7982e-05 lr: 3.6226e-07 eta: 1 day, 19:13:29 time: 0.9994 data_time: 0.0047 memory: 8462 loss: 0.0339 decode.loss_ce: 0.0196 decode.acc_seg: 99.4043 aux.loss_ce: 0.0144 aux.acc_seg: 98.8764 +04/17 09:50:57 - mmengine - INFO - Iter(train) [ 4750/160000] base_lr: 9.7950e-05 lr: 3.6214e-07 eta: 1 day, 19:12:33 time: 0.9985 data_time: 0.0049 memory: 8462 loss: 0.0379 decode.loss_ce: 0.0226 decode.acc_seg: 98.9700 aux.loss_ce: 0.0153 aux.acc_seg: 98.1937 +04/17 09:51:47 - mmengine - INFO - Iter(train) [ 4800/160000] base_lr: 9.7919e-05 lr: 3.6203e-07 eta: 1 day, 19:11:40 time: 0.9991 data_time: 0.0042 memory: 8462 loss: 0.0353 decode.loss_ce: 0.0215 decode.acc_seg: 99.1190 aux.loss_ce: 0.0139 aux.acc_seg: 98.4636 +04/17 09:52:37 - mmengine - INFO - Iter(train) [ 4850/160000] base_lr: 9.7887e-05 lr: 3.6191e-07 eta: 1 day, 19:10:46 time: 0.9983 data_time: 0.0041 memory: 8462 loss: 0.0354 decode.loss_ce: 0.0206 decode.acc_seg: 99.3814 aux.loss_ce: 0.0148 aux.acc_seg: 98.8951 +04/17 09:53:27 - mmengine - INFO - Iter(train) [ 4900/160000] base_lr: 9.7856e-05 lr: 3.6179e-07 eta: 1 day, 19:09:53 time: 1.0004 data_time: 0.0043 memory: 8462 loss: 0.0361 decode.loss_ce: 0.0214 decode.acc_seg: 98.6313 aux.loss_ce: 0.0147 aux.acc_seg: 98.2702 +04/17 09:54:17 - mmengine - INFO - Iter(train) [ 4950/160000] base_lr: 9.7824e-05 lr: 3.6168e-07 eta: 1 day, 19:08:59 time: 0.9996 data_time: 0.0043 memory: 8462 loss: 0.0347 decode.loss_ce: 0.0205 decode.acc_seg: 99.5302 aux.loss_ce: 0.0142 aux.acc_seg: 99.1716 +04/17 09:55:07 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 09:55:07 - mmengine - INFO - Iter(train) [ 5000/160000] base_lr: 9.7792e-05 lr: 3.6156e-07 eta: 1 day, 19:08:07 time: 1.0003 data_time: 0.0046 memory: 8462 loss: 0.0372 decode.loss_ce: 0.0226 decode.acc_seg: 99.1566 aux.loss_ce: 0.0146 aux.acc_seg: 98.8195 +04/17 09:55:57 - mmengine - INFO - Iter(train) [ 5050/160000] base_lr: 9.7761e-05 lr: 3.6144e-07 eta: 1 day, 19:07:15 time: 1.0014 data_time: 0.0046 memory: 8462 loss: 0.0359 decode.loss_ce: 0.0222 decode.acc_seg: 99.2836 aux.loss_ce: 0.0138 aux.acc_seg: 98.8277 +04/17 09:56:47 - mmengine - INFO - Iter(train) [ 5100/160000] base_lr: 9.7729e-05 lr: 3.6133e-07 eta: 1 day, 19:06:21 time: 0.9992 data_time: 0.0044 memory: 8462 loss: 0.0312 decode.loss_ce: 0.0186 decode.acc_seg: 99.5804 aux.loss_ce: 0.0127 aux.acc_seg: 99.2104 +04/17 09:57:37 - mmengine - INFO - Iter(train) [ 5150/160000] base_lr: 9.7698e-05 lr: 3.6121e-07 eta: 1 day, 19:05:29 time: 1.0001 data_time: 0.0047 memory: 8462 loss: 0.0433 decode.loss_ce: 0.0268 decode.acc_seg: 99.1711 aux.loss_ce: 0.0166 aux.acc_seg: 98.8379 +04/17 09:58:27 - mmengine - INFO - Iter(train) [ 5200/160000] base_lr: 9.7666e-05 lr: 3.6109e-07 eta: 1 day, 19:04:35 time: 0.9999 data_time: 0.0041 memory: 8462 loss: 0.0383 decode.loss_ce: 0.0227 decode.acc_seg: 99.3835 aux.loss_ce: 0.0156 aux.acc_seg: 98.8422 +04/17 09:59:17 - mmengine - INFO - Iter(train) [ 5250/160000] base_lr: 9.7635e-05 lr: 3.6098e-07 eta: 1 day, 19:03:44 time: 0.9997 data_time: 0.0043 memory: 8462 loss: 0.0380 decode.loss_ce: 0.0231 decode.acc_seg: 99.4253 aux.loss_ce: 0.0149 aux.acc_seg: 98.9494 +04/17 10:00:07 - mmengine - INFO - Iter(train) [ 5300/160000] base_lr: 9.7603e-05 lr: 3.6086e-07 eta: 1 day, 19:02:51 time: 1.0010 data_time: 0.0041 memory: 8462 loss: 0.0373 decode.loss_ce: 0.0229 decode.acc_seg: 98.6557 aux.loss_ce: 0.0145 aux.acc_seg: 98.1289 +04/17 10:00:57 - mmengine - INFO - Iter(train) [ 5350/160000] base_lr: 9.7572e-05 lr: 3.6074e-07 eta: 1 day, 19:01:58 time: 1.0000 data_time: 0.0040 memory: 8462 loss: 0.0357 decode.loss_ce: 0.0208 decode.acc_seg: 98.8726 aux.loss_ce: 0.0148 aux.acc_seg: 98.3179 +04/17 10:01:47 - mmengine - INFO - Iter(train) [ 5400/160000] base_lr: 9.7540e-05 lr: 3.6063e-07 eta: 1 day, 19:01:06 time: 1.0000 data_time: 0.0042 memory: 8462 loss: 0.0377 decode.loss_ce: 0.0226 decode.acc_seg: 99.2432 aux.loss_ce: 0.0151 aux.acc_seg: 98.8401 +04/17 10:02:37 - mmengine - INFO - Iter(train) [ 5450/160000] base_lr: 9.7509e-05 lr: 3.6051e-07 eta: 1 day, 19:00:14 time: 0.9995 data_time: 0.0044 memory: 8462 loss: 0.0334 decode.loss_ce: 0.0194 decode.acc_seg: 99.0490 aux.loss_ce: 0.0141 aux.acc_seg: 98.4489 +04/17 10:03:27 - mmengine - INFO - Iter(train) [ 5500/160000] base_lr: 9.7477e-05 lr: 3.6039e-07 eta: 1 day, 18:59:21 time: 0.9999 data_time: 0.0045 memory: 8462 loss: 0.0403 decode.loss_ce: 0.0243 decode.acc_seg: 99.5821 aux.loss_ce: 0.0160 aux.acc_seg: 98.7394 +04/17 10:04:17 - mmengine - INFO - Iter(train) [ 5550/160000] base_lr: 9.7445e-05 lr: 3.6028e-07 eta: 1 day, 18:58:29 time: 1.0009 data_time: 0.0042 memory: 8462 loss: 0.0361 decode.loss_ce: 0.0219 decode.acc_seg: 99.4492 aux.loss_ce: 0.0143 aux.acc_seg: 98.9521 +04/17 10:05:07 - mmengine - INFO - Iter(train) [ 5600/160000] base_lr: 9.7414e-05 lr: 3.6016e-07 eta: 1 day, 18:57:36 time: 1.0006 data_time: 0.0040 memory: 8462 loss: 0.0342 decode.loss_ce: 0.0203 decode.acc_seg: 99.4709 aux.loss_ce: 0.0139 aux.acc_seg: 99.1426 +04/17 10:05:57 - mmengine - INFO - Iter(train) [ 5650/160000] base_lr: 9.7382e-05 lr: 3.6004e-07 eta: 1 day, 18:56:45 time: 0.9997 data_time: 0.0045 memory: 8462 loss: 0.0376 decode.loss_ce: 0.0223 decode.acc_seg: 98.4533 aux.loss_ce: 0.0153 aux.acc_seg: 97.8525 +04/17 10:06:47 - mmengine - INFO - Iter(train) [ 5700/160000] base_lr: 9.7351e-05 lr: 3.5993e-07 eta: 1 day, 18:55:53 time: 1.0012 data_time: 0.0042 memory: 8462 loss: 0.0358 decode.loss_ce: 0.0211 decode.acc_seg: 99.3319 aux.loss_ce: 0.0148 aux.acc_seg: 98.8008 +04/17 10:07:37 - mmengine - INFO - Iter(train) [ 5750/160000] base_lr: 9.7319e-05 lr: 3.5981e-07 eta: 1 day, 18:55:01 time: 0.9993 data_time: 0.0043 memory: 8462 loss: 0.0328 decode.loss_ce: 0.0194 decode.acc_seg: 99.6233 aux.loss_ce: 0.0133 aux.acc_seg: 98.9773 +04/17 10:08:27 - mmengine - INFO - Iter(train) [ 5800/160000] base_lr: 9.7288e-05 lr: 3.5969e-07 eta: 1 day, 18:54:10 time: 1.0010 data_time: 0.0042 memory: 8462 loss: 0.0342 decode.loss_ce: 0.0208 decode.acc_seg: 99.3694 aux.loss_ce: 0.0133 aux.acc_seg: 99.0971 +04/17 10:09:17 - mmengine - INFO - Iter(train) [ 5850/160000] base_lr: 9.7256e-05 lr: 3.5958e-07 eta: 1 day, 18:53:18 time: 0.9987 data_time: 0.0044 memory: 8462 loss: 0.0315 decode.loss_ce: 0.0187 decode.acc_seg: 99.3813 aux.loss_ce: 0.0128 aux.acc_seg: 99.0602 +04/17 10:10:07 - mmengine - INFO - Iter(train) [ 5900/160000] base_lr: 9.7225e-05 lr: 3.5946e-07 eta: 1 day, 18:52:26 time: 1.0002 data_time: 0.0042 memory: 8462 loss: 0.0377 decode.loss_ce: 0.0230 decode.acc_seg: 99.0547 aux.loss_ce: 0.0146 aux.acc_seg: 99.2113 +04/17 10:10:57 - mmengine - INFO - Iter(train) [ 5950/160000] base_lr: 9.7193e-05 lr: 3.5934e-07 eta: 1 day, 18:51:35 time: 1.0009 data_time: 0.0045 memory: 8462 loss: 0.0346 decode.loss_ce: 0.0212 decode.acc_seg: 99.1407 aux.loss_ce: 0.0134 aux.acc_seg: 98.7675 +04/17 10:11:47 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 10:11:47 - mmengine - INFO - Iter(train) [ 6000/160000] base_lr: 9.7161e-05 lr: 3.5923e-07 eta: 1 day, 18:50:43 time: 0.9991 data_time: 0.0042 memory: 8462 loss: 0.0308 decode.loss_ce: 0.0179 decode.acc_seg: 99.2178 aux.loss_ce: 0.0129 aux.acc_seg: 98.6378 +04/17 10:12:37 - mmengine - INFO - Iter(train) [ 6050/160000] base_lr: 9.7130e-05 lr: 3.5911e-07 eta: 1 day, 18:49:51 time: 1.0007 data_time: 0.0044 memory: 8462 loss: 0.0298 decode.loss_ce: 0.0170 decode.acc_seg: 99.5426 aux.loss_ce: 0.0128 aux.acc_seg: 98.9246 +04/17 10:13:27 - mmengine - INFO - Iter(train) [ 6100/160000] base_lr: 9.7098e-05 lr: 3.5899e-07 eta: 1 day, 18:49:00 time: 0.9998 data_time: 0.0043 memory: 8462 loss: 0.0350 decode.loss_ce: 0.0213 decode.acc_seg: 99.6117 aux.loss_ce: 0.0137 aux.acc_seg: 99.2361 +04/17 10:14:17 - mmengine - INFO - Iter(train) [ 6150/160000] base_lr: 9.7067e-05 lr: 3.5888e-07 eta: 1 day, 18:48:09 time: 1.0003 data_time: 0.0042 memory: 8462 loss: 0.0323 decode.loss_ce: 0.0194 decode.acc_seg: 99.6401 aux.loss_ce: 0.0129 aux.acc_seg: 99.0118 +04/17 10:15:07 - mmengine - INFO - Iter(train) [ 6200/160000] base_lr: 9.7035e-05 lr: 3.5876e-07 eta: 1 day, 18:47:18 time: 1.0015 data_time: 0.0045 memory: 8462 loss: 0.0245 decode.loss_ce: 0.0145 decode.acc_seg: 99.0974 aux.loss_ce: 0.0099 aux.acc_seg: 98.7240 +04/17 10:15:57 - mmengine - INFO - Iter(train) [ 6250/160000] base_lr: 9.7004e-05 lr: 3.5864e-07 eta: 1 day, 18:46:25 time: 1.0000 data_time: 0.0047 memory: 8462 loss: 0.0372 decode.loss_ce: 0.0233 decode.acc_seg: 99.3490 aux.loss_ce: 0.0139 aux.acc_seg: 98.9359 +04/17 10:16:47 - mmengine - INFO - Iter(train) [ 6300/160000] base_lr: 9.6972e-05 lr: 3.5853e-07 eta: 1 day, 18:45:34 time: 1.0004 data_time: 0.0047 memory: 8462 loss: 0.0360 decode.loss_ce: 0.0216 decode.acc_seg: 99.5167 aux.loss_ce: 0.0144 aux.acc_seg: 99.0215 +04/17 10:17:37 - mmengine - INFO - Iter(train) [ 6350/160000] base_lr: 9.6941e-05 lr: 3.5841e-07 eta: 1 day, 18:44:43 time: 1.0010 data_time: 0.0049 memory: 8462 loss: 0.0320 decode.loss_ce: 0.0187 decode.acc_seg: 99.2607 aux.loss_ce: 0.0133 aux.acc_seg: 98.5270 +04/17 10:18:27 - mmengine - INFO - Iter(train) [ 6400/160000] base_lr: 9.6909e-05 lr: 3.5829e-07 eta: 1 day, 18:43:52 time: 1.0014 data_time: 0.0045 memory: 8462 loss: 0.0435 decode.loss_ce: 0.0272 decode.acc_seg: 98.5983 aux.loss_ce: 0.0163 aux.acc_seg: 98.1480 +04/17 10:19:17 - mmengine - INFO - Iter(train) [ 6450/160000] base_lr: 9.6878e-05 lr: 3.5818e-07 eta: 1 day, 18:42:59 time: 0.9994 data_time: 0.0048 memory: 8462 loss: 0.0364 decode.loss_ce: 0.0225 decode.acc_seg: 99.4532 aux.loss_ce: 0.0139 aux.acc_seg: 98.9351 +04/17 10:20:07 - mmengine - INFO - Iter(train) [ 6500/160000] base_lr: 9.6846e-05 lr: 3.5806e-07 eta: 1 day, 18:42:08 time: 1.0001 data_time: 0.0046 memory: 8462 loss: 0.0339 decode.loss_ce: 0.0201 decode.acc_seg: 99.1266 aux.loss_ce: 0.0138 aux.acc_seg: 98.5552 +04/17 10:20:57 - mmengine - INFO - Iter(train) [ 6550/160000] base_lr: 9.6814e-05 lr: 3.5794e-07 eta: 1 day, 18:41:16 time: 0.9999 data_time: 0.0044 memory: 8462 loss: 0.0331 decode.loss_ce: 0.0196 decode.acc_seg: 98.7511 aux.loss_ce: 0.0135 aux.acc_seg: 98.2067 +04/17 10:21:47 - mmengine - INFO - Iter(train) [ 6600/160000] base_lr: 9.6783e-05 lr: 3.5783e-07 eta: 1 day, 18:40:24 time: 0.9997 data_time: 0.0042 memory: 8462 loss: 0.0330 decode.loss_ce: 0.0203 decode.acc_seg: 99.3629 aux.loss_ce: 0.0127 aux.acc_seg: 98.8394 +04/17 10:22:37 - mmengine - INFO - Iter(train) [ 6650/160000] base_lr: 9.6751e-05 lr: 3.5771e-07 eta: 1 day, 18:39:32 time: 0.9991 data_time: 0.0041 memory: 8462 loss: 0.0314 decode.loss_ce: 0.0184 decode.acc_seg: 99.6651 aux.loss_ce: 0.0130 aux.acc_seg: 99.2207 +04/17 10:23:27 - mmengine - INFO - Iter(train) [ 6700/160000] base_lr: 9.6720e-05 lr: 3.5759e-07 eta: 1 day, 18:38:41 time: 1.0009 data_time: 0.0047 memory: 8462 loss: 0.0288 decode.loss_ce: 0.0163 decode.acc_seg: 99.4678 aux.loss_ce: 0.0125 aux.acc_seg: 98.8937 +04/17 10:24:17 - mmengine - INFO - Iter(train) [ 6750/160000] base_lr: 9.6688e-05 lr: 3.5748e-07 eta: 1 day, 18:37:50 time: 1.0009 data_time: 0.0044 memory: 8462 loss: 0.0323 decode.loss_ce: 0.0190 decode.acc_seg: 99.3738 aux.loss_ce: 0.0133 aux.acc_seg: 98.7343 +04/17 10:25:07 - mmengine - INFO - Iter(train) [ 6800/160000] base_lr: 9.6657e-05 lr: 3.5736e-07 eta: 1 day, 18:36:59 time: 1.0006 data_time: 0.0046 memory: 8462 loss: 0.0381 decode.loss_ce: 0.0228 decode.acc_seg: 99.1611 aux.loss_ce: 0.0153 aux.acc_seg: 98.4055 +04/17 10:25:57 - mmengine - INFO - Iter(train) [ 6850/160000] base_lr: 9.6625e-05 lr: 3.5724e-07 eta: 1 day, 18:36:07 time: 0.9994 data_time: 0.0043 memory: 8462 loss: 0.0337 decode.loss_ce: 0.0199 decode.acc_seg: 99.4860 aux.loss_ce: 0.0138 aux.acc_seg: 98.8110 +04/17 10:26:47 - mmengine - INFO - Iter(train) [ 6900/160000] base_lr: 9.6594e-05 lr: 3.5713e-07 eta: 1 day, 18:35:16 time: 1.0015 data_time: 0.0045 memory: 8462 loss: 0.0276 decode.loss_ce: 0.0161 decode.acc_seg: 99.5693 aux.loss_ce: 0.0116 aux.acc_seg: 99.0717 +04/17 10:27:37 - mmengine - INFO - Iter(train) [ 6950/160000] base_lr: 9.6562e-05 lr: 3.5701e-07 eta: 1 day, 18:34:25 time: 1.0019 data_time: 0.0044 memory: 8462 loss: 0.0346 decode.loss_ce: 0.0203 decode.acc_seg: 98.7282 aux.loss_ce: 0.0143 aux.acc_seg: 97.7823 +04/17 10:28:27 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 10:28:27 - mmengine - INFO - Iter(train) [ 7000/160000] base_lr: 9.6531e-05 lr: 3.5689e-07 eta: 1 day, 18:33:33 time: 0.9999 data_time: 0.0044 memory: 8462 loss: 0.0356 decode.loss_ce: 0.0214 decode.acc_seg: 99.3029 aux.loss_ce: 0.0143 aux.acc_seg: 98.7858 +04/17 10:29:17 - mmengine - INFO - Iter(train) [ 7050/160000] base_lr: 9.6499e-05 lr: 3.5678e-07 eta: 1 day, 18:32:42 time: 0.9999 data_time: 0.0045 memory: 8462 loss: 0.0337 decode.loss_ce: 0.0205 decode.acc_seg: 99.5481 aux.loss_ce: 0.0132 aux.acc_seg: 99.1684 +04/17 10:30:07 - mmengine - INFO - Iter(train) [ 7100/160000] base_lr: 9.6467e-05 lr: 3.5666e-07 eta: 1 day, 18:31:51 time: 1.0014 data_time: 0.0043 memory: 8462 loss: 0.0319 decode.loss_ce: 0.0187 decode.acc_seg: 99.1455 aux.loss_ce: 0.0132 aux.acc_seg: 98.8676 +04/17 10:30:57 - mmengine - INFO - Iter(train) [ 7150/160000] base_lr: 9.6436e-05 lr: 3.5654e-07 eta: 1 day, 18:31:00 time: 1.0008 data_time: 0.0043 memory: 8462 loss: 0.0319 decode.loss_ce: 0.0187 decode.acc_seg: 98.9529 aux.loss_ce: 0.0131 aux.acc_seg: 98.1979 +04/17 10:31:47 - mmengine - INFO - Iter(train) [ 7200/160000] base_lr: 9.6404e-05 lr: 3.5643e-07 eta: 1 day, 18:30:09 time: 1.0000 data_time: 0.0047 memory: 8462 loss: 0.0312 decode.loss_ce: 0.0187 decode.acc_seg: 99.6557 aux.loss_ce: 0.0125 aux.acc_seg: 99.3261 +04/17 10:32:37 - mmengine - INFO - Iter(train) [ 7250/160000] base_lr: 9.6373e-05 lr: 3.5631e-07 eta: 1 day, 18:29:18 time: 1.0010 data_time: 0.0047 memory: 8462 loss: 0.0336 decode.loss_ce: 0.0197 decode.acc_seg: 98.9107 aux.loss_ce: 0.0139 aux.acc_seg: 98.4201 +04/17 10:33:27 - mmengine - INFO - Iter(train) [ 7300/160000] base_lr: 9.6341e-05 lr: 3.5619e-07 eta: 1 day, 18:28:28 time: 1.0011 data_time: 0.0045 memory: 8462 loss: 0.0378 decode.loss_ce: 0.0227 decode.acc_seg: 98.6927 aux.loss_ce: 0.0151 aux.acc_seg: 98.0427 +04/17 10:34:18 - mmengine - INFO - Iter(train) [ 7350/160000] base_lr: 9.6310e-05 lr: 3.5608e-07 eta: 1 day, 18:27:37 time: 1.0004 data_time: 0.0045 memory: 8462 loss: 0.0358 decode.loss_ce: 0.0215 decode.acc_seg: 99.4621 aux.loss_ce: 0.0142 aux.acc_seg: 98.7711 +04/17 10:35:08 - mmengine - INFO - Iter(train) [ 7400/160000] base_lr: 9.6278e-05 lr: 3.5596e-07 eta: 1 day, 18:26:46 time: 1.0027 data_time: 0.0047 memory: 8462 loss: 0.0302 decode.loss_ce: 0.0181 decode.acc_seg: 99.4499 aux.loss_ce: 0.0122 aux.acc_seg: 99.2104 +04/17 10:35:58 - mmengine - INFO - Iter(train) [ 7450/160000] base_lr: 9.6247e-05 lr: 3.5584e-07 eta: 1 day, 18:25:55 time: 1.0011 data_time: 0.0042 memory: 8462 loss: 0.0323 decode.loss_ce: 0.0190 decode.acc_seg: 98.8752 aux.loss_ce: 0.0133 aux.acc_seg: 98.3652 +04/17 10:36:48 - mmengine - INFO - Iter(train) [ 7500/160000] base_lr: 9.6215e-05 lr: 3.5573e-07 eta: 1 day, 18:25:05 time: 1.0014 data_time: 0.0042 memory: 8462 loss: 0.0277 decode.loss_ce: 0.0164 decode.acc_seg: 99.1823 aux.loss_ce: 0.0114 aux.acc_seg: 98.7297 +04/17 10:37:38 - mmengine - INFO - Iter(train) [ 7550/160000] base_lr: 9.6184e-05 lr: 3.5561e-07 eta: 1 day, 18:24:15 time: 1.0003 data_time: 0.0039 memory: 8462 loss: 0.0301 decode.loss_ce: 0.0178 decode.acc_seg: 99.1760 aux.loss_ce: 0.0123 aux.acc_seg: 98.8791 +04/17 10:38:28 - mmengine - INFO - Iter(train) [ 7600/160000] base_lr: 9.6152e-05 lr: 3.5549e-07 eta: 1 day, 18:23:24 time: 0.9997 data_time: 0.0040 memory: 8462 loss: 0.0308 decode.loss_ce: 0.0181 decode.acc_seg: 99.1236 aux.loss_ce: 0.0127 aux.acc_seg: 98.4756 +04/17 10:39:18 - mmengine - INFO - Iter(train) [ 7650/160000] base_lr: 9.6120e-05 lr: 3.5538e-07 eta: 1 day, 18:22:33 time: 1.0000 data_time: 0.0039 memory: 8462 loss: 0.0333 decode.loss_ce: 0.0200 decode.acc_seg: 99.0524 aux.loss_ce: 0.0133 aux.acc_seg: 98.4222 +04/17 10:40:08 - mmengine - INFO - Iter(train) [ 7700/160000] base_lr: 9.6089e-05 lr: 3.5526e-07 eta: 1 day, 18:21:42 time: 0.9994 data_time: 0.0040 memory: 8462 loss: 0.0285 decode.loss_ce: 0.0160 decode.acc_seg: 99.4146 aux.loss_ce: 0.0125 aux.acc_seg: 98.5973 +04/17 10:40:58 - mmengine - INFO - Iter(train) [ 7750/160000] base_lr: 9.6057e-05 lr: 3.5514e-07 eta: 1 day, 18:20:51 time: 1.0002 data_time: 0.0047 memory: 8462 loss: 0.0324 decode.loss_ce: 0.0193 decode.acc_seg: 99.3982 aux.loss_ce: 0.0131 aux.acc_seg: 98.7888 +04/17 10:41:48 - mmengine - INFO - Iter(train) [ 7800/160000] base_lr: 9.6026e-05 lr: 3.5503e-07 eta: 1 day, 18:20:00 time: 1.0006 data_time: 0.0040 memory: 8462 loss: 0.0360 decode.loss_ce: 0.0224 decode.acc_seg: 99.0635 aux.loss_ce: 0.0136 aux.acc_seg: 98.8470 +04/17 10:42:38 - mmengine - INFO - Iter(train) [ 7850/160000] base_lr: 9.5994e-05 lr: 3.5491e-07 eta: 1 day, 18:19:10 time: 1.0003 data_time: 0.0044 memory: 8462 loss: 0.0255 decode.loss_ce: 0.0145 decode.acc_seg: 99.1734 aux.loss_ce: 0.0110 aux.acc_seg: 98.4362 +04/17 10:43:28 - mmengine - INFO - Iter(train) [ 7900/160000] base_lr: 9.5963e-05 lr: 3.5479e-07 eta: 1 day, 18:18:19 time: 0.9994 data_time: 0.0042 memory: 8462 loss: 0.0285 decode.loss_ce: 0.0167 decode.acc_seg: 99.1795 aux.loss_ce: 0.0118 aux.acc_seg: 98.7524 +04/17 10:44:18 - mmengine - INFO - Iter(train) [ 7950/160000] base_lr: 9.5931e-05 lr: 3.5468e-07 eta: 1 day, 18:17:27 time: 0.9990 data_time: 0.0043 memory: 8462 loss: 0.0262 decode.loss_ce: 0.0151 decode.acc_seg: 99.2907 aux.loss_ce: 0.0111 aux.acc_seg: 98.7049 +04/17 10:45:08 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 10:45:08 - mmengine - INFO - Iter(train) [ 8000/160000] base_lr: 9.5900e-05 lr: 3.5456e-07 eta: 1 day, 18:16:36 time: 1.0002 data_time: 0.0047 memory: 8462 loss: 0.0325 decode.loss_ce: 0.0191 decode.acc_seg: 99.4884 aux.loss_ce: 0.0134 aux.acc_seg: 98.7900 +04/17 10:45:58 - mmengine - INFO - Iter(train) [ 8050/160000] base_lr: 9.5868e-05 lr: 3.5444e-07 eta: 1 day, 18:15:44 time: 0.9992 data_time: 0.0045 memory: 8462 loss: 0.0257 decode.loss_ce: 0.0149 decode.acc_seg: 99.4730 aux.loss_ce: 0.0108 aux.acc_seg: 99.1138 +04/17 10:46:48 - mmengine - INFO - Iter(train) [ 8100/160000] base_lr: 9.5837e-05 lr: 3.5433e-07 eta: 1 day, 18:14:52 time: 0.9996 data_time: 0.0044 memory: 8462 loss: 0.0312 decode.loss_ce: 0.0180 decode.acc_seg: 98.8304 aux.loss_ce: 0.0131 aux.acc_seg: 97.3690 +04/17 10:47:38 - mmengine - INFO - Iter(train) [ 8150/160000] base_lr: 9.5805e-05 lr: 3.5421e-07 eta: 1 day, 18:14:01 time: 0.9994 data_time: 0.0043 memory: 8462 loss: 0.0284 decode.loss_ce: 0.0165 decode.acc_seg: 99.4873 aux.loss_ce: 0.0118 aux.acc_seg: 99.1148 +04/17 10:48:28 - mmengine - INFO - Iter(train) [ 8200/160000] base_lr: 9.5773e-05 lr: 3.5409e-07 eta: 1 day, 18:13:10 time: 1.0002 data_time: 0.0042 memory: 8462 loss: 0.0321 decode.loss_ce: 0.0184 decode.acc_seg: 99.5493 aux.loss_ce: 0.0136 aux.acc_seg: 99.0458 +04/17 10:49:18 - mmengine - INFO - Iter(train) [ 8250/160000] base_lr: 9.5742e-05 lr: 3.5398e-07 eta: 1 day, 18:12:18 time: 0.9994 data_time: 0.0042 memory: 8462 loss: 0.0264 decode.loss_ce: 0.0148 decode.acc_seg: 99.6235 aux.loss_ce: 0.0116 aux.acc_seg: 99.0587 +04/17 10:50:08 - mmengine - INFO - Iter(train) [ 8300/160000] base_lr: 9.5710e-05 lr: 3.5386e-07 eta: 1 day, 18:11:27 time: 0.9996 data_time: 0.0042 memory: 8462 loss: 0.0318 decode.loss_ce: 0.0187 decode.acc_seg: 99.6046 aux.loss_ce: 0.0132 aux.acc_seg: 99.2018 +04/17 10:50:58 - mmengine - INFO - Iter(train) [ 8350/160000] base_lr: 9.5679e-05 lr: 3.5374e-07 eta: 1 day, 18:10:35 time: 1.0004 data_time: 0.0048 memory: 8462 loss: 0.0312 decode.loss_ce: 0.0187 decode.acc_seg: 99.0238 aux.loss_ce: 0.0125 aux.acc_seg: 98.4592 +04/17 10:51:48 - mmengine - INFO - Iter(train) [ 8400/160000] base_lr: 9.5647e-05 lr: 3.5363e-07 eta: 1 day, 18:09:44 time: 0.9997 data_time: 0.0042 memory: 8462 loss: 0.0297 decode.loss_ce: 0.0168 decode.acc_seg: 99.4942 aux.loss_ce: 0.0129 aux.acc_seg: 99.0360 +04/17 10:52:38 - mmengine - INFO - Iter(train) [ 8450/160000] base_lr: 9.5616e-05 lr: 3.5351e-07 eta: 1 day, 18:08:53 time: 1.0005 data_time: 0.0047 memory: 8462 loss: 0.0286 decode.loss_ce: 0.0164 decode.acc_seg: 99.4204 aux.loss_ce: 0.0121 aux.acc_seg: 98.8857 +04/17 10:53:28 - mmengine - INFO - Iter(train) [ 8500/160000] base_lr: 9.5584e-05 lr: 3.5339e-07 eta: 1 day, 18:08:02 time: 1.0005 data_time: 0.0050 memory: 8462 loss: 0.0294 decode.loss_ce: 0.0174 decode.acc_seg: 99.3834 aux.loss_ce: 0.0120 aux.acc_seg: 98.7679 +04/17 10:54:18 - mmengine - INFO - Iter(train) [ 8550/160000] base_lr: 9.5553e-05 lr: 3.5328e-07 eta: 1 day, 18:07:11 time: 0.9991 data_time: 0.0042 memory: 8462 loss: 0.0322 decode.loss_ce: 0.0191 decode.acc_seg: 99.6374 aux.loss_ce: 0.0131 aux.acc_seg: 99.1426 +04/17 10:55:08 - mmengine - INFO - Iter(train) [ 8600/160000] base_lr: 9.5521e-05 lr: 3.5316e-07 eta: 1 day, 18:06:20 time: 0.9991 data_time: 0.0048 memory: 8462 loss: 0.0302 decode.loss_ce: 0.0176 decode.acc_seg: 99.0908 aux.loss_ce: 0.0126 aux.acc_seg: 98.6433 +04/17 10:55:58 - mmengine - INFO - Iter(train) [ 8650/160000] base_lr: 9.5490e-05 lr: 3.5304e-07 eta: 1 day, 18:05:28 time: 0.9985 data_time: 0.0044 memory: 8462 loss: 0.0315 decode.loss_ce: 0.0184 decode.acc_seg: 99.4024 aux.loss_ce: 0.0130 aux.acc_seg: 98.8716 +04/17 10:56:48 - mmengine - INFO - Iter(train) [ 8700/160000] base_lr: 9.5458e-05 lr: 3.5293e-07 eta: 1 day, 18:04:36 time: 1.0000 data_time: 0.0043 memory: 8462 loss: 0.0312 decode.loss_ce: 0.0186 decode.acc_seg: 99.1198 aux.loss_ce: 0.0126 aux.acc_seg: 98.9529 +04/17 10:57:38 - mmengine - INFO - Iter(train) [ 8750/160000] base_lr: 9.5426e-05 lr: 3.5281e-07 eta: 1 day, 18:03:45 time: 0.9992 data_time: 0.0045 memory: 8462 loss: 0.0291 decode.loss_ce: 0.0169 decode.acc_seg: 98.7726 aux.loss_ce: 0.0121 aux.acc_seg: 97.8586 +04/17 10:58:28 - mmengine - INFO - Iter(train) [ 8800/160000] base_lr: 9.5395e-05 lr: 3.5269e-07 eta: 1 day, 18:02:53 time: 0.9992 data_time: 0.0042 memory: 8462 loss: 0.0293 decode.loss_ce: 0.0173 decode.acc_seg: 99.5106 aux.loss_ce: 0.0119 aux.acc_seg: 99.0864 +04/17 10:59:18 - mmengine - INFO - Iter(train) [ 8850/160000] base_lr: 9.5363e-05 lr: 3.5258e-07 eta: 1 day, 18:02:02 time: 0.9989 data_time: 0.0044 memory: 8462 loss: 0.0286 decode.loss_ce: 0.0170 decode.acc_seg: 98.9819 aux.loss_ce: 0.0117 aux.acc_seg: 98.1556 +04/17 11:00:08 - mmengine - INFO - Iter(train) [ 8900/160000] base_lr: 9.5332e-05 lr: 3.5246e-07 eta: 1 day, 18:01:11 time: 0.9989 data_time: 0.0042 memory: 8462 loss: 0.0326 decode.loss_ce: 0.0199 decode.acc_seg: 99.2109 aux.loss_ce: 0.0127 aux.acc_seg: 98.8785 +04/17 11:00:58 - mmengine - INFO - Iter(train) [ 8950/160000] base_lr: 9.5300e-05 lr: 3.5234e-07 eta: 1 day, 18:00:19 time: 1.0001 data_time: 0.0045 memory: 8462 loss: 0.0297 decode.loss_ce: 0.0174 decode.acc_seg: 99.3111 aux.loss_ce: 0.0123 aux.acc_seg: 98.6567 +04/17 11:01:48 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 11:01:48 - mmengine - INFO - Iter(train) [ 9000/160000] base_lr: 9.5269e-05 lr: 3.5223e-07 eta: 1 day, 17:59:29 time: 1.0004 data_time: 0.0043 memory: 8462 loss: 0.0280 decode.loss_ce: 0.0168 decode.acc_seg: 99.1098 aux.loss_ce: 0.0111 aux.acc_seg: 98.6355 +04/17 11:02:38 - mmengine - INFO - Iter(train) [ 9050/160000] base_lr: 9.5237e-05 lr: 3.5211e-07 eta: 1 day, 17:58:37 time: 0.9995 data_time: 0.0041 memory: 8462 loss: 0.0279 decode.loss_ce: 0.0163 decode.acc_seg: 99.2556 aux.loss_ce: 0.0117 aux.acc_seg: 98.5273 +04/17 11:03:28 - mmengine - INFO - Iter(train) [ 9100/160000] base_lr: 9.5206e-05 lr: 3.5199e-07 eta: 1 day, 17:57:46 time: 0.9984 data_time: 0.0045 memory: 8462 loss: 0.0283 decode.loss_ce: 0.0171 decode.acc_seg: 99.2605 aux.loss_ce: 0.0112 aux.acc_seg: 98.7671 +04/17 11:04:18 - mmengine - INFO - Iter(train) [ 9150/160000] base_lr: 9.5174e-05 lr: 3.5188e-07 eta: 1 day, 17:56:54 time: 0.9985 data_time: 0.0044 memory: 8462 loss: 0.0292 decode.loss_ce: 0.0174 decode.acc_seg: 99.2044 aux.loss_ce: 0.0118 aux.acc_seg: 98.6469 +04/17 11:05:08 - mmengine - INFO - Iter(train) [ 9200/160000] base_lr: 9.5143e-05 lr: 3.5176e-07 eta: 1 day, 17:56:03 time: 0.9999 data_time: 0.0041 memory: 8462 loss: 0.0288 decode.loss_ce: 0.0166 decode.acc_seg: 99.1257 aux.loss_ce: 0.0122 aux.acc_seg: 98.5540 +04/17 11:05:58 - mmengine - INFO - Iter(train) [ 9250/160000] base_lr: 9.5111e-05 lr: 3.5164e-07 eta: 1 day, 17:55:12 time: 0.9992 data_time: 0.0044 memory: 8462 loss: 0.0361 decode.loss_ce: 0.0221 decode.acc_seg: 99.4940 aux.loss_ce: 0.0140 aux.acc_seg: 99.1802 +04/17 11:06:47 - mmengine - INFO - Iter(train) [ 9300/160000] base_lr: 9.5079e-05 lr: 3.5153e-07 eta: 1 day, 17:54:20 time: 0.9994 data_time: 0.0041 memory: 8462 loss: 0.0266 decode.loss_ce: 0.0153 decode.acc_seg: 99.6422 aux.loss_ce: 0.0114 aux.acc_seg: 99.3658 +04/17 11:07:37 - mmengine - INFO - Iter(train) [ 9350/160000] base_lr: 9.5048e-05 lr: 3.5141e-07 eta: 1 day, 17:53:29 time: 0.9984 data_time: 0.0042 memory: 8462 loss: 0.0290 decode.loss_ce: 0.0168 decode.acc_seg: 99.5070 aux.loss_ce: 0.0122 aux.acc_seg: 99.0034 +04/17 11:08:27 - mmengine - INFO - Iter(train) [ 9400/160000] base_lr: 9.5016e-05 lr: 3.5130e-07 eta: 1 day, 17:52:37 time: 0.9990 data_time: 0.0046 memory: 8462 loss: 0.0286 decode.loss_ce: 0.0172 decode.acc_seg: 99.3139 aux.loss_ce: 0.0114 aux.acc_seg: 99.0477 +04/17 11:09:17 - mmengine - INFO - Iter(train) [ 9450/160000] base_lr: 9.4985e-05 lr: 3.5118e-07 eta: 1 day, 17:51:46 time: 0.9999 data_time: 0.0042 memory: 8462 loss: 0.0280 decode.loss_ce: 0.0161 decode.acc_seg: 99.5781 aux.loss_ce: 0.0120 aux.acc_seg: 99.3572 +04/17 11:10:07 - mmengine - INFO - Iter(train) [ 9500/160000] base_lr: 9.4953e-05 lr: 3.5106e-07 eta: 1 day, 17:50:55 time: 0.9997 data_time: 0.0042 memory: 8462 loss: 0.0311 decode.loss_ce: 0.0183 decode.acc_seg: 99.4741 aux.loss_ce: 0.0128 aux.acc_seg: 98.6847 +04/17 11:10:57 - mmengine - INFO - Iter(train) [ 9550/160000] base_lr: 9.4922e-05 lr: 3.5095e-07 eta: 1 day, 17:50:04 time: 0.9996 data_time: 0.0047 memory: 8462 loss: 0.0341 decode.loss_ce: 0.0208 decode.acc_seg: 99.3027 aux.loss_ce: 0.0133 aux.acc_seg: 98.7852 +04/17 11:11:47 - mmengine - INFO - Iter(train) [ 9600/160000] base_lr: 9.4890e-05 lr: 3.5083e-07 eta: 1 day, 17:49:12 time: 0.9985 data_time: 0.0044 memory: 8462 loss: 0.0271 decode.loss_ce: 0.0162 decode.acc_seg: 99.4505 aux.loss_ce: 0.0109 aux.acc_seg: 98.7850 +04/17 11:12:37 - mmengine - INFO - Iter(train) [ 9650/160000] base_lr: 9.4859e-05 lr: 3.5071e-07 eta: 1 day, 17:48:21 time: 0.9994 data_time: 0.0044 memory: 8462 loss: 0.0422 decode.loss_ce: 0.0263 decode.acc_seg: 97.1151 aux.loss_ce: 0.0158 aux.acc_seg: 97.0732 +04/17 11:13:27 - mmengine - INFO - Iter(train) [ 9700/160000] base_lr: 9.4827e-05 lr: 3.5060e-07 eta: 1 day, 17:47:29 time: 0.9986 data_time: 0.0044 memory: 8462 loss: 0.0245 decode.loss_ce: 0.0143 decode.acc_seg: 99.3326 aux.loss_ce: 0.0102 aux.acc_seg: 98.6275 +04/17 11:14:17 - mmengine - INFO - Iter(train) [ 9750/160000] base_lr: 9.4796e-05 lr: 3.5048e-07 eta: 1 day, 17:46:38 time: 0.9988 data_time: 0.0042 memory: 8462 loss: 0.0313 decode.loss_ce: 0.0178 decode.acc_seg: 99.2683 aux.loss_ce: 0.0135 aux.acc_seg: 98.4152 +04/17 11:15:07 - mmengine - INFO - Iter(train) [ 9800/160000] base_lr: 9.4764e-05 lr: 3.5036e-07 eta: 1 day, 17:45:46 time: 0.9999 data_time: 0.0042 memory: 8462 loss: 0.0285 decode.loss_ce: 0.0174 decode.acc_seg: 99.5544 aux.loss_ce: 0.0111 aux.acc_seg: 99.2006 +04/17 11:15:57 - mmengine - INFO - Iter(train) [ 9850/160000] base_lr: 9.4732e-05 lr: 3.5025e-07 eta: 1 day, 17:44:55 time: 0.9983 data_time: 0.0043 memory: 8462 loss: 0.0267 decode.loss_ce: 0.0149 decode.acc_seg: 99.2363 aux.loss_ce: 0.0118 aux.acc_seg: 98.6767 +04/17 11:16:47 - mmengine - INFO - Iter(train) [ 9900/160000] base_lr: 9.4701e-05 lr: 3.5013e-07 eta: 1 day, 17:44:03 time: 0.9983 data_time: 0.0044 memory: 8462 loss: 0.0267 decode.loss_ce: 0.0154 decode.acc_seg: 98.6429 aux.loss_ce: 0.0113 aux.acc_seg: 98.1266 +04/17 11:17:37 - mmengine - INFO - Iter(train) [ 9950/160000] base_lr: 9.4669e-05 lr: 3.5001e-07 eta: 1 day, 17:43:11 time: 0.9989 data_time: 0.0045 memory: 8462 loss: 0.0329 decode.loss_ce: 0.0202 decode.acc_seg: 99.0934 aux.loss_ce: 0.0127 aux.acc_seg: 98.5346 +04/17 11:18:27 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 11:18:27 - mmengine - INFO - Iter(train) [ 10000/160000] base_lr: 9.4638e-05 lr: 3.4990e-07 eta: 1 day, 17:42:20 time: 0.9986 data_time: 0.0045 memory: 8462 loss: 0.0318 decode.loss_ce: 0.0197 decode.acc_seg: 99.1579 aux.loss_ce: 0.0122 aux.acc_seg: 98.8224 +04/17 11:18:27 - mmengine - INFO - Saving checkpoint at 10000 iterations +04/17 11:18:38 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:21 time: 0.1155 data_time: 0.0014 memory: 7125 +04/17 11:18:44 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:13 time: 0.1156 data_time: 0.0014 memory: 4004 +04/17 11:18:50 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:06 time: 0.1157 data_time: 0.0014 memory: 4004 +04/17 11:18:56 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.1154 data_time: 0.0012 memory: 4004 +04/17 11:18:56 - mmengine - INFO - per class results: +04/17 11:18:56 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 99.08 | 99.54 | 99.54 | 99.54 | 99.54 | +| contrast | 80.1 | 89.03 | 88.95 | 88.87 | 89.03 | ++------------+-------+-------+--------+-----------+--------+ +04/17 11:18:56 - mmengine - INFO - Iter(val) [200/200] aAcc: 99.1200 mIoU: 89.5900 mAcc: 94.2800 mFscore: 94.2400 mPrecision: 94.2100 mRecall: 94.2800 data_time: 0.0025 time: 0.1230 +04/17 11:19:46 - mmengine - INFO - Iter(train) [ 10050/160000] base_lr: 9.4606e-05 lr: 3.4978e-07 eta: 1 day, 17:41:31 time: 0.9982 data_time: 0.0045 memory: 8462 loss: 0.0228 decode.loss_ce: 0.0127 decode.acc_seg: 99.5184 aux.loss_ce: 0.0101 aux.acc_seg: 98.6654 +04/17 11:20:36 - mmengine - INFO - Iter(train) [ 10100/160000] base_lr: 9.4575e-05 lr: 3.4966e-07 eta: 1 day, 17:40:38 time: 0.9976 data_time: 0.0043 memory: 8462 loss: 0.0265 decode.loss_ce: 0.0152 decode.acc_seg: 99.2083 aux.loss_ce: 0.0113 aux.acc_seg: 98.8071 +04/17 11:21:26 - mmengine - INFO - Iter(train) [ 10150/160000] base_lr: 9.4543e-05 lr: 3.4955e-07 eta: 1 day, 17:39:46 time: 0.9978 data_time: 0.0042 memory: 8462 loss: 0.0320 decode.loss_ce: 0.0193 decode.acc_seg: 99.5743 aux.loss_ce: 0.0127 aux.acc_seg: 98.9920 +04/17 11:22:16 - mmengine - INFO - Iter(train) [ 10200/160000] base_lr: 9.4512e-05 lr: 3.4943e-07 eta: 1 day, 17:38:55 time: 0.9988 data_time: 0.0042 memory: 8462 loss: 0.0263 decode.loss_ce: 0.0156 decode.acc_seg: 99.2430 aux.loss_ce: 0.0108 aux.acc_seg: 98.4165 +04/17 11:23:05 - mmengine - INFO - Iter(train) [ 10250/160000] base_lr: 9.4480e-05 lr: 3.4931e-07 eta: 1 day, 17:38:03 time: 0.9993 data_time: 0.0048 memory: 8462 loss: 0.0263 decode.loss_ce: 0.0153 decode.acc_seg: 99.4232 aux.loss_ce: 0.0110 aux.acc_seg: 98.7839 +04/17 11:23:55 - mmengine - INFO - Iter(train) [ 10300/160000] base_lr: 9.4449e-05 lr: 3.4920e-07 eta: 1 day, 17:37:12 time: 0.9985 data_time: 0.0049 memory: 8462 loss: 0.0240 decode.loss_ce: 0.0137 decode.acc_seg: 99.3811 aux.loss_ce: 0.0103 aux.acc_seg: 98.7280 +04/17 11:24:45 - mmengine - INFO - Iter(train) [ 10350/160000] base_lr: 9.4417e-05 lr: 3.4908e-07 eta: 1 day, 17:36:20 time: 0.9986 data_time: 0.0046 memory: 8462 loss: 0.0265 decode.loss_ce: 0.0156 decode.acc_seg: 99.4926 aux.loss_ce: 0.0109 aux.acc_seg: 99.3280 +04/17 11:25:35 - mmengine - INFO - Iter(train) [ 10400/160000] base_lr: 9.4385e-05 lr: 3.4896e-07 eta: 1 day, 17:35:29 time: 0.9999 data_time: 0.0047 memory: 8462 loss: 0.0279 decode.loss_ce: 0.0161 decode.acc_seg: 99.0896 aux.loss_ce: 0.0118 aux.acc_seg: 98.7675 +04/17 11:26:25 - mmengine - INFO - Iter(train) [ 10450/160000] base_lr: 9.4354e-05 lr: 3.4885e-07 eta: 1 day, 17:34:38 time: 0.9985 data_time: 0.0045 memory: 8462 loss: 0.0307 decode.loss_ce: 0.0181 decode.acc_seg: 99.5108 aux.loss_ce: 0.0127 aux.acc_seg: 99.0753 +04/17 11:27:15 - mmengine - INFO - Iter(train) [ 10500/160000] base_lr: 9.4322e-05 lr: 3.4873e-07 eta: 1 day, 17:33:46 time: 0.9986 data_time: 0.0043 memory: 8462 loss: 0.0291 decode.loss_ce: 0.0175 decode.acc_seg: 99.3355 aux.loss_ce: 0.0116 aux.acc_seg: 98.9653 +04/17 11:28:05 - mmengine - INFO - Iter(train) [ 10550/160000] base_lr: 9.4291e-05 lr: 3.4861e-07 eta: 1 day, 17:32:55 time: 1.0003 data_time: 0.0043 memory: 8462 loss: 0.0286 decode.loss_ce: 0.0176 decode.acc_seg: 99.3546 aux.loss_ce: 0.0111 aux.acc_seg: 98.6609 +04/17 11:28:55 - mmengine - INFO - Iter(train) [ 10600/160000] base_lr: 9.4259e-05 lr: 3.4850e-07 eta: 1 day, 17:32:03 time: 0.9968 data_time: 0.0045 memory: 8462 loss: 0.0275 decode.loss_ce: 0.0156 decode.acc_seg: 99.4614 aux.loss_ce: 0.0119 aux.acc_seg: 98.6715 +04/17 11:29:45 - mmengine - INFO - Iter(train) [ 10650/160000] base_lr: 9.4228e-05 lr: 3.4838e-07 eta: 1 day, 17:31:11 time: 0.9983 data_time: 0.0043 memory: 8462 loss: 0.0283 decode.loss_ce: 0.0164 decode.acc_seg: 99.5975 aux.loss_ce: 0.0119 aux.acc_seg: 99.0761 +04/17 11:30:35 - mmengine - INFO - Iter(train) [ 10700/160000] base_lr: 9.4196e-05 lr: 3.4826e-07 eta: 1 day, 17:30:20 time: 0.9991 data_time: 0.0043 memory: 8462 loss: 0.0267 decode.loss_ce: 0.0154 decode.acc_seg: 99.5754 aux.loss_ce: 0.0113 aux.acc_seg: 99.1217 +04/17 11:31:25 - mmengine - INFO - Iter(train) [ 10750/160000] base_lr: 9.4165e-05 lr: 3.4815e-07 eta: 1 day, 17:29:28 time: 0.9996 data_time: 0.0049 memory: 8462 loss: 0.0224 decode.loss_ce: 0.0124 decode.acc_seg: 99.7696 aux.loss_ce: 0.0100 aux.acc_seg: 99.5125 +04/17 11:32:15 - mmengine - INFO - Iter(train) [ 10800/160000] base_lr: 9.4133e-05 lr: 3.4803e-07 eta: 1 day, 17:28:37 time: 0.9997 data_time: 0.0041 memory: 8462 loss: 0.0295 decode.loss_ce: 0.0172 decode.acc_seg: 99.4450 aux.loss_ce: 0.0124 aux.acc_seg: 98.9927 +04/17 11:33:05 - mmengine - INFO - Iter(train) [ 10850/160000] base_lr: 9.4102e-05 lr: 3.4791e-07 eta: 1 day, 17:27:46 time: 0.9995 data_time: 0.0043 memory: 8462 loss: 0.0328 decode.loss_ce: 0.0201 decode.acc_seg: 99.3118 aux.loss_ce: 0.0128 aux.acc_seg: 99.0999 +04/17 11:33:55 - mmengine - INFO - Iter(train) [ 10900/160000] base_lr: 9.4070e-05 lr: 3.4780e-07 eta: 1 day, 17:26:54 time: 0.9985 data_time: 0.0047 memory: 8462 loss: 0.0293 decode.loss_ce: 0.0176 decode.acc_seg: 99.5304 aux.loss_ce: 0.0117 aux.acc_seg: 99.0887 +04/17 11:34:45 - mmengine - INFO - Iter(train) [ 10950/160000] base_lr: 9.4038e-05 lr: 3.4768e-07 eta: 1 day, 17:26:03 time: 0.9986 data_time: 0.0042 memory: 8462 loss: 0.0242 decode.loss_ce: 0.0135 decode.acc_seg: 99.4892 aux.loss_ce: 0.0107 aux.acc_seg: 98.9141 +04/17 11:35:35 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 11:35:35 - mmengine - INFO - Iter(train) [ 11000/160000] base_lr: 9.4007e-05 lr: 3.4756e-07 eta: 1 day, 17:25:11 time: 0.9977 data_time: 0.0049 memory: 8462 loss: 0.0291 decode.loss_ce: 0.0168 decode.acc_seg: 99.1938 aux.loss_ce: 0.0123 aux.acc_seg: 98.6153 +04/17 11:36:25 - mmengine - INFO - Iter(train) [ 11050/160000] base_lr: 9.3975e-05 lr: 3.4745e-07 eta: 1 day, 17:24:20 time: 0.9982 data_time: 0.0048 memory: 8462 loss: 0.0313 decode.loss_ce: 0.0184 decode.acc_seg: 99.2914 aux.loss_ce: 0.0129 aux.acc_seg: 98.7675 +04/17 11:37:14 - mmengine - INFO - Iter(train) [ 11100/160000] base_lr: 9.3944e-05 lr: 3.4733e-07 eta: 1 day, 17:23:29 time: 0.9997 data_time: 0.0051 memory: 8462 loss: 0.0285 decode.loss_ce: 0.0160 decode.acc_seg: 99.6382 aux.loss_ce: 0.0124 aux.acc_seg: 99.2136 +04/17 11:38:04 - mmengine - INFO - Iter(train) [ 11150/160000] base_lr: 9.3912e-05 lr: 3.4721e-07 eta: 1 day, 17:22:37 time: 0.9979 data_time: 0.0044 memory: 8462 loss: 0.0252 decode.loss_ce: 0.0143 decode.acc_seg: 99.3307 aux.loss_ce: 0.0109 aux.acc_seg: 98.5315 +04/17 11:38:54 - mmengine - INFO - Iter(train) [ 11200/160000] base_lr: 9.3881e-05 lr: 3.4710e-07 eta: 1 day, 17:21:46 time: 0.9979 data_time: 0.0042 memory: 8462 loss: 0.0291 decode.loss_ce: 0.0178 decode.acc_seg: 99.6355 aux.loss_ce: 0.0113 aux.acc_seg: 99.2823 +04/17 11:39:44 - mmengine - INFO - Iter(train) [ 11250/160000] base_lr: 9.3849e-05 lr: 3.4698e-07 eta: 1 day, 17:20:54 time: 0.9981 data_time: 0.0045 memory: 8462 loss: 0.0241 decode.loss_ce: 0.0136 decode.acc_seg: 99.3332 aux.loss_ce: 0.0106 aux.acc_seg: 98.6628 +04/17 11:40:34 - mmengine - INFO - Iter(train) [ 11300/160000] base_lr: 9.3818e-05 lr: 3.4686e-07 eta: 1 day, 17:20:02 time: 0.9988 data_time: 0.0047 memory: 8462 loss: 0.0240 decode.loss_ce: 0.0133 decode.acc_seg: 99.3959 aux.loss_ce: 0.0107 aux.acc_seg: 98.8110 +04/17 11:41:24 - mmengine - INFO - Iter(train) [ 11350/160000] base_lr: 9.3786e-05 lr: 3.4675e-07 eta: 1 day, 17:19:10 time: 0.9985 data_time: 0.0041 memory: 8462 loss: 0.0247 decode.loss_ce: 0.0141 decode.acc_seg: 99.1455 aux.loss_ce: 0.0106 aux.acc_seg: 98.6660 +04/17 11:42:14 - mmengine - INFO - Iter(train) [ 11400/160000] base_lr: 9.3755e-05 lr: 3.4663e-07 eta: 1 day, 17:18:19 time: 0.9975 data_time: 0.0051 memory: 8462 loss: 0.0301 decode.loss_ce: 0.0177 decode.acc_seg: 99.1589 aux.loss_ce: 0.0124 aux.acc_seg: 98.4291 +04/17 11:43:04 - mmengine - INFO - Iter(train) [ 11450/160000] base_lr: 9.3723e-05 lr: 3.4651e-07 eta: 1 day, 17:17:27 time: 0.9990 data_time: 0.0048 memory: 8462 loss: 0.0251 decode.loss_ce: 0.0149 decode.acc_seg: 99.0963 aux.loss_ce: 0.0102 aux.acc_seg: 98.6616 +04/17 11:43:54 - mmengine - INFO - Iter(train) [ 11500/160000] base_lr: 9.3691e-05 lr: 3.4640e-07 eta: 1 day, 17:16:36 time: 1.0002 data_time: 0.0045 memory: 8462 loss: 0.0246 decode.loss_ce: 0.0141 decode.acc_seg: 99.2422 aux.loss_ce: 0.0105 aux.acc_seg: 98.6483 +04/17 11:44:44 - mmengine - INFO - Iter(train) [ 11550/160000] base_lr: 9.3660e-05 lr: 3.4628e-07 eta: 1 day, 17:15:45 time: 0.9990 data_time: 0.0043 memory: 8462 loss: 0.0245 decode.loss_ce: 0.0138 decode.acc_seg: 99.5718 aux.loss_ce: 0.0108 aux.acc_seg: 98.8657 +04/17 11:45:34 - mmengine - INFO - Iter(train) [ 11600/160000] base_lr: 9.3628e-05 lr: 3.4616e-07 eta: 1 day, 17:14:53 time: 0.9977 data_time: 0.0041 memory: 8462 loss: 0.0267 decode.loss_ce: 0.0157 decode.acc_seg: 99.7395 aux.loss_ce: 0.0110 aux.acc_seg: 99.3423 +04/17 11:46:24 - mmengine - INFO - Iter(train) [ 11650/160000] base_lr: 9.3597e-05 lr: 3.4605e-07 eta: 1 day, 17:14:02 time: 1.0000 data_time: 0.0046 memory: 8462 loss: 0.0245 decode.loss_ce: 0.0138 decode.acc_seg: 99.1066 aux.loss_ce: 0.0108 aux.acc_seg: 98.3194 +04/17 11:47:14 - mmengine - INFO - Iter(train) [ 11700/160000] base_lr: 9.3565e-05 lr: 3.4593e-07 eta: 1 day, 17:13:11 time: 0.9998 data_time: 0.0043 memory: 8462 loss: 0.0288 decode.loss_ce: 0.0169 decode.acc_seg: 98.5687 aux.loss_ce: 0.0119 aux.acc_seg: 98.0122 +04/17 11:48:03 - mmengine - INFO - Iter(train) [ 11750/160000] base_lr: 9.3534e-05 lr: 3.4581e-07 eta: 1 day, 17:12:19 time: 0.9987 data_time: 0.0043 memory: 8462 loss: 0.0266 decode.loss_ce: 0.0157 decode.acc_seg: 99.4154 aux.loss_ce: 0.0109 aux.acc_seg: 98.7604 +04/17 11:48:53 - mmengine - INFO - Iter(train) [ 11800/160000] base_lr: 9.3502e-05 lr: 3.4570e-07 eta: 1 day, 17:11:28 time: 0.9971 data_time: 0.0042 memory: 8462 loss: 0.0273 decode.loss_ce: 0.0153 decode.acc_seg: 99.6025 aux.loss_ce: 0.0120 aux.acc_seg: 99.0688 +04/17 11:49:43 - mmengine - INFO - Iter(train) [ 11850/160000] base_lr: 9.3471e-05 lr: 3.4558e-07 eta: 1 day, 17:10:36 time: 0.9984 data_time: 0.0043 memory: 8462 loss: 0.0260 decode.loss_ce: 0.0149 decode.acc_seg: 99.0017 aux.loss_ce: 0.0111 aux.acc_seg: 98.8382 +04/17 11:50:33 - mmengine - INFO - Iter(train) [ 11900/160000] base_lr: 9.3439e-05 lr: 3.4546e-07 eta: 1 day, 17:09:45 time: 0.9980 data_time: 0.0044 memory: 8462 loss: 0.0258 decode.loss_ce: 0.0150 decode.acc_seg: 99.1013 aux.loss_ce: 0.0108 aux.acc_seg: 98.4045 +04/17 11:51:23 - mmengine - INFO - Iter(train) [ 11950/160000] base_lr: 9.3408e-05 lr: 3.4535e-07 eta: 1 day, 17:08:54 time: 0.9983 data_time: 0.0045 memory: 8462 loss: 0.0275 decode.loss_ce: 0.0159 decode.acc_seg: 99.4509 aux.loss_ce: 0.0116 aux.acc_seg: 98.7490 +04/17 11:52:13 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 11:52:13 - mmengine - INFO - Iter(train) [ 12000/160000] base_lr: 9.3376e-05 lr: 3.4523e-07 eta: 1 day, 17:08:02 time: 0.9975 data_time: 0.0043 memory: 8462 loss: 0.0261 decode.loss_ce: 0.0151 decode.acc_seg: 98.8495 aux.loss_ce: 0.0110 aux.acc_seg: 98.4133 +04/17 11:53:03 - mmengine - INFO - Iter(train) [ 12050/160000] base_lr: 9.3344e-05 lr: 3.4511e-07 eta: 1 day, 17:07:10 time: 0.9981 data_time: 0.0046 memory: 8462 loss: 0.0233 decode.loss_ce: 0.0136 decode.acc_seg: 99.4600 aux.loss_ce: 0.0097 aux.acc_seg: 99.2146 +04/17 11:53:53 - mmengine - INFO - Iter(train) [ 12100/160000] base_lr: 9.3313e-05 lr: 3.4500e-07 eta: 1 day, 17:06:19 time: 0.9968 data_time: 0.0051 memory: 8462 loss: 0.0287 decode.loss_ce: 0.0161 decode.acc_seg: 99.6464 aux.loss_ce: 0.0126 aux.acc_seg: 99.0677 +04/17 11:54:43 - mmengine - INFO - Iter(train) [ 12150/160000] base_lr: 9.3281e-05 lr: 3.4488e-07 eta: 1 day, 17:05:27 time: 0.9973 data_time: 0.0045 memory: 8462 loss: 0.0291 decode.loss_ce: 0.0168 decode.acc_seg: 99.0984 aux.loss_ce: 0.0123 aux.acc_seg: 98.5958 +04/17 11:55:33 - mmengine - INFO - Iter(train) [ 12200/160000] base_lr: 9.3250e-05 lr: 3.4476e-07 eta: 1 day, 17:04:35 time: 0.9968 data_time: 0.0042 memory: 8462 loss: 0.0271 decode.loss_ce: 0.0160 decode.acc_seg: 99.2754 aux.loss_ce: 0.0111 aux.acc_seg: 98.7045 +04/17 11:56:22 - mmengine - INFO - Iter(train) [ 12250/160000] base_lr: 9.3218e-05 lr: 3.4465e-07 eta: 1 day, 17:03:43 time: 0.9983 data_time: 0.0043 memory: 8462 loss: 0.0274 decode.loss_ce: 0.0156 decode.acc_seg: 99.5762 aux.loss_ce: 0.0118 aux.acc_seg: 98.3437 +04/17 11:57:12 - mmengine - INFO - Iter(train) [ 12300/160000] base_lr: 9.3187e-05 lr: 3.4453e-07 eta: 1 day, 17:02:52 time: 0.9974 data_time: 0.0045 memory: 8462 loss: 0.0259 decode.loss_ce: 0.0148 decode.acc_seg: 99.4024 aux.loss_ce: 0.0111 aux.acc_seg: 98.6334 +04/17 11:58:02 - mmengine - INFO - Iter(train) [ 12350/160000] base_lr: 9.3155e-05 lr: 3.4441e-07 eta: 1 day, 17:02:01 time: 0.9980 data_time: 0.0047 memory: 8462 loss: 0.0252 decode.loss_ce: 0.0144 decode.acc_seg: 99.4419 aux.loss_ce: 0.0107 aux.acc_seg: 98.6288 +04/17 11:58:52 - mmengine - INFO - Iter(train) [ 12400/160000] base_lr: 9.3124e-05 lr: 3.4430e-07 eta: 1 day, 17:01:09 time: 0.9977 data_time: 0.0047 memory: 8462 loss: 0.0268 decode.loss_ce: 0.0153 decode.acc_seg: 99.0204 aux.loss_ce: 0.0115 aux.acc_seg: 98.3780 +04/17 11:59:42 - mmengine - INFO - Iter(train) [ 12450/160000] base_lr: 9.3092e-05 lr: 3.4418e-07 eta: 1 day, 17:00:18 time: 0.9977 data_time: 0.0048 memory: 8462 loss: 0.0252 decode.loss_ce: 0.0142 decode.acc_seg: 99.4844 aux.loss_ce: 0.0110 aux.acc_seg: 98.9780 +04/17 12:00:32 - mmengine - INFO - Iter(train) [ 12500/160000] base_lr: 9.3061e-05 lr: 3.4406e-07 eta: 1 day, 16:59:26 time: 0.9965 data_time: 0.0042 memory: 8462 loss: 0.0316 decode.loss_ce: 0.0184 decode.acc_seg: 99.4486 aux.loss_ce: 0.0132 aux.acc_seg: 98.5220 +04/17 12:01:22 - mmengine - INFO - Iter(train) [ 12550/160000] base_lr: 9.3029e-05 lr: 3.4395e-07 eta: 1 day, 16:58:35 time: 0.9981 data_time: 0.0044 memory: 8462 loss: 0.0269 decode.loss_ce: 0.0160 decode.acc_seg: 99.4404 aux.loss_ce: 0.0109 aux.acc_seg: 99.0669 +04/17 12:02:12 - mmengine - INFO - Iter(train) [ 12600/160000] base_lr: 9.2997e-05 lr: 3.4383e-07 eta: 1 day, 16:57:43 time: 0.9981 data_time: 0.0044 memory: 8462 loss: 0.0245 decode.loss_ce: 0.0137 decode.acc_seg: 99.5785 aux.loss_ce: 0.0108 aux.acc_seg: 99.0877 +04/17 12:03:02 - mmengine - INFO - Iter(train) [ 12650/160000] base_lr: 9.2966e-05 lr: 3.4371e-07 eta: 1 day, 16:56:52 time: 0.9991 data_time: 0.0043 memory: 8462 loss: 0.0250 decode.loss_ce: 0.0136 decode.acc_seg: 99.3414 aux.loss_ce: 0.0114 aux.acc_seg: 98.4335 +04/17 12:03:52 - mmengine - INFO - Iter(train) [ 12700/160000] base_lr: 9.2934e-05 lr: 3.4360e-07 eta: 1 day, 16:56:01 time: 0.9980 data_time: 0.0045 memory: 8462 loss: 0.0310 decode.loss_ce: 0.0179 decode.acc_seg: 99.0231 aux.loss_ce: 0.0131 aux.acc_seg: 98.6908 +04/17 12:04:42 - mmengine - INFO - Iter(train) [ 12750/160000] base_lr: 9.2903e-05 lr: 3.4348e-07 eta: 1 day, 16:55:09 time: 0.9977 data_time: 0.0047 memory: 8462 loss: 0.0256 decode.loss_ce: 0.0142 decode.acc_seg: 99.5392 aux.loss_ce: 0.0114 aux.acc_seg: 99.0322 +04/17 12:05:31 - mmengine - INFO - Iter(train) [ 12800/160000] base_lr: 9.2871e-05 lr: 3.4336e-07 eta: 1 day, 16:54:18 time: 0.9983 data_time: 0.0048 memory: 8462 loss: 0.0259 decode.loss_ce: 0.0146 decode.acc_seg: 99.4265 aux.loss_ce: 0.0113 aux.acc_seg: 98.6423 +04/17 12:06:21 - mmengine - INFO - Iter(train) [ 12850/160000] base_lr: 9.2840e-05 lr: 3.4325e-07 eta: 1 day, 16:53:26 time: 0.9972 data_time: 0.0046 memory: 8462 loss: 0.0269 decode.loss_ce: 0.0159 decode.acc_seg: 99.3744 aux.loss_ce: 0.0110 aux.acc_seg: 98.5023 +04/17 12:07:11 - mmengine - INFO - Iter(train) [ 12900/160000] base_lr: 9.2808e-05 lr: 3.4313e-07 eta: 1 day, 16:52:35 time: 0.9987 data_time: 0.0041 memory: 8462 loss: 0.0240 decode.loss_ce: 0.0135 decode.acc_seg: 99.3826 aux.loss_ce: 0.0106 aux.acc_seg: 98.7181 +04/17 12:08:01 - mmengine - INFO - Iter(train) [ 12950/160000] base_lr: 9.2777e-05 lr: 3.4301e-07 eta: 1 day, 16:51:43 time: 0.9990 data_time: 0.0058 memory: 8462 loss: 0.0213 decode.loss_ce: 0.0118 decode.acc_seg: 99.4797 aux.loss_ce: 0.0095 aux.acc_seg: 98.8075 +04/17 12:08:51 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 12:08:51 - mmengine - INFO - Iter(train) [ 13000/160000] base_lr: 9.2745e-05 lr: 3.4290e-07 eta: 1 day, 16:50:52 time: 0.9986 data_time: 0.0047 memory: 8462 loss: 0.0247 decode.loss_ce: 0.0142 decode.acc_seg: 99.0213 aux.loss_ce: 0.0105 aux.acc_seg: 98.4758 +04/17 12:09:41 - mmengine - INFO - Iter(train) [ 13050/160000] base_lr: 9.2714e-05 lr: 3.4278e-07 eta: 1 day, 16:50:00 time: 0.9998 data_time: 0.0043 memory: 8462 loss: 0.0242 decode.loss_ce: 0.0134 decode.acc_seg: 99.3568 aux.loss_ce: 0.0108 aux.acc_seg: 98.4947 +04/17 12:10:31 - mmengine - INFO - Iter(train) [ 13100/160000] base_lr: 9.2682e-05 lr: 3.4266e-07 eta: 1 day, 16:49:09 time: 0.9990 data_time: 0.0042 memory: 8462 loss: 0.0222 decode.loss_ce: 0.0125 decode.acc_seg: 99.0801 aux.loss_ce: 0.0097 aux.acc_seg: 98.3044 +04/17 12:11:21 - mmengine - INFO - Iter(train) [ 13150/160000] base_lr: 9.2650e-05 lr: 3.4255e-07 eta: 1 day, 16:48:18 time: 0.9983 data_time: 0.0041 memory: 8462 loss: 0.0304 decode.loss_ce: 0.0169 decode.acc_seg: 99.1022 aux.loss_ce: 0.0135 aux.acc_seg: 98.2176 +04/17 12:12:11 - mmengine - INFO - Iter(train) [ 13200/160000] base_lr: 9.2619e-05 lr: 3.4243e-07 eta: 1 day, 16:47:26 time: 0.9979 data_time: 0.0046 memory: 8462 loss: 0.0278 decode.loss_ce: 0.0157 decode.acc_seg: 99.3277 aux.loss_ce: 0.0122 aux.acc_seg: 98.3519 +04/17 12:13:00 - mmengine - INFO - Iter(train) [ 13250/160000] base_lr: 9.2587e-05 lr: 3.4231e-07 eta: 1 day, 16:46:35 time: 0.9980 data_time: 0.0043 memory: 8462 loss: 0.0255 decode.loss_ce: 0.0143 decode.acc_seg: 99.5480 aux.loss_ce: 0.0112 aux.acc_seg: 98.9100 +04/17 12:13:50 - mmengine - INFO - Iter(train) [ 13300/160000] base_lr: 9.2556e-05 lr: 3.4220e-07 eta: 1 day, 16:45:43 time: 0.9966 data_time: 0.0045 memory: 8462 loss: 0.0245 decode.loss_ce: 0.0140 decode.acc_seg: 99.0860 aux.loss_ce: 0.0104 aux.acc_seg: 98.6025 +04/17 12:14:40 - mmengine - INFO - Iter(train) [ 13350/160000] base_lr: 9.2524e-05 lr: 3.4208e-07 eta: 1 day, 16:44:52 time: 0.9980 data_time: 0.0041 memory: 8462 loss: 0.0238 decode.loss_ce: 0.0135 decode.acc_seg: 99.5161 aux.loss_ce: 0.0103 aux.acc_seg: 99.2218 +04/17 12:15:30 - mmengine - INFO - Iter(train) [ 13400/160000] base_lr: 9.2493e-05 lr: 3.4196e-07 eta: 1 day, 16:44:00 time: 0.9974 data_time: 0.0045 memory: 8462 loss: 0.0247 decode.loss_ce: 0.0134 decode.acc_seg: 99.6300 aux.loss_ce: 0.0113 aux.acc_seg: 99.0601 +04/17 12:16:20 - mmengine - INFO - Iter(train) [ 13450/160000] base_lr: 9.2461e-05 lr: 3.4185e-07 eta: 1 day, 16:43:09 time: 0.9979 data_time: 0.0045 memory: 8462 loss: 0.0240 decode.loss_ce: 0.0135 decode.acc_seg: 99.5245 aux.loss_ce: 0.0105 aux.acc_seg: 98.8018 +04/17 12:17:10 - mmengine - INFO - Iter(train) [ 13500/160000] base_lr: 9.2430e-05 lr: 3.4173e-07 eta: 1 day, 16:42:18 time: 0.9969 data_time: 0.0043 memory: 8462 loss: 0.0278 decode.loss_ce: 0.0156 decode.acc_seg: 99.4808 aux.loss_ce: 0.0123 aux.acc_seg: 98.9613 +04/17 12:18:00 - mmengine - INFO - Iter(train) [ 13550/160000] base_lr: 9.2398e-05 lr: 3.4161e-07 eta: 1 day, 16:41:26 time: 0.9970 data_time: 0.0047 memory: 8462 loss: 0.0221 decode.loss_ce: 0.0125 decode.acc_seg: 98.9977 aux.loss_ce: 0.0096 aux.acc_seg: 98.8420 +04/17 12:18:50 - mmengine - INFO - Iter(train) [ 13600/160000] base_lr: 9.2367e-05 lr: 3.4150e-07 eta: 1 day, 16:40:35 time: 0.9968 data_time: 0.0048 memory: 8462 loss: 0.0210 decode.loss_ce: 0.0115 decode.acc_seg: 99.4091 aux.loss_ce: 0.0095 aux.acc_seg: 98.9174 +04/17 12:19:40 - mmengine - INFO - Iter(train) [ 13650/160000] base_lr: 9.2335e-05 lr: 3.4138e-07 eta: 1 day, 16:39:43 time: 0.9975 data_time: 0.0044 memory: 8462 loss: 0.0219 decode.loss_ce: 0.0118 decode.acc_seg: 99.1947 aux.loss_ce: 0.0101 aux.acc_seg: 98.6771 +04/17 12:20:29 - mmengine - INFO - Iter(train) [ 13700/160000] base_lr: 9.2303e-05 lr: 3.4126e-07 eta: 1 day, 16:38:52 time: 0.9994 data_time: 0.0046 memory: 8462 loss: 0.0219 decode.loss_ce: 0.0123 decode.acc_seg: 99.5081 aux.loss_ce: 0.0096 aux.acc_seg: 98.7627 +04/17 12:21:19 - mmengine - INFO - Iter(train) [ 13750/160000] base_lr: 9.2272e-05 lr: 3.4115e-07 eta: 1 day, 16:38:00 time: 0.9972 data_time: 0.0048 memory: 8462 loss: 0.0248 decode.loss_ce: 0.0138 decode.acc_seg: 99.3843 aux.loss_ce: 0.0110 aux.acc_seg: 98.3892 +04/17 12:22:09 - mmengine - INFO - Iter(train) [ 13800/160000] base_lr: 9.2240e-05 lr: 3.4103e-07 eta: 1 day, 16:37:09 time: 0.9985 data_time: 0.0048 memory: 8462 loss: 0.0283 decode.loss_ce: 0.0167 decode.acc_seg: 99.5398 aux.loss_ce: 0.0116 aux.acc_seg: 99.1119 +04/17 12:22:59 - mmengine - INFO - Iter(train) [ 13850/160000] base_lr: 9.2209e-05 lr: 3.4091e-07 eta: 1 day, 16:36:18 time: 0.9953 data_time: 0.0045 memory: 8462 loss: 0.0256 decode.loss_ce: 0.0142 decode.acc_seg: 98.9668 aux.loss_ce: 0.0114 aux.acc_seg: 98.2203 +04/17 12:23:49 - mmengine - INFO - Iter(train) [ 13900/160000] base_lr: 9.2177e-05 lr: 3.4080e-07 eta: 1 day, 16:35:26 time: 0.9975 data_time: 0.0042 memory: 8462 loss: 0.0242 decode.loss_ce: 0.0139 decode.acc_seg: 99.4198 aux.loss_ce: 0.0104 aux.acc_seg: 98.5403 +04/17 12:24:39 - mmengine - INFO - Iter(train) [ 13950/160000] base_lr: 9.2146e-05 lr: 3.4068e-07 eta: 1 day, 16:34:35 time: 0.9975 data_time: 0.0045 memory: 8462 loss: 0.0241 decode.loss_ce: 0.0139 decode.acc_seg: 99.3166 aux.loss_ce: 0.0102 aux.acc_seg: 98.9058 +04/17 12:25:29 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 12:25:29 - mmengine - INFO - Iter(train) [ 14000/160000] base_lr: 9.2114e-05 lr: 3.4056e-07 eta: 1 day, 16:33:43 time: 0.9971 data_time: 0.0042 memory: 8462 loss: 0.0258 decode.loss_ce: 0.0149 decode.acc_seg: 99.5991 aux.loss_ce: 0.0108 aux.acc_seg: 99.1245 +04/17 12:26:19 - mmengine - INFO - Iter(train) [ 14050/160000] base_lr: 9.2083e-05 lr: 3.4045e-07 eta: 1 day, 16:32:52 time: 0.9998 data_time: 0.0049 memory: 8462 loss: 0.0270 decode.loss_ce: 0.0141 decode.acc_seg: 99.3965 aux.loss_ce: 0.0128 aux.acc_seg: 98.5229 +04/17 12:27:09 - mmengine - INFO - Iter(train) [ 14100/160000] base_lr: 9.2051e-05 lr: 3.4033e-07 eta: 1 day, 16:32:01 time: 0.9974 data_time: 0.0043 memory: 8462 loss: 0.0208 decode.loss_ce: 0.0118 decode.acc_seg: 99.5802 aux.loss_ce: 0.0090 aux.acc_seg: 99.3126 +04/17 12:27:58 - mmengine - INFO - Iter(train) [ 14150/160000] base_lr: 9.2020e-05 lr: 3.4022e-07 eta: 1 day, 16:31:10 time: 0.9978 data_time: 0.0044 memory: 8462 loss: 0.0228 decode.loss_ce: 0.0126 decode.acc_seg: 99.6016 aux.loss_ce: 0.0102 aux.acc_seg: 98.9491 +04/17 12:28:48 - mmengine - INFO - Iter(train) [ 14200/160000] base_lr: 9.1988e-05 lr: 3.4010e-07 eta: 1 day, 16:30:18 time: 0.9979 data_time: 0.0048 memory: 8462 loss: 0.0248 decode.loss_ce: 0.0138 decode.acc_seg: 99.4162 aux.loss_ce: 0.0109 aux.acc_seg: 98.5853 +04/17 12:29:38 - mmengine - INFO - Iter(train) [ 14250/160000] base_lr: 9.1956e-05 lr: 3.3998e-07 eta: 1 day, 16:29:27 time: 0.9979 data_time: 0.0042 memory: 8462 loss: 0.0226 decode.loss_ce: 0.0123 decode.acc_seg: 99.3723 aux.loss_ce: 0.0103 aux.acc_seg: 98.7762 +04/17 12:30:28 - mmengine - INFO - Iter(train) [ 14300/160000] base_lr: 9.1925e-05 lr: 3.3987e-07 eta: 1 day, 16:28:35 time: 0.9959 data_time: 0.0042 memory: 8462 loss: 0.0210 decode.loss_ce: 0.0114 decode.acc_seg: 99.5321 aux.loss_ce: 0.0096 aux.acc_seg: 98.9382 +04/17 12:31:18 - mmengine - INFO - Iter(train) [ 14350/160000] base_lr: 9.1893e-05 lr: 3.3975e-07 eta: 1 day, 16:27:44 time: 0.9986 data_time: 0.0045 memory: 8462 loss: 0.0255 decode.loss_ce: 0.0144 decode.acc_seg: 99.4427 aux.loss_ce: 0.0111 aux.acc_seg: 98.8909 +04/17 12:32:08 - mmengine - INFO - Iter(train) [ 14400/160000] base_lr: 9.1862e-05 lr: 3.3963e-07 eta: 1 day, 16:26:53 time: 0.9988 data_time: 0.0046 memory: 8462 loss: 0.0243 decode.loss_ce: 0.0139 decode.acc_seg: 99.5283 aux.loss_ce: 0.0105 aux.acc_seg: 99.2180 +04/17 12:32:58 - mmengine - INFO - Iter(train) [ 14450/160000] base_lr: 9.1830e-05 lr: 3.3952e-07 eta: 1 day, 16:26:02 time: 0.9968 data_time: 0.0044 memory: 8462 loss: 0.0227 decode.loss_ce: 0.0125 decode.acc_seg: 99.5022 aux.loss_ce: 0.0101 aux.acc_seg: 98.7011 +04/17 12:33:48 - mmengine - INFO - Iter(train) [ 14500/160000] base_lr: 9.1799e-05 lr: 3.3940e-07 eta: 1 day, 16:25:10 time: 0.9973 data_time: 0.0047 memory: 8462 loss: 0.0254 decode.loss_ce: 0.0139 decode.acc_seg: 99.5682 aux.loss_ce: 0.0115 aux.acc_seg: 99.0284 +04/17 12:34:37 - mmengine - INFO - Iter(train) [ 14550/160000] base_lr: 9.1767e-05 lr: 3.3928e-07 eta: 1 day, 16:24:19 time: 0.9985 data_time: 0.0046 memory: 8462 loss: 0.0246 decode.loss_ce: 0.0136 decode.acc_seg: 99.6719 aux.loss_ce: 0.0110 aux.acc_seg: 99.3185 +04/17 12:35:27 - mmengine - INFO - Iter(train) [ 14600/160000] base_lr: 9.1736e-05 lr: 3.3917e-07 eta: 1 day, 16:23:28 time: 0.9979 data_time: 0.0049 memory: 8462 loss: 0.0225 decode.loss_ce: 0.0125 decode.acc_seg: 99.6122 aux.loss_ce: 0.0099 aux.acc_seg: 99.1463 +04/17 12:36:17 - mmengine - INFO - Iter(train) [ 14650/160000] base_lr: 9.1704e-05 lr: 3.3905e-07 eta: 1 day, 16:22:37 time: 0.9967 data_time: 0.0044 memory: 8462 loss: 0.0225 decode.loss_ce: 0.0127 decode.acc_seg: 99.5100 aux.loss_ce: 0.0098 aux.acc_seg: 99.2855 +04/17 12:37:07 - mmengine - INFO - Iter(train) [ 14700/160000] base_lr: 9.1673e-05 lr: 3.3893e-07 eta: 1 day, 16:21:46 time: 0.9975 data_time: 0.0043 memory: 8462 loss: 0.0222 decode.loss_ce: 0.0120 decode.acc_seg: 99.6143 aux.loss_ce: 0.0102 aux.acc_seg: 99.1068 +04/17 12:37:57 - mmengine - INFO - Iter(train) [ 14750/160000] base_lr: 9.1641e-05 lr: 3.3882e-07 eta: 1 day, 16:20:55 time: 0.9978 data_time: 0.0043 memory: 8462 loss: 0.0283 decode.loss_ce: 0.0164 decode.acc_seg: 99.5216 aux.loss_ce: 0.0120 aux.acc_seg: 98.9532 +04/17 12:38:47 - mmengine - INFO - Iter(train) [ 14800/160000] base_lr: 9.1609e-05 lr: 3.3870e-07 eta: 1 day, 16:20:04 time: 0.9989 data_time: 0.0042 memory: 8462 loss: 0.0275 decode.loss_ce: 0.0154 decode.acc_seg: 99.6338 aux.loss_ce: 0.0122 aux.acc_seg: 99.1535 +04/17 12:39:37 - mmengine - INFO - Iter(train) [ 14850/160000] base_lr: 9.1578e-05 lr: 3.3858e-07 eta: 1 day, 16:19:13 time: 0.9978 data_time: 0.0044 memory: 8462 loss: 0.0235 decode.loss_ce: 0.0129 decode.acc_seg: 99.4799 aux.loss_ce: 0.0106 aux.acc_seg: 98.4684 +04/17 12:40:27 - mmengine - INFO - Iter(train) [ 14900/160000] base_lr: 9.1546e-05 lr: 3.3847e-07 eta: 1 day, 16:18:22 time: 0.9977 data_time: 0.0044 memory: 8462 loss: 0.0240 decode.loss_ce: 0.0132 decode.acc_seg: 99.5035 aux.loss_ce: 0.0108 aux.acc_seg: 99.1489 +04/17 12:41:17 - mmengine - INFO - Iter(train) [ 14950/160000] base_lr: 9.1515e-05 lr: 3.3835e-07 eta: 1 day, 16:17:31 time: 0.9975 data_time: 0.0046 memory: 8462 loss: 0.0240 decode.loss_ce: 0.0135 decode.acc_seg: 99.4625 aux.loss_ce: 0.0105 aux.acc_seg: 99.0116 +04/17 12:42:06 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 12:42:06 - mmengine - INFO - Iter(train) [ 15000/160000] base_lr: 9.1483e-05 lr: 3.3823e-07 eta: 1 day, 16:16:40 time: 0.9989 data_time: 0.0046 memory: 8462 loss: 0.0261 decode.loss_ce: 0.0151 decode.acc_seg: 99.4242 aux.loss_ce: 0.0111 aux.acc_seg: 99.0814 +04/17 12:42:56 - mmengine - INFO - Iter(train) [ 15050/160000] base_lr: 9.1452e-05 lr: 3.3812e-07 eta: 1 day, 16:15:48 time: 0.9970 data_time: 0.0042 memory: 8462 loss: 0.0240 decode.loss_ce: 0.0142 decode.acc_seg: 99.3740 aux.loss_ce: 0.0098 aux.acc_seg: 99.0076 +04/17 12:43:46 - mmengine - INFO - Iter(train) [ 15100/160000] base_lr: 9.1420e-05 lr: 3.3800e-07 eta: 1 day, 16:14:57 time: 0.9973 data_time: 0.0043 memory: 8462 loss: 0.0253 decode.loss_ce: 0.0133 decode.acc_seg: 99.4009 aux.loss_ce: 0.0120 aux.acc_seg: 98.8981 +04/17 12:44:36 - mmengine - INFO - Iter(train) [ 15150/160000] base_lr: 9.1389e-05 lr: 3.3788e-07 eta: 1 day, 16:14:06 time: 0.9979 data_time: 0.0042 memory: 8462 loss: 0.0229 decode.loss_ce: 0.0128 decode.acc_seg: 99.4537 aux.loss_ce: 0.0101 aux.acc_seg: 98.9344 +04/17 12:45:26 - mmengine - INFO - Iter(train) [ 15200/160000] base_lr: 9.1357e-05 lr: 3.3777e-07 eta: 1 day, 16:13:15 time: 0.9968 data_time: 0.0042 memory: 8462 loss: 0.0211 decode.loss_ce: 0.0114 decode.acc_seg: 99.4286 aux.loss_ce: 0.0097 aux.acc_seg: 98.8831 +04/17 12:46:16 - mmengine - INFO - Iter(train) [ 15250/160000] base_lr: 9.1325e-05 lr: 3.3765e-07 eta: 1 day, 16:12:23 time: 0.9980 data_time: 0.0043 memory: 8462 loss: 0.0251 decode.loss_ce: 0.0142 decode.acc_seg: 99.4448 aux.loss_ce: 0.0109 aux.acc_seg: 98.9090 +04/17 12:47:06 - mmengine - INFO - Iter(train) [ 15300/160000] base_lr: 9.1294e-05 lr: 3.3753e-07 eta: 1 day, 16:11:33 time: 0.9967 data_time: 0.0042 memory: 8462 loss: 0.0245 decode.loss_ce: 0.0134 decode.acc_seg: 99.2613 aux.loss_ce: 0.0111 aux.acc_seg: 98.5228 +04/17 12:47:56 - mmengine - INFO - Iter(train) [ 15350/160000] base_lr: 9.1262e-05 lr: 3.3742e-07 eta: 1 day, 16:10:41 time: 0.9972 data_time: 0.0045 memory: 8462 loss: 0.0223 decode.loss_ce: 0.0122 decode.acc_seg: 99.6096 aux.loss_ce: 0.0101 aux.acc_seg: 99.0072 +04/17 12:48:45 - mmengine - INFO - Iter(train) [ 15400/160000] base_lr: 9.1231e-05 lr: 3.3730e-07 eta: 1 day, 16:09:50 time: 0.9982 data_time: 0.0052 memory: 8462 loss: 0.0191 decode.loss_ce: 0.0105 decode.acc_seg: 99.5132 aux.loss_ce: 0.0086 aux.acc_seg: 99.0273 +04/17 12:49:35 - mmengine - INFO - Iter(train) [ 15450/160000] base_lr: 9.1199e-05 lr: 3.3718e-07 eta: 1 day, 16:08:59 time: 0.9986 data_time: 0.0045 memory: 8462 loss: 0.0234 decode.loss_ce: 0.0126 decode.acc_seg: 99.6105 aux.loss_ce: 0.0108 aux.acc_seg: 98.9338 +04/17 12:50:25 - mmengine - INFO - Iter(train) [ 15500/160000] base_lr: 9.1168e-05 lr: 3.3707e-07 eta: 1 day, 16:08:08 time: 0.9982 data_time: 0.0046 memory: 8462 loss: 0.0237 decode.loss_ce: 0.0137 decode.acc_seg: 99.3944 aux.loss_ce: 0.0101 aux.acc_seg: 98.6105 +04/17 12:51:15 - mmengine - INFO - Iter(train) [ 15550/160000] base_lr: 9.1136e-05 lr: 3.3695e-07 eta: 1 day, 16:07:17 time: 0.9970 data_time: 0.0041 memory: 8462 loss: 0.0255 decode.loss_ce: 0.0140 decode.acc_seg: 99.3401 aux.loss_ce: 0.0115 aux.acc_seg: 99.0372 +04/17 12:52:05 - mmengine - INFO - Iter(train) [ 15600/160000] base_lr: 9.1105e-05 lr: 3.3683e-07 eta: 1 day, 16:06:26 time: 0.9961 data_time: 0.0042 memory: 8462 loss: 0.0236 decode.loss_ce: 0.0133 decode.acc_seg: 99.5707 aux.loss_ce: 0.0103 aux.acc_seg: 99.0843 +04/17 12:52:55 - mmengine - INFO - Iter(train) [ 15650/160000] base_lr: 9.1073e-05 lr: 3.3672e-07 eta: 1 day, 16:05:35 time: 0.9966 data_time: 0.0045 memory: 8462 loss: 0.0208 decode.loss_ce: 0.0115 decode.acc_seg: 99.3523 aux.loss_ce: 0.0093 aux.acc_seg: 98.6834 +04/17 12:53:45 - mmengine - INFO - Iter(train) [ 15700/160000] base_lr: 9.1042e-05 lr: 3.3660e-07 eta: 1 day, 16:04:44 time: 0.9970 data_time: 0.0047 memory: 8462 loss: 0.0225 decode.loss_ce: 0.0122 decode.acc_seg: 99.3650 aux.loss_ce: 0.0103 aux.acc_seg: 98.6662 +04/17 12:54:35 - mmengine - INFO - Iter(train) [ 15750/160000] base_lr: 9.1010e-05 lr: 3.3648e-07 eta: 1 day, 16:03:52 time: 0.9969 data_time: 0.0046 memory: 8462 loss: 0.0229 decode.loss_ce: 0.0129 decode.acc_seg: 99.5964 aux.loss_ce: 0.0099 aux.acc_seg: 99.0538 +04/17 12:55:24 - mmengine - INFO - Iter(train) [ 15800/160000] base_lr: 9.0978e-05 lr: 3.3637e-07 eta: 1 day, 16:03:01 time: 0.9977 data_time: 0.0044 memory: 8462 loss: 0.0240 decode.loss_ce: 0.0131 decode.acc_seg: 99.5281 aux.loss_ce: 0.0108 aux.acc_seg: 98.9275 +04/17 12:56:14 - mmengine - INFO - Iter(train) [ 15850/160000] base_lr: 9.0947e-05 lr: 3.3625e-07 eta: 1 day, 16:02:10 time: 0.9989 data_time: 0.0049 memory: 8462 loss: 0.0237 decode.loss_ce: 0.0130 decode.acc_seg: 99.7799 aux.loss_ce: 0.0108 aux.acc_seg: 99.4410 +04/17 12:57:04 - mmengine - INFO - Iter(train) [ 15900/160000] base_lr: 9.0915e-05 lr: 3.3613e-07 eta: 1 day, 16:01:19 time: 0.9970 data_time: 0.0045 memory: 8462 loss: 0.0247 decode.loss_ce: 0.0138 decode.acc_seg: 99.0255 aux.loss_ce: 0.0109 aux.acc_seg: 98.0291 +04/17 12:57:54 - mmengine - INFO - Iter(train) [ 15950/160000] base_lr: 9.0884e-05 lr: 3.3602e-07 eta: 1 day, 16:00:28 time: 0.9987 data_time: 0.0048 memory: 8462 loss: 0.0214 decode.loss_ce: 0.0121 decode.acc_seg: 99.2308 aux.loss_ce: 0.0093 aux.acc_seg: 98.8634 +04/17 12:58:44 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 12:58:44 - mmengine - INFO - Iter(train) [ 16000/160000] base_lr: 9.0852e-05 lr: 3.3590e-07 eta: 1 day, 15:59:36 time: 0.9974 data_time: 0.0051 memory: 8462 loss: 0.0224 decode.loss_ce: 0.0126 decode.acc_seg: 98.7751 aux.loss_ce: 0.0098 aux.acc_seg: 98.1810 +04/17 12:59:34 - mmengine - INFO - Iter(train) [ 16050/160000] base_lr: 9.0821e-05 lr: 3.3578e-07 eta: 1 day, 15:58:45 time: 0.9982 data_time: 0.0048 memory: 8462 loss: 0.0217 decode.loss_ce: 0.0117 decode.acc_seg: 99.1167 aux.loss_ce: 0.0099 aux.acc_seg: 98.3572 +04/17 13:00:24 - mmengine - INFO - Iter(train) [ 16100/160000] base_lr: 9.0789e-05 lr: 3.3567e-07 eta: 1 day, 15:57:54 time: 0.9979 data_time: 0.0042 memory: 8462 loss: 0.0205 decode.loss_ce: 0.0111 decode.acc_seg: 99.3414 aux.loss_ce: 0.0094 aux.acc_seg: 98.5754 +04/17 13:01:14 - mmengine - INFO - Iter(train) [ 16150/160000] base_lr: 9.0758e-05 lr: 3.3555e-07 eta: 1 day, 15:57:03 time: 0.9976 data_time: 0.0043 memory: 8462 loss: 0.0244 decode.loss_ce: 0.0141 decode.acc_seg: 99.2956 aux.loss_ce: 0.0103 aux.acc_seg: 99.0616 +04/17 13:02:03 - mmengine - INFO - Iter(train) [ 16200/160000] base_lr: 9.0726e-05 lr: 3.3543e-07 eta: 1 day, 15:56:12 time: 0.9976 data_time: 0.0039 memory: 8462 loss: 0.0227 decode.loss_ce: 0.0127 decode.acc_seg: 99.5726 aux.loss_ce: 0.0100 aux.acc_seg: 99.1488 +04/17 13:02:53 - mmengine - INFO - Iter(train) [ 16250/160000] base_lr: 9.0695e-05 lr: 3.3532e-07 eta: 1 day, 15:55:21 time: 0.9975 data_time: 0.0045 memory: 8462 loss: 0.0220 decode.loss_ce: 0.0120 decode.acc_seg: 99.4024 aux.loss_ce: 0.0100 aux.acc_seg: 98.6151 +04/17 13:03:43 - mmengine - INFO - Iter(train) [ 16300/160000] base_lr: 9.0663e-05 lr: 3.3520e-07 eta: 1 day, 15:54:30 time: 0.9963 data_time: 0.0044 memory: 8462 loss: 0.0241 decode.loss_ce: 0.0135 decode.acc_seg: 99.3757 aux.loss_ce: 0.0107 aux.acc_seg: 98.7549 +04/17 13:04:33 - mmengine - INFO - Iter(train) [ 16350/160000] base_lr: 9.0631e-05 lr: 3.3508e-07 eta: 1 day, 15:53:39 time: 0.9961 data_time: 0.0043 memory: 8462 loss: 0.0276 decode.loss_ce: 0.0151 decode.acc_seg: 99.6042 aux.loss_ce: 0.0125 aux.acc_seg: 99.2022 +04/17 13:05:23 - mmengine - INFO - Iter(train) [ 16400/160000] base_lr: 9.0600e-05 lr: 3.3497e-07 eta: 1 day, 15:52:48 time: 0.9965 data_time: 0.0045 memory: 8462 loss: 0.0253 decode.loss_ce: 0.0138 decode.acc_seg: 99.4143 aux.loss_ce: 0.0114 aux.acc_seg: 98.7661 +04/17 13:06:13 - mmengine - INFO - Iter(train) [ 16450/160000] base_lr: 9.0568e-05 lr: 3.3485e-07 eta: 1 day, 15:51:57 time: 0.9980 data_time: 0.0046 memory: 8462 loss: 0.0194 decode.loss_ce: 0.0107 decode.acc_seg: 99.5127 aux.loss_ce: 0.0087 aux.acc_seg: 98.9428 +04/17 13:07:03 - mmengine - INFO - Iter(train) [ 16500/160000] base_lr: 9.0537e-05 lr: 3.3473e-07 eta: 1 day, 15:51:06 time: 0.9970 data_time: 0.0043 memory: 8462 loss: 0.0213 decode.loss_ce: 0.0118 decode.acc_seg: 99.8001 aux.loss_ce: 0.0096 aux.acc_seg: 99.4225 +04/17 13:07:53 - mmengine - INFO - Iter(train) [ 16550/160000] base_lr: 9.0505e-05 lr: 3.3462e-07 eta: 1 day, 15:50:15 time: 0.9988 data_time: 0.0047 memory: 8462 loss: 0.0220 decode.loss_ce: 0.0119 decode.acc_seg: 99.6180 aux.loss_ce: 0.0100 aux.acc_seg: 99.0452 +04/17 13:08:42 - mmengine - INFO - Iter(train) [ 16600/160000] base_lr: 9.0474e-05 lr: 3.3450e-07 eta: 1 day, 15:49:25 time: 0.9987 data_time: 0.0050 memory: 8462 loss: 0.0230 decode.loss_ce: 0.0127 decode.acc_seg: 99.4692 aux.loss_ce: 0.0104 aux.acc_seg: 99.0362 +04/17 13:09:32 - mmengine - INFO - Iter(train) [ 16650/160000] base_lr: 9.0442e-05 lr: 3.3438e-07 eta: 1 day, 15:48:34 time: 0.9996 data_time: 0.0043 memory: 8462 loss: 0.0220 decode.loss_ce: 0.0117 decode.acc_seg: 99.4780 aux.loss_ce: 0.0104 aux.acc_seg: 98.8773 +04/17 13:10:22 - mmengine - INFO - Iter(train) [ 16700/160000] base_lr: 9.0411e-05 lr: 3.3427e-07 eta: 1 day, 15:47:44 time: 0.9975 data_time: 0.0045 memory: 8462 loss: 0.0245 decode.loss_ce: 0.0137 decode.acc_seg: 99.4667 aux.loss_ce: 0.0107 aux.acc_seg: 98.9105 +04/17 13:11:12 - mmengine - INFO - Iter(train) [ 16750/160000] base_lr: 9.0379e-05 lr: 3.3415e-07 eta: 1 day, 15:46:53 time: 0.9969 data_time: 0.0043 memory: 8462 loss: 0.0222 decode.loss_ce: 0.0121 decode.acc_seg: 99.6586 aux.loss_ce: 0.0101 aux.acc_seg: 99.2302 +04/17 13:12:02 - mmengine - INFO - Iter(train) [ 16800/160000] base_lr: 9.0348e-05 lr: 3.3403e-07 eta: 1 day, 15:46:02 time: 0.9977 data_time: 0.0043 memory: 8462 loss: 0.0223 decode.loss_ce: 0.0122 decode.acc_seg: 99.7305 aux.loss_ce: 0.0101 aux.acc_seg: 99.3217 +04/17 13:12:52 - mmengine - INFO - Iter(train) [ 16850/160000] base_lr: 9.0316e-05 lr: 3.3392e-07 eta: 1 day, 15:45:11 time: 0.9971 data_time: 0.0045 memory: 8462 loss: 0.0228 decode.loss_ce: 0.0124 decode.acc_seg: 99.4852 aux.loss_ce: 0.0104 aux.acc_seg: 98.7169 +04/17 13:13:42 - mmengine - INFO - Iter(train) [ 16900/160000] base_lr: 9.0284e-05 lr: 3.3380e-07 eta: 1 day, 15:44:20 time: 0.9984 data_time: 0.0043 memory: 8462 loss: 0.0221 decode.loss_ce: 0.0115 decode.acc_seg: 99.1451 aux.loss_ce: 0.0106 aux.acc_seg: 97.9866 +04/17 13:14:32 - mmengine - INFO - Iter(train) [ 16950/160000] base_lr: 9.0253e-05 lr: 3.3368e-07 eta: 1 day, 15:43:30 time: 0.9992 data_time: 0.0050 memory: 8462 loss: 0.0222 decode.loss_ce: 0.0122 decode.acc_seg: 99.4707 aux.loss_ce: 0.0099 aux.acc_seg: 98.7959 +04/17 13:15:22 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 13:15:22 - mmengine - INFO - Iter(train) [ 17000/160000] base_lr: 9.0221e-05 lr: 3.3357e-07 eta: 1 day, 15:42:39 time: 0.9988 data_time: 0.0044 memory: 8462 loss: 0.0238 decode.loss_ce: 0.0127 decode.acc_seg: 99.7061 aux.loss_ce: 0.0111 aux.acc_seg: 99.3080 +04/17 13:16:12 - mmengine - INFO - Iter(train) [ 17050/160000] base_lr: 9.0190e-05 lr: 3.3345e-07 eta: 1 day, 15:41:48 time: 0.9982 data_time: 0.0047 memory: 8462 loss: 0.0186 decode.loss_ce: 0.0107 decode.acc_seg: 99.4898 aux.loss_ce: 0.0079 aux.acc_seg: 99.2535 +04/17 13:17:02 - mmengine - INFO - Iter(train) [ 17100/160000] base_lr: 9.0158e-05 lr: 3.3333e-07 eta: 1 day, 15:40:58 time: 0.9988 data_time: 0.0043 memory: 8462 loss: 0.0222 decode.loss_ce: 0.0125 decode.acc_seg: 99.1219 aux.loss_ce: 0.0097 aux.acc_seg: 98.7671 +04/17 13:17:51 - mmengine - INFO - Iter(train) [ 17150/160000] base_lr: 9.0127e-05 lr: 3.3322e-07 eta: 1 day, 15:40:07 time: 0.9980 data_time: 0.0045 memory: 8462 loss: 0.0215 decode.loss_ce: 0.0113 decode.acc_seg: 99.5983 aux.loss_ce: 0.0102 aux.acc_seg: 99.0562 +04/17 13:18:41 - mmengine - INFO - Iter(train) [ 17200/160000] base_lr: 9.0095e-05 lr: 3.3310e-07 eta: 1 day, 15:39:17 time: 0.9983 data_time: 0.0044 memory: 8462 loss: 0.0220 decode.loss_ce: 0.0116 decode.acc_seg: 99.5598 aux.loss_ce: 0.0103 aux.acc_seg: 98.6620 +04/17 13:19:31 - mmengine - INFO - Iter(train) [ 17250/160000] base_lr: 9.0064e-05 lr: 3.3298e-07 eta: 1 day, 15:38:26 time: 0.9996 data_time: 0.0041 memory: 8462 loss: 0.0242 decode.loss_ce: 0.0135 decode.acc_seg: 99.4719 aux.loss_ce: 0.0107 aux.acc_seg: 99.0494 +04/17 13:20:21 - mmengine - INFO - Iter(train) [ 17300/160000] base_lr: 9.0032e-05 lr: 3.3287e-07 eta: 1 day, 15:37:36 time: 0.9993 data_time: 0.0048 memory: 8462 loss: 0.0220 decode.loss_ce: 0.0120 decode.acc_seg: 99.6330 aux.loss_ce: 0.0100 aux.acc_seg: 99.3477 +04/17 13:21:11 - mmengine - INFO - Iter(train) [ 17350/160000] base_lr: 9.0001e-05 lr: 3.3275e-07 eta: 1 day, 15:36:45 time: 0.9984 data_time: 0.0045 memory: 8462 loss: 0.0218 decode.loss_ce: 0.0114 decode.acc_seg: 99.5409 aux.loss_ce: 0.0104 aux.acc_seg: 98.9855 +04/17 13:22:01 - mmengine - INFO - Iter(train) [ 17400/160000] base_lr: 8.9969e-05 lr: 3.3263e-07 eta: 1 day, 15:35:55 time: 0.9998 data_time: 0.0045 memory: 8462 loss: 0.0220 decode.loss_ce: 0.0126 decode.acc_seg: 99.4459 aux.loss_ce: 0.0094 aux.acc_seg: 98.9763 +04/17 13:22:51 - mmengine - INFO - Iter(train) [ 17450/160000] base_lr: 8.9937e-05 lr: 3.3252e-07 eta: 1 day, 15:35:05 time: 0.9987 data_time: 0.0049 memory: 8462 loss: 0.0242 decode.loss_ce: 0.0139 decode.acc_seg: 99.3793 aux.loss_ce: 0.0102 aux.acc_seg: 98.7186 +04/17 13:23:41 - mmengine - INFO - Iter(train) [ 17500/160000] base_lr: 8.9906e-05 lr: 3.3240e-07 eta: 1 day, 15:34:14 time: 0.9980 data_time: 0.0042 memory: 8462 loss: 0.0218 decode.loss_ce: 0.0117 decode.acc_seg: 99.6786 aux.loss_ce: 0.0101 aux.acc_seg: 99.2096 +04/17 13:24:31 - mmengine - INFO - Iter(train) [ 17550/160000] base_lr: 8.9874e-05 lr: 3.3228e-07 eta: 1 day, 15:33:24 time: 0.9995 data_time: 0.0051 memory: 8462 loss: 0.0234 decode.loss_ce: 0.0125 decode.acc_seg: 99.6319 aux.loss_ce: 0.0108 aux.acc_seg: 99.2407 +04/17 13:25:21 - mmengine - INFO - Iter(train) [ 17600/160000] base_lr: 8.9843e-05 lr: 3.3217e-07 eta: 1 day, 15:32:33 time: 0.9965 data_time: 0.0044 memory: 8462 loss: 0.0224 decode.loss_ce: 0.0117 decode.acc_seg: 99.3477 aux.loss_ce: 0.0106 aux.acc_seg: 98.6540 +04/17 13:26:11 - mmengine - INFO - Iter(train) [ 17650/160000] base_lr: 8.9811e-05 lr: 3.3205e-07 eta: 1 day, 15:31:43 time: 0.9986 data_time: 0.0049 memory: 8462 loss: 0.0271 decode.loss_ce: 0.0147 decode.acc_seg: 99.2949 aux.loss_ce: 0.0124 aux.acc_seg: 98.1636 +04/17 13:27:01 - mmengine - INFO - Iter(train) [ 17700/160000] base_lr: 8.9780e-05 lr: 3.3193e-07 eta: 1 day, 15:30:52 time: 0.9986 data_time: 0.0044 memory: 8462 loss: 0.0241 decode.loss_ce: 0.0135 decode.acc_seg: 99.1665 aux.loss_ce: 0.0106 aux.acc_seg: 98.5540 +04/17 13:27:51 - mmengine - INFO - Iter(train) [ 17750/160000] base_lr: 8.9748e-05 lr: 3.3182e-07 eta: 1 day, 15:30:02 time: 0.9992 data_time: 0.0042 memory: 8462 loss: 0.0214 decode.loss_ce: 0.0111 decode.acc_seg: 99.6120 aux.loss_ce: 0.0103 aux.acc_seg: 98.6225 +04/17 13:28:41 - mmengine - INFO - Iter(train) [ 17800/160000] base_lr: 8.9717e-05 lr: 3.3170e-07 eta: 1 day, 15:29:12 time: 0.9992 data_time: 0.0042 memory: 8462 loss: 0.0186 decode.loss_ce: 0.0101 decode.acc_seg: 99.6502 aux.loss_ce: 0.0085 aux.acc_seg: 99.2655 +04/17 13:29:31 - mmengine - INFO - Iter(train) [ 17850/160000] base_lr: 8.9685e-05 lr: 3.3158e-07 eta: 1 day, 15:28:21 time: 0.9986 data_time: 0.0048 memory: 8462 loss: 0.0245 decode.loss_ce: 0.0132 decode.acc_seg: 99.6231 aux.loss_ce: 0.0113 aux.acc_seg: 99.1957 +04/17 13:30:20 - mmengine - INFO - Iter(train) [ 17900/160000] base_lr: 8.9654e-05 lr: 3.3147e-07 eta: 1 day, 15:27:31 time: 0.9986 data_time: 0.0046 memory: 8462 loss: 0.0199 decode.loss_ce: 0.0108 decode.acc_seg: 99.6162 aux.loss_ce: 0.0091 aux.acc_seg: 99.0400 +04/17 13:31:10 - mmengine - INFO - Iter(train) [ 17950/160000] base_lr: 8.9622e-05 lr: 3.3135e-07 eta: 1 day, 15:26:40 time: 0.9975 data_time: 0.0043 memory: 8462 loss: 0.0218 decode.loss_ce: 0.0121 decode.acc_seg: 99.8589 aux.loss_ce: 0.0097 aux.acc_seg: 99.5493 +04/17 13:32:00 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 13:32:00 - mmengine - INFO - Iter(train) [ 18000/160000] base_lr: 8.9590e-05 lr: 3.3123e-07 eta: 1 day, 15:25:50 time: 0.9972 data_time: 0.0043 memory: 8462 loss: 0.0230 decode.loss_ce: 0.0125 decode.acc_seg: 99.6740 aux.loss_ce: 0.0104 aux.acc_seg: 99.0997 +04/17 13:32:50 - mmengine - INFO - Iter(train) [ 18050/160000] base_lr: 8.9559e-05 lr: 3.3112e-07 eta: 1 day, 15:25:00 time: 0.9987 data_time: 0.0043 memory: 8462 loss: 0.0236 decode.loss_ce: 0.0131 decode.acc_seg: 99.5947 aux.loss_ce: 0.0105 aux.acc_seg: 99.1249 +04/17 13:33:40 - mmengine - INFO - Iter(train) [ 18100/160000] base_lr: 8.9527e-05 lr: 3.3100e-07 eta: 1 day, 15:24:09 time: 0.9989 data_time: 0.0044 memory: 8462 loss: 0.0232 decode.loss_ce: 0.0118 decode.acc_seg: 99.4345 aux.loss_ce: 0.0113 aux.acc_seg: 98.7534 +04/17 13:34:30 - mmengine - INFO - Iter(train) [ 18150/160000] base_lr: 8.9496e-05 lr: 3.3088e-07 eta: 1 day, 15:23:19 time: 0.9981 data_time: 0.0045 memory: 8462 loss: 0.0228 decode.loss_ce: 0.0121 decode.acc_seg: 99.5770 aux.loss_ce: 0.0107 aux.acc_seg: 99.1978 +04/17 13:35:20 - mmengine - INFO - Iter(train) [ 18200/160000] base_lr: 8.9464e-05 lr: 3.3077e-07 eta: 1 day, 15:22:28 time: 0.9992 data_time: 0.0042 memory: 8462 loss: 0.0214 decode.loss_ce: 0.0116 decode.acc_seg: 99.7555 aux.loss_ce: 0.0099 aux.acc_seg: 99.4345 +04/17 13:36:10 - mmengine - INFO - Iter(train) [ 18250/160000] base_lr: 8.9433e-05 lr: 3.3065e-07 eta: 1 day, 15:21:38 time: 0.9981 data_time: 0.0043 memory: 8462 loss: 0.0219 decode.loss_ce: 0.0124 decode.acc_seg: 99.6183 aux.loss_ce: 0.0094 aux.acc_seg: 98.8300 +04/17 13:37:00 - mmengine - INFO - Iter(train) [ 18300/160000] base_lr: 8.9401e-05 lr: 3.3053e-07 eta: 1 day, 15:20:48 time: 0.9982 data_time: 0.0048 memory: 8462 loss: 0.0239 decode.loss_ce: 0.0133 decode.acc_seg: 99.3641 aux.loss_ce: 0.0106 aux.acc_seg: 98.5109 +04/17 13:37:50 - mmengine - INFO - Iter(train) [ 18350/160000] base_lr: 8.9370e-05 lr: 3.3042e-07 eta: 1 day, 15:19:57 time: 0.9981 data_time: 0.0045 memory: 8462 loss: 0.0230 decode.loss_ce: 0.0124 decode.acc_seg: 99.4347 aux.loss_ce: 0.0107 aux.acc_seg: 98.7696 +04/17 13:38:40 - mmengine - INFO - Iter(train) [ 18400/160000] base_lr: 8.9338e-05 lr: 3.3030e-07 eta: 1 day, 15:19:07 time: 0.9994 data_time: 0.0046 memory: 8462 loss: 0.0277 decode.loss_ce: 0.0151 decode.acc_seg: 99.4741 aux.loss_ce: 0.0126 aux.acc_seg: 99.0673 +04/17 13:39:30 - mmengine - INFO - Iter(train) [ 18450/160000] base_lr: 8.9307e-05 lr: 3.3018e-07 eta: 1 day, 15:18:17 time: 0.9995 data_time: 0.0050 memory: 8462 loss: 0.0189 decode.loss_ce: 0.0097 decode.acc_seg: 99.5771 aux.loss_ce: 0.0091 aux.acc_seg: 98.8585 +04/17 13:40:20 - mmengine - INFO - Iter(train) [ 18500/160000] base_lr: 8.9275e-05 lr: 3.3007e-07 eta: 1 day, 15:17:27 time: 0.9988 data_time: 0.0044 memory: 8462 loss: 0.0213 decode.loss_ce: 0.0111 decode.acc_seg: 99.6395 aux.loss_ce: 0.0102 aux.acc_seg: 99.1505 +04/17 13:41:10 - mmengine - INFO - Iter(train) [ 18550/160000] base_lr: 8.9243e-05 lr: 3.2995e-07 eta: 1 day, 15:16:37 time: 0.9982 data_time: 0.0044 memory: 8462 loss: 0.0195 decode.loss_ce: 0.0100 decode.acc_seg: 99.6601 aux.loss_ce: 0.0095 aux.acc_seg: 99.2523 +04/17 13:42:00 - mmengine - INFO - Iter(train) [ 18600/160000] base_lr: 8.9212e-05 lr: 3.2983e-07 eta: 1 day, 15:15:47 time: 1.0000 data_time: 0.0051 memory: 8462 loss: 0.0218 decode.loss_ce: 0.0120 decode.acc_seg: 99.4564 aux.loss_ce: 0.0099 aux.acc_seg: 98.5744 +04/17 13:42:50 - mmengine - INFO - Iter(train) [ 18650/160000] base_lr: 8.9180e-05 lr: 3.2972e-07 eta: 1 day, 15:14:56 time: 1.0002 data_time: 0.0047 memory: 8462 loss: 0.0213 decode.loss_ce: 0.0111 decode.acc_seg: 99.4984 aux.loss_ce: 0.0102 aux.acc_seg: 98.9391 +04/17 13:43:40 - mmengine - INFO - Iter(train) [ 18700/160000] base_lr: 8.9149e-05 lr: 3.2960e-07 eta: 1 day, 15:14:06 time: 0.9989 data_time: 0.0046 memory: 8462 loss: 0.0200 decode.loss_ce: 0.0109 decode.acc_seg: 99.7986 aux.loss_ce: 0.0092 aux.acc_seg: 99.4448 +04/17 13:44:29 - mmengine - INFO - Iter(train) [ 18750/160000] base_lr: 8.9117e-05 lr: 3.2948e-07 eta: 1 day, 15:13:16 time: 0.9984 data_time: 0.0045 memory: 8462 loss: 0.0213 decode.loss_ce: 0.0116 decode.acc_seg: 99.5569 aux.loss_ce: 0.0097 aux.acc_seg: 99.1232 +04/17 13:45:19 - mmengine - INFO - Iter(train) [ 18800/160000] base_lr: 8.9086e-05 lr: 3.2937e-07 eta: 1 day, 15:12:26 time: 0.9985 data_time: 0.0042 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0092 decode.acc_seg: 99.5239 aux.loss_ce: 0.0087 aux.acc_seg: 98.6853 +04/17 13:46:09 - mmengine - INFO - Iter(train) [ 18850/160000] base_lr: 8.9054e-05 lr: 3.2925e-07 eta: 1 day, 15:11:35 time: 0.9988 data_time: 0.0046 memory: 8462 loss: 0.0216 decode.loss_ce: 0.0113 decode.acc_seg: 99.4699 aux.loss_ce: 0.0103 aux.acc_seg: 98.9473 +04/17 13:46:59 - mmengine - INFO - Iter(train) [ 18900/160000] base_lr: 8.9023e-05 lr: 3.2914e-07 eta: 1 day, 15:10:45 time: 0.9991 data_time: 0.0049 memory: 8462 loss: 0.0249 decode.loss_ce: 0.0132 decode.acc_seg: 99.6851 aux.loss_ce: 0.0116 aux.acc_seg: 99.2382 +04/17 13:47:49 - mmengine - INFO - Iter(train) [ 18950/160000] base_lr: 8.8991e-05 lr: 3.2902e-07 eta: 1 day, 15:09:54 time: 0.9983 data_time: 0.0046 memory: 8462 loss: 0.0206 decode.loss_ce: 0.0108 decode.acc_seg: 99.3652 aux.loss_ce: 0.0098 aux.acc_seg: 98.5264 +04/17 13:48:39 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 13:48:39 - mmengine - INFO - Iter(train) [ 19000/160000] base_lr: 8.8960e-05 lr: 3.2890e-07 eta: 1 day, 15:09:04 time: 0.9995 data_time: 0.0044 memory: 8462 loss: 0.0222 decode.loss_ce: 0.0120 decode.acc_seg: 99.6391 aux.loss_ce: 0.0103 aux.acc_seg: 99.2020 +04/17 13:49:29 - mmengine - INFO - Iter(train) [ 19050/160000] base_lr: 8.8928e-05 lr: 3.2879e-07 eta: 1 day, 15:08:14 time: 0.9976 data_time: 0.0042 memory: 8462 loss: 0.0216 decode.loss_ce: 0.0114 decode.acc_seg: 99.5697 aux.loss_ce: 0.0102 aux.acc_seg: 98.9531 +04/17 13:50:19 - mmengine - INFO - Iter(train) [ 19100/160000] base_lr: 8.8896e-05 lr: 3.2867e-07 eta: 1 day, 15:07:23 time: 0.9977 data_time: 0.0042 memory: 8462 loss: 0.0204 decode.loss_ce: 0.0113 decode.acc_seg: 99.6140 aux.loss_ce: 0.0091 aux.acc_seg: 99.0669 +04/17 13:51:09 - mmengine - INFO - Iter(train) [ 19150/160000] base_lr: 8.8865e-05 lr: 3.2855e-07 eta: 1 day, 15:06:33 time: 0.9992 data_time: 0.0044 memory: 8462 loss: 0.0224 decode.loss_ce: 0.0126 decode.acc_seg: 99.3668 aux.loss_ce: 0.0098 aux.acc_seg: 98.8646 +04/17 13:51:59 - mmengine - INFO - Iter(train) [ 19200/160000] base_lr: 8.8833e-05 lr: 3.2844e-07 eta: 1 day, 15:05:43 time: 0.9993 data_time: 0.0044 memory: 8462 loss: 0.0228 decode.loss_ce: 0.0120 decode.acc_seg: 99.4331 aux.loss_ce: 0.0108 aux.acc_seg: 98.8338 +04/17 13:52:49 - mmengine - INFO - Iter(train) [ 19250/160000] base_lr: 8.8802e-05 lr: 3.2832e-07 eta: 1 day, 15:04:53 time: 0.9982 data_time: 0.0043 memory: 8462 loss: 0.0219 decode.loss_ce: 0.0117 decode.acc_seg: 99.3053 aux.loss_ce: 0.0102 aux.acc_seg: 98.6570 +04/17 13:53:39 - mmengine - INFO - Iter(train) [ 19300/160000] base_lr: 8.8770e-05 lr: 3.2820e-07 eta: 1 day, 15:04:03 time: 0.9992 data_time: 0.0043 memory: 8462 loss: 0.0220 decode.loss_ce: 0.0118 decode.acc_seg: 99.6469 aux.loss_ce: 0.0102 aux.acc_seg: 99.2630 +04/17 13:54:29 - mmengine - INFO - Iter(train) [ 19350/160000] base_lr: 8.8739e-05 lr: 3.2809e-07 eta: 1 day, 15:03:13 time: 0.9986 data_time: 0.0042 memory: 8462 loss: 0.0227 decode.loss_ce: 0.0125 decode.acc_seg: 99.3605 aux.loss_ce: 0.0102 aux.acc_seg: 98.8811 +04/17 13:55:19 - mmengine - INFO - Iter(train) [ 19400/160000] base_lr: 8.8707e-05 lr: 3.2797e-07 eta: 1 day, 15:02:22 time: 1.0004 data_time: 0.0043 memory: 8462 loss: 0.0233 decode.loss_ce: 0.0126 decode.acc_seg: 99.1262 aux.loss_ce: 0.0106 aux.acc_seg: 98.6568 +04/17 13:56:09 - mmengine - INFO - Iter(train) [ 19450/160000] base_lr: 8.8676e-05 lr: 3.2785e-07 eta: 1 day, 15:01:32 time: 1.0000 data_time: 0.0043 memory: 8462 loss: 0.0219 decode.loss_ce: 0.0112 decode.acc_seg: 99.6866 aux.loss_ce: 0.0107 aux.acc_seg: 99.1751 +04/17 13:56:59 - mmengine - INFO - Iter(train) [ 19500/160000] base_lr: 8.8644e-05 lr: 3.2774e-07 eta: 1 day, 15:00:42 time: 0.9976 data_time: 0.0044 memory: 8462 loss: 0.0247 decode.loss_ce: 0.0137 decode.acc_seg: 99.2142 aux.loss_ce: 0.0111 aux.acc_seg: 98.6185 +04/17 13:57:49 - mmengine - INFO - Iter(train) [ 19550/160000] base_lr: 8.8613e-05 lr: 3.2762e-07 eta: 1 day, 14:59:52 time: 0.9992 data_time: 0.0041 memory: 8462 loss: 0.0207 decode.loss_ce: 0.0116 decode.acc_seg: 99.1375 aux.loss_ce: 0.0091 aux.acc_seg: 98.7877 +04/17 13:58:39 - mmengine - INFO - Iter(train) [ 19600/160000] base_lr: 8.8581e-05 lr: 3.2750e-07 eta: 1 day, 14:59:02 time: 0.9983 data_time: 0.0042 memory: 8462 loss: 0.0182 decode.loss_ce: 0.0093 decode.acc_seg: 99.5197 aux.loss_ce: 0.0089 aux.acc_seg: 99.0948 +04/17 13:59:28 - mmengine - INFO - Iter(train) [ 19650/160000] base_lr: 8.8549e-05 lr: 3.2739e-07 eta: 1 day, 14:58:11 time: 0.9975 data_time: 0.0044 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0097 decode.acc_seg: 99.7725 aux.loss_ce: 0.0090 aux.acc_seg: 99.3301 +04/17 14:00:18 - mmengine - INFO - Iter(train) [ 19700/160000] base_lr: 8.8518e-05 lr: 3.2727e-07 eta: 1 day, 14:57:21 time: 0.9988 data_time: 0.0046 memory: 8462 loss: 0.0210 decode.loss_ce: 0.0114 decode.acc_seg: 99.7171 aux.loss_ce: 0.0096 aux.acc_seg: 99.4770 +04/17 14:01:08 - mmengine - INFO - Iter(train) [ 19750/160000] base_lr: 8.8486e-05 lr: 3.2715e-07 eta: 1 day, 14:56:31 time: 0.9991 data_time: 0.0043 memory: 8462 loss: 0.0195 decode.loss_ce: 0.0104 decode.acc_seg: 99.5567 aux.loss_ce: 0.0091 aux.acc_seg: 99.2126 +04/17 14:01:58 - mmengine - INFO - Iter(train) [ 19800/160000] base_lr: 8.8455e-05 lr: 3.2704e-07 eta: 1 day, 14:55:41 time: 0.9990 data_time: 0.0044 memory: 8462 loss: 0.0245 decode.loss_ce: 0.0133 decode.acc_seg: 99.5886 aux.loss_ce: 0.0112 aux.acc_seg: 98.8911 +04/17 14:02:48 - mmengine - INFO - Iter(train) [ 19850/160000] base_lr: 8.8423e-05 lr: 3.2692e-07 eta: 1 day, 14:54:51 time: 0.9998 data_time: 0.0044 memory: 8462 loss: 0.0230 decode.loss_ce: 0.0129 decode.acc_seg: 99.3370 aux.loss_ce: 0.0101 aux.acc_seg: 98.9658 +04/17 14:03:38 - mmengine - INFO - Iter(train) [ 19900/160000] base_lr: 8.8392e-05 lr: 3.2680e-07 eta: 1 day, 14:54:01 time: 0.9988 data_time: 0.0045 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0098 decode.acc_seg: 99.7274 aux.loss_ce: 0.0091 aux.acc_seg: 99.2981 +04/17 14:04:28 - mmengine - INFO - Iter(train) [ 19950/160000] base_lr: 8.8360e-05 lr: 3.2669e-07 eta: 1 day, 14:53:11 time: 0.9984 data_time: 0.0044 memory: 8462 loss: 0.0244 decode.loss_ce: 0.0134 decode.acc_seg: 98.7705 aux.loss_ce: 0.0110 aux.acc_seg: 98.3341 +04/17 14:05:18 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 14:05:18 - mmengine - INFO - Iter(train) [ 20000/160000] base_lr: 8.8329e-05 lr: 3.2657e-07 eta: 1 day, 14:52:21 time: 0.9999 data_time: 0.0046 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0089 decode.acc_seg: 99.6473 aux.loss_ce: 0.0090 aux.acc_seg: 99.0765 +04/17 14:05:18 - mmengine - INFO - Saving checkpoint at 20000 iterations +04/17 14:05:28 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:17 time: 0.1155 data_time: 0.0014 memory: 4004 +04/17 14:05:34 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:11 time: 0.1158 data_time: 0.0015 memory: 4004 +04/17 14:05:40 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:05 time: 0.1156 data_time: 0.0014 memory: 4004 +04/17 14:05:46 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.1153 data_time: 0.0013 memory: 4004 +04/17 14:05:46 - mmengine - INFO - per class results: +04/17 14:05:46 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 99.14 | 99.55 | 99.57 | 99.58 | 99.55 | +| contrast | 81.18 | 89.96 | 89.61 | 89.28 | 89.96 | ++------------+-------+-------+--------+-----------+--------+ +04/17 14:05:46 - mmengine - INFO - Iter(val) [200/200] aAcc: 99.1700 mIoU: 90.1600 mAcc: 94.7500 mFscore: 94.5900 mPrecision: 94.4300 mRecall: 94.7500 data_time: 0.0015 time: 0.1160 +04/17 14:06:36 - mmengine - INFO - Iter(train) [ 20050/160000] base_lr: 8.8297e-05 lr: 3.2645e-07 eta: 1 day, 14:51:31 time: 0.9981 data_time: 0.0046 memory: 8462 loss: 0.0221 decode.loss_ce: 0.0120 decode.acc_seg: 99.3975 aux.loss_ce: 0.0101 aux.acc_seg: 98.7370 +04/17 14:07:26 - mmengine - INFO - Iter(train) [ 20100/160000] base_lr: 8.8266e-05 lr: 3.2634e-07 eta: 1 day, 14:50:41 time: 0.9985 data_time: 0.0046 memory: 8462 loss: 0.0203 decode.loss_ce: 0.0107 decode.acc_seg: 99.5319 aux.loss_ce: 0.0096 aux.acc_seg: 98.8304 +04/17 14:08:16 - mmengine - INFO - Iter(train) [ 20150/160000] base_lr: 8.8234e-05 lr: 3.2622e-07 eta: 1 day, 14:49:51 time: 0.9998 data_time: 0.0048 memory: 8462 loss: 0.0225 decode.loss_ce: 0.0112 decode.acc_seg: 99.5188 aux.loss_ce: 0.0113 aux.acc_seg: 98.8132 +04/17 14:09:06 - mmengine - INFO - Iter(train) [ 20200/160000] base_lr: 8.8202e-05 lr: 3.2610e-07 eta: 1 day, 14:49:01 time: 1.0004 data_time: 0.0047 memory: 8462 loss: 0.0215 decode.loss_ce: 0.0116 decode.acc_seg: 99.5768 aux.loss_ce: 0.0099 aux.acc_seg: 98.7598 +04/17 14:09:55 - mmengine - INFO - Iter(train) [ 20250/160000] base_lr: 8.8171e-05 lr: 3.2599e-07 eta: 1 day, 14:48:11 time: 0.9997 data_time: 0.0050 memory: 8462 loss: 0.0212 decode.loss_ce: 0.0112 decode.acc_seg: 99.4320 aux.loss_ce: 0.0100 aux.acc_seg: 98.8493 +04/17 14:10:45 - mmengine - INFO - Iter(train) [ 20300/160000] base_lr: 8.8139e-05 lr: 3.2587e-07 eta: 1 day, 14:47:21 time: 0.9996 data_time: 0.0045 memory: 8462 loss: 0.0236 decode.loss_ce: 0.0124 decode.acc_seg: 99.3404 aux.loss_ce: 0.0112 aux.acc_seg: 98.4577 +04/17 14:11:35 - mmengine - INFO - Iter(train) [ 20350/160000] base_lr: 8.8108e-05 lr: 3.2575e-07 eta: 1 day, 14:46:31 time: 0.9991 data_time: 0.0045 memory: 8462 loss: 0.0197 decode.loss_ce: 0.0104 decode.acc_seg: 99.5899 aux.loss_ce: 0.0093 aux.acc_seg: 99.0639 +04/17 14:12:25 - mmengine - INFO - Iter(train) [ 20400/160000] base_lr: 8.8076e-05 lr: 3.2564e-07 eta: 1 day, 14:45:41 time: 0.9993 data_time: 0.0042 memory: 8462 loss: 0.0211 decode.loss_ce: 0.0110 decode.acc_seg: 99.4804 aux.loss_ce: 0.0101 aux.acc_seg: 98.7062 +04/17 14:13:15 - mmengine - INFO - Iter(train) [ 20450/160000] base_lr: 8.8045e-05 lr: 3.2552e-07 eta: 1 day, 14:44:51 time: 0.9996 data_time: 0.0051 memory: 8462 loss: 0.0206 decode.loss_ce: 0.0105 decode.acc_seg: 99.5016 aux.loss_ce: 0.0100 aux.acc_seg: 98.5600 +04/17 14:14:05 - mmengine - INFO - Iter(train) [ 20500/160000] base_lr: 8.8013e-05 lr: 3.2540e-07 eta: 1 day, 14:44:01 time: 0.9985 data_time: 0.0045 memory: 8462 loss: 0.0201 decode.loss_ce: 0.0106 decode.acc_seg: 99.5970 aux.loss_ce: 0.0094 aux.acc_seg: 98.8459 +04/17 14:14:55 - mmengine - INFO - Iter(train) [ 20550/160000] base_lr: 8.7982e-05 lr: 3.2529e-07 eta: 1 day, 14:43:11 time: 1.0003 data_time: 0.0043 memory: 8462 loss: 0.0205 decode.loss_ce: 0.0110 decode.acc_seg: 99.0553 aux.loss_ce: 0.0095 aux.acc_seg: 98.2803 +04/17 14:15:45 - mmengine - INFO - Iter(train) [ 20600/160000] base_lr: 8.7950e-05 lr: 3.2517e-07 eta: 1 day, 14:42:21 time: 0.9983 data_time: 0.0047 memory: 8462 loss: 0.0200 decode.loss_ce: 0.0100 decode.acc_seg: 99.6210 aux.loss_ce: 0.0100 aux.acc_seg: 99.0175 +04/17 14:16:35 - mmengine - INFO - Iter(train) [ 20650/160000] base_lr: 8.7919e-05 lr: 3.2505e-07 eta: 1 day, 14:41:31 time: 0.9992 data_time: 0.0042 memory: 8462 loss: 0.0239 decode.loss_ce: 0.0135 decode.acc_seg: 98.8693 aux.loss_ce: 0.0104 aux.acc_seg: 98.3368 +04/17 14:17:25 - mmengine - INFO - Iter(train) [ 20700/160000] base_lr: 8.7887e-05 lr: 3.2494e-07 eta: 1 day, 14:40:41 time: 0.9992 data_time: 0.0044 memory: 8462 loss: 0.0225 decode.loss_ce: 0.0122 decode.acc_seg: 99.4473 aux.loss_ce: 0.0103 aux.acc_seg: 98.7219 +04/17 14:18:15 - mmengine - INFO - Iter(train) [ 20750/160000] base_lr: 8.7855e-05 lr: 3.2482e-07 eta: 1 day, 14:39:51 time: 0.9991 data_time: 0.0046 memory: 8462 loss: 0.0184 decode.loss_ce: 0.0098 decode.acc_seg: 99.6065 aux.loss_ce: 0.0086 aux.acc_seg: 99.1484 +04/17 14:19:05 - mmengine - INFO - Iter(train) [ 20800/160000] base_lr: 8.7824e-05 lr: 3.2470e-07 eta: 1 day, 14:39:01 time: 0.9999 data_time: 0.0046 memory: 8462 loss: 0.0219 decode.loss_ce: 0.0115 decode.acc_seg: 99.6258 aux.loss_ce: 0.0105 aux.acc_seg: 99.0911 +04/17 14:19:55 - mmengine - INFO - Iter(train) [ 20850/160000] base_lr: 8.7792e-05 lr: 3.2459e-07 eta: 1 day, 14:38:11 time: 1.0000 data_time: 0.0046 memory: 8462 loss: 0.0227 decode.loss_ce: 0.0121 decode.acc_seg: 99.5569 aux.loss_ce: 0.0106 aux.acc_seg: 99.1344 +04/17 14:20:45 - mmengine - INFO - Iter(train) [ 20900/160000] base_lr: 8.7761e-05 lr: 3.2447e-07 eta: 1 day, 14:37:21 time: 0.9985 data_time: 0.0044 memory: 8462 loss: 0.0223 decode.loss_ce: 0.0117 decode.acc_seg: 99.2662 aux.loss_ce: 0.0106 aux.acc_seg: 98.4653 +04/17 14:21:35 - mmengine - INFO - Iter(train) [ 20950/160000] base_lr: 8.7729e-05 lr: 3.2435e-07 eta: 1 day, 14:36:31 time: 1.0004 data_time: 0.0043 memory: 8462 loss: 0.0216 decode.loss_ce: 0.0111 decode.acc_seg: 99.4967 aux.loss_ce: 0.0105 aux.acc_seg: 98.8319 +04/17 14:22:25 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 14:22:25 - mmengine - INFO - Iter(train) [ 21000/160000] base_lr: 8.7698e-05 lr: 3.2424e-07 eta: 1 day, 14:35:41 time: 0.9986 data_time: 0.0046 memory: 8462 loss: 0.0207 decode.loss_ce: 0.0109 decode.acc_seg: 99.6061 aux.loss_ce: 0.0099 aux.acc_seg: 99.0068 +04/17 14:23:15 - mmengine - INFO - Iter(train) [ 21050/160000] base_lr: 8.7666e-05 lr: 3.2412e-07 eta: 1 day, 14:34:51 time: 0.9994 data_time: 0.0047 memory: 8462 loss: 0.0223 decode.loss_ce: 0.0108 decode.acc_seg: 99.4978 aux.loss_ce: 0.0115 aux.acc_seg: 99.0391 +04/17 14:24:05 - mmengine - INFO - Iter(train) [ 21100/160000] base_lr: 8.7635e-05 lr: 3.2400e-07 eta: 1 day, 14:34:01 time: 1.0004 data_time: 0.0046 memory: 8462 loss: 0.0204 decode.loss_ce: 0.0109 decode.acc_seg: 99.3053 aux.loss_ce: 0.0095 aux.acc_seg: 98.8251 +04/17 14:24:55 - mmengine - INFO - Iter(train) [ 21150/160000] base_lr: 8.7603e-05 lr: 3.2389e-07 eta: 1 day, 14:33:11 time: 1.0008 data_time: 0.0047 memory: 8462 loss: 0.0216 decode.loss_ce: 0.0116 decode.acc_seg: 99.6618 aux.loss_ce: 0.0101 aux.acc_seg: 99.1035 +04/17 14:25:45 - mmengine - INFO - Iter(train) [ 21200/160000] base_lr: 8.7572e-05 lr: 3.2377e-07 eta: 1 day, 14:32:21 time: 1.0010 data_time: 0.0047 memory: 8462 loss: 0.0201 decode.loss_ce: 0.0108 decode.acc_seg: 99.4436 aux.loss_ce: 0.0094 aux.acc_seg: 98.6851 +04/17 14:26:35 - mmengine - INFO - Iter(train) [ 21250/160000] base_lr: 8.7540e-05 lr: 3.2365e-07 eta: 1 day, 14:31:31 time: 0.9999 data_time: 0.0046 memory: 8462 loss: 0.0194 decode.loss_ce: 0.0099 decode.acc_seg: 99.5758 aux.loss_ce: 0.0095 aux.acc_seg: 99.0086 +04/17 14:27:25 - mmengine - INFO - Iter(train) [ 21300/160000] base_lr: 8.7508e-05 lr: 3.2354e-07 eta: 1 day, 14:30:41 time: 0.9984 data_time: 0.0044 memory: 8462 loss: 0.0211 decode.loss_ce: 0.0109 decode.acc_seg: 99.5089 aux.loss_ce: 0.0102 aux.acc_seg: 98.9325 +04/17 14:28:15 - mmengine - INFO - Iter(train) [ 21350/160000] base_lr: 8.7477e-05 lr: 3.2342e-07 eta: 1 day, 14:29:50 time: 0.9979 data_time: 0.0044 memory: 8462 loss: 0.0240 decode.loss_ce: 0.0124 decode.acc_seg: 99.1619 aux.loss_ce: 0.0117 aux.acc_seg: 97.6984 +04/17 14:29:05 - mmengine - INFO - Iter(train) [ 21400/160000] base_lr: 8.7445e-05 lr: 3.2330e-07 eta: 1 day, 14:29:00 time: 1.0005 data_time: 0.0045 memory: 8462 loss: 0.0201 decode.loss_ce: 0.0106 decode.acc_seg: 99.6611 aux.loss_ce: 0.0095 aux.acc_seg: 99.1283 +04/17 14:29:55 - mmengine - INFO - Iter(train) [ 21450/160000] base_lr: 8.7414e-05 lr: 3.2319e-07 eta: 1 day, 14:28:10 time: 0.9994 data_time: 0.0043 memory: 8462 loss: 0.0186 decode.loss_ce: 0.0095 decode.acc_seg: 99.4682 aux.loss_ce: 0.0091 aux.acc_seg: 98.8892 +04/17 14:30:45 - mmengine - INFO - Iter(train) [ 21500/160000] base_lr: 8.7382e-05 lr: 3.2307e-07 eta: 1 day, 14:27:20 time: 0.9987 data_time: 0.0045 memory: 8462 loss: 0.0205 decode.loss_ce: 0.0114 decode.acc_seg: 99.5121 aux.loss_ce: 0.0091 aux.acc_seg: 99.0715 +04/17 14:31:35 - mmengine - INFO - Iter(train) [ 21550/160000] base_lr: 8.7351e-05 lr: 3.2295e-07 eta: 1 day, 14:26:30 time: 0.9996 data_time: 0.0045 memory: 8462 loss: 0.0211 decode.loss_ce: 0.0112 decode.acc_seg: 99.4154 aux.loss_ce: 0.0099 aux.acc_seg: 99.1652 +04/17 14:32:25 - mmengine - INFO - Iter(train) [ 21600/160000] base_lr: 8.7319e-05 lr: 3.2284e-07 eta: 1 day, 14:25:40 time: 1.0001 data_time: 0.0053 memory: 8462 loss: 0.0215 decode.loss_ce: 0.0113 decode.acc_seg: 99.2371 aux.loss_ce: 0.0102 aux.acc_seg: 98.2595 +04/17 14:33:15 - mmengine - INFO - Iter(train) [ 21650/160000] base_lr: 8.7288e-05 lr: 3.2272e-07 eta: 1 day, 14:24:50 time: 0.9995 data_time: 0.0043 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0088 decode.acc_seg: 99.5977 aux.loss_ce: 0.0086 aux.acc_seg: 98.9861 +04/17 14:34:05 - mmengine - INFO - Iter(train) [ 21700/160000] base_lr: 8.7256e-05 lr: 3.2260e-07 eta: 1 day, 14:24:00 time: 0.9986 data_time: 0.0043 memory: 8462 loss: 0.0192 decode.loss_ce: 0.0099 decode.acc_seg: 99.7166 aux.loss_ce: 0.0093 aux.acc_seg: 99.2840 +04/17 14:34:55 - mmengine - INFO - Iter(train) [ 21750/160000] base_lr: 8.7225e-05 lr: 3.2249e-07 eta: 1 day, 14:23:10 time: 1.0003 data_time: 0.0045 memory: 8462 loss: 0.0229 decode.loss_ce: 0.0120 decode.acc_seg: 99.5123 aux.loss_ce: 0.0109 aux.acc_seg: 98.7047 +04/17 14:35:45 - mmengine - INFO - Iter(train) [ 21800/160000] base_lr: 8.7193e-05 lr: 3.2237e-07 eta: 1 day, 14:22:20 time: 0.9993 data_time: 0.0042 memory: 8462 loss: 0.0218 decode.loss_ce: 0.0117 decode.acc_seg: 99.4455 aux.loss_ce: 0.0101 aux.acc_seg: 98.8354 +04/17 14:36:35 - mmengine - INFO - Iter(train) [ 21850/160000] base_lr: 8.7161e-05 lr: 3.2225e-07 eta: 1 day, 14:21:30 time: 0.9993 data_time: 0.0042 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0099 decode.acc_seg: 99.6090 aux.loss_ce: 0.0091 aux.acc_seg: 99.1480 +04/17 14:37:25 - mmengine - INFO - Iter(train) [ 21900/160000] base_lr: 8.7130e-05 lr: 3.2214e-07 eta: 1 day, 14:20:40 time: 0.9999 data_time: 0.0048 memory: 8462 loss: 0.0214 decode.loss_ce: 0.0112 decode.acc_seg: 99.3156 aux.loss_ce: 0.0103 aux.acc_seg: 98.5327 +04/17 14:38:15 - mmengine - INFO - Iter(train) [ 21950/160000] base_lr: 8.7098e-05 lr: 3.2202e-07 eta: 1 day, 14:19:50 time: 0.9998 data_time: 0.0047 memory: 8462 loss: 0.0185 decode.loss_ce: 0.0095 decode.acc_seg: 99.6925 aux.loss_ce: 0.0090 aux.acc_seg: 99.2367 +04/17 14:39:05 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 14:39:05 - mmengine - INFO - Iter(train) [ 22000/160000] base_lr: 8.7067e-05 lr: 3.2190e-07 eta: 1 day, 14:19:00 time: 0.9991 data_time: 0.0045 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0087 decode.acc_seg: 99.7269 aux.loss_ce: 0.0086 aux.acc_seg: 99.2006 +04/17 14:39:55 - mmengine - INFO - Iter(train) [ 22050/160000] base_lr: 8.7035e-05 lr: 3.2179e-07 eta: 1 day, 14:18:10 time: 0.9997 data_time: 0.0046 memory: 8462 loss: 0.0239 decode.loss_ce: 0.0123 decode.acc_seg: 99.4081 aux.loss_ce: 0.0116 aux.acc_seg: 98.3019 +04/17 14:40:45 - mmengine - INFO - Iter(train) [ 22100/160000] base_lr: 8.7004e-05 lr: 3.2167e-07 eta: 1 day, 14:17:21 time: 1.0005 data_time: 0.0046 memory: 8462 loss: 0.0233 decode.loss_ce: 0.0126 decode.acc_seg: 99.6626 aux.loss_ce: 0.0107 aux.acc_seg: 99.1590 +04/17 14:41:35 - mmengine - INFO - Iter(train) [ 22150/160000] base_lr: 8.6972e-05 lr: 3.2155e-07 eta: 1 day, 14:16:31 time: 1.0003 data_time: 0.0047 memory: 8462 loss: 0.0198 decode.loss_ce: 0.0105 decode.acc_seg: 99.7681 aux.loss_ce: 0.0093 aux.acc_seg: 99.4343 +04/17 14:42:25 - mmengine - INFO - Iter(train) [ 22200/160000] base_lr: 8.6941e-05 lr: 3.2144e-07 eta: 1 day, 14:15:41 time: 0.9997 data_time: 0.0048 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0085 decode.acc_seg: 99.6647 aux.loss_ce: 0.0089 aux.acc_seg: 99.1859 +04/17 14:43:15 - mmengine - INFO - Iter(train) [ 22250/160000] base_lr: 8.6909e-05 lr: 3.2132e-07 eta: 1 day, 14:14:51 time: 0.9980 data_time: 0.0045 memory: 8462 loss: 0.0213 decode.loss_ce: 0.0112 decode.acc_seg: 99.6481 aux.loss_ce: 0.0101 aux.acc_seg: 99.2300 +04/17 14:44:05 - mmengine - INFO - Iter(train) [ 22300/160000] base_lr: 8.6878e-05 lr: 3.2120e-07 eta: 1 day, 14:14:01 time: 0.9993 data_time: 0.0046 memory: 8462 loss: 0.0169 decode.loss_ce: 0.0089 decode.acc_seg: 99.6572 aux.loss_ce: 0.0080 aux.acc_seg: 99.1638 +04/17 14:44:55 - mmengine - INFO - Iter(train) [ 22350/160000] base_lr: 8.6846e-05 lr: 3.2109e-07 eta: 1 day, 14:13:11 time: 0.9998 data_time: 0.0043 memory: 8462 loss: 0.0200 decode.loss_ce: 0.0098 decode.acc_seg: 99.5817 aux.loss_ce: 0.0102 aux.acc_seg: 99.0046 +04/17 14:45:45 - mmengine - INFO - Iter(train) [ 22400/160000] base_lr: 8.6814e-05 lr: 3.2097e-07 eta: 1 day, 14:12:21 time: 1.0010 data_time: 0.0053 memory: 8462 loss: 0.0213 decode.loss_ce: 0.0114 decode.acc_seg: 99.5935 aux.loss_ce: 0.0100 aux.acc_seg: 98.8609 +04/17 14:46:35 - mmengine - INFO - Iter(train) [ 22450/160000] base_lr: 8.6783e-05 lr: 3.2085e-07 eta: 1 day, 14:11:31 time: 1.0001 data_time: 0.0044 memory: 8462 loss: 0.0209 decode.loss_ce: 0.0114 decode.acc_seg: 99.6431 aux.loss_ce: 0.0095 aux.acc_seg: 99.1547 +04/17 14:47:25 - mmengine - INFO - Iter(train) [ 22500/160000] base_lr: 8.6751e-05 lr: 3.2074e-07 eta: 1 day, 14:10:41 time: 0.9988 data_time: 0.0043 memory: 8462 loss: 0.0192 decode.loss_ce: 0.0099 decode.acc_seg: 99.6481 aux.loss_ce: 0.0093 aux.acc_seg: 98.8970 +04/17 14:48:15 - mmengine - INFO - Iter(train) [ 22550/160000] base_lr: 8.6720e-05 lr: 3.2062e-07 eta: 1 day, 14:09:51 time: 1.0005 data_time: 0.0049 memory: 8462 loss: 0.0199 decode.loss_ce: 0.0107 decode.acc_seg: 99.6973 aux.loss_ce: 0.0092 aux.acc_seg: 98.9571 +04/17 14:49:05 - mmengine - INFO - Iter(train) [ 22600/160000] base_lr: 8.6688e-05 lr: 3.2050e-07 eta: 1 day, 14:09:01 time: 0.9994 data_time: 0.0042 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0088 decode.acc_seg: 99.5554 aux.loss_ce: 0.0086 aux.acc_seg: 99.0030 +04/17 14:49:55 - mmengine - INFO - Iter(train) [ 22650/160000] base_lr: 8.6657e-05 lr: 3.2039e-07 eta: 1 day, 14:08:12 time: 1.0015 data_time: 0.0052 memory: 8462 loss: 0.0217 decode.loss_ce: 0.0113 decode.acc_seg: 99.2714 aux.loss_ce: 0.0104 aux.acc_seg: 98.3213 +04/17 14:50:45 - mmengine - INFO - Iter(train) [ 22700/160000] base_lr: 8.6625e-05 lr: 3.2027e-07 eta: 1 day, 14:07:22 time: 1.0001 data_time: 0.0046 memory: 8462 loss: 0.0232 decode.loss_ce: 0.0122 decode.acc_seg: 99.4961 aux.loss_ce: 0.0110 aux.acc_seg: 99.1377 +04/17 14:51:34 - mmengine - INFO - Iter(train) [ 22750/160000] base_lr: 8.6594e-05 lr: 3.2015e-07 eta: 1 day, 14:06:32 time: 0.9989 data_time: 0.0042 memory: 8462 loss: 0.0165 decode.loss_ce: 0.0087 decode.acc_seg: 99.7097 aux.loss_ce: 0.0078 aux.acc_seg: 99.4293 +04/17 14:52:24 - mmengine - INFO - Iter(train) [ 22800/160000] base_lr: 8.6562e-05 lr: 3.2004e-07 eta: 1 day, 14:05:42 time: 1.0002 data_time: 0.0046 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0096 decode.acc_seg: 99.4398 aux.loss_ce: 0.0094 aux.acc_seg: 99.2588 +04/17 14:53:14 - mmengine - INFO - Iter(train) [ 22850/160000] base_lr: 8.6531e-05 lr: 3.1992e-07 eta: 1 day, 14:04:51 time: 0.9991 data_time: 0.0043 memory: 8462 loss: 0.0224 decode.loss_ce: 0.0119 decode.acc_seg: 99.4925 aux.loss_ce: 0.0105 aux.acc_seg: 98.6650 +04/17 14:54:04 - mmengine - INFO - Iter(train) [ 22900/160000] base_lr: 8.6499e-05 lr: 3.1980e-07 eta: 1 day, 14:04:02 time: 1.0002 data_time: 0.0047 memory: 8462 loss: 0.0209 decode.loss_ce: 0.0105 decode.acc_seg: 99.4282 aux.loss_ce: 0.0103 aux.acc_seg: 98.7370 +04/17 14:54:54 - mmengine - INFO - Iter(train) [ 22950/160000] base_lr: 8.6467e-05 lr: 3.1969e-07 eta: 1 day, 14:03:11 time: 0.9989 data_time: 0.0044 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0095 decode.acc_seg: 99.5348 aux.loss_ce: 0.0084 aux.acc_seg: 98.6467 +04/17 14:55:44 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 14:55:44 - mmengine - INFO - Iter(train) [ 23000/160000] base_lr: 8.6436e-05 lr: 3.1957e-07 eta: 1 day, 14:02:21 time: 0.9997 data_time: 0.0043 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0089 decode.acc_seg: 99.5611 aux.loss_ce: 0.0087 aux.acc_seg: 99.0721 +04/17 14:56:34 - mmengine - INFO - Iter(train) [ 23050/160000] base_lr: 8.6404e-05 lr: 3.1945e-07 eta: 1 day, 14:01:32 time: 1.0006 data_time: 0.0043 memory: 8462 loss: 0.0197 decode.loss_ce: 0.0100 decode.acc_seg: 99.4980 aux.loss_ce: 0.0097 aux.acc_seg: 98.8644 +04/17 14:57:24 - mmengine - INFO - Iter(train) [ 23100/160000] base_lr: 8.6373e-05 lr: 3.1934e-07 eta: 1 day, 14:00:42 time: 0.9997 data_time: 0.0043 memory: 8462 loss: 0.0189 decode.loss_ce: 0.0101 decode.acc_seg: 99.5207 aux.loss_ce: 0.0088 aux.acc_seg: 98.8394 +04/17 14:58:14 - mmengine - INFO - Iter(train) [ 23150/160000] base_lr: 8.6341e-05 lr: 3.1922e-07 eta: 1 day, 13:59:52 time: 0.9995 data_time: 0.0043 memory: 8462 loss: 0.0194 decode.loss_ce: 0.0097 decode.acc_seg: 99.5272 aux.loss_ce: 0.0097 aux.acc_seg: 98.7566 +04/17 14:59:04 - mmengine - INFO - Iter(train) [ 23200/160000] base_lr: 8.6310e-05 lr: 3.1910e-07 eta: 1 day, 13:59:02 time: 0.9996 data_time: 0.0041 memory: 8462 loss: 0.0202 decode.loss_ce: 0.0103 decode.acc_seg: 99.6855 aux.loss_ce: 0.0099 aux.acc_seg: 99.0540 +04/17 14:59:54 - mmengine - INFO - Iter(train) [ 23250/160000] base_lr: 8.6278e-05 lr: 3.1899e-07 eta: 1 day, 13:58:12 time: 0.9996 data_time: 0.0041 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0088 decode.acc_seg: 99.7057 aux.loss_ce: 0.0089 aux.acc_seg: 99.1768 +04/17 15:00:44 - mmengine - INFO - Iter(train) [ 23300/160000] base_lr: 8.6247e-05 lr: 3.1887e-07 eta: 1 day, 13:57:22 time: 0.9991 data_time: 0.0051 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0100 decode.acc_seg: 99.4694 aux.loss_ce: 0.0090 aux.acc_seg: 99.0191 +04/17 15:01:34 - mmengine - INFO - Iter(train) [ 23350/160000] base_lr: 8.6215e-05 lr: 3.1875e-07 eta: 1 day, 13:56:32 time: 0.9996 data_time: 0.0047 memory: 8462 loss: 0.0165 decode.loss_ce: 0.0086 decode.acc_seg: 99.7654 aux.loss_ce: 0.0079 aux.acc_seg: 99.2838 +04/17 15:02:24 - mmengine - INFO - Iter(train) [ 23400/160000] base_lr: 8.6184e-05 lr: 3.1864e-07 eta: 1 day, 13:55:42 time: 1.0003 data_time: 0.0049 memory: 8462 loss: 0.0213 decode.loss_ce: 0.0108 decode.acc_seg: 99.2485 aux.loss_ce: 0.0105 aux.acc_seg: 98.4438 +04/17 15:03:14 - mmengine - INFO - Iter(train) [ 23450/160000] base_lr: 8.6152e-05 lr: 3.1852e-07 eta: 1 day, 13:54:52 time: 1.0005 data_time: 0.0043 memory: 8462 loss: 0.0228 decode.loss_ce: 0.0119 decode.acc_seg: 99.4175 aux.loss_ce: 0.0108 aux.acc_seg: 98.8270 +04/17 15:04:04 - mmengine - INFO - Iter(train) [ 23500/160000] base_lr: 8.6120e-05 lr: 3.1840e-07 eta: 1 day, 13:54:03 time: 0.9993 data_time: 0.0049 memory: 8462 loss: 0.0209 decode.loss_ce: 0.0110 decode.acc_seg: 99.1472 aux.loss_ce: 0.0099 aux.acc_seg: 97.9836 +04/17 15:04:54 - mmengine - INFO - Iter(train) [ 23550/160000] base_lr: 8.6089e-05 lr: 3.1829e-07 eta: 1 day, 13:53:13 time: 0.9997 data_time: 0.0045 memory: 8462 loss: 0.0239 decode.loss_ce: 0.0121 decode.acc_seg: 99.1421 aux.loss_ce: 0.0118 aux.acc_seg: 98.3263 +04/17 15:05:44 - mmengine - INFO - Iter(train) [ 23600/160000] base_lr: 8.6057e-05 lr: 3.1817e-07 eta: 1 day, 13:52:23 time: 1.0008 data_time: 0.0045 memory: 8462 loss: 0.0199 decode.loss_ce: 0.0108 decode.acc_seg: 99.6548 aux.loss_ce: 0.0091 aux.acc_seg: 99.1447 +04/17 15:06:34 - mmengine - INFO - Iter(train) [ 23650/160000] base_lr: 8.6026e-05 lr: 3.1806e-07 eta: 1 day, 13:51:33 time: 1.0012 data_time: 0.0044 memory: 8462 loss: 0.0213 decode.loss_ce: 0.0116 decode.acc_seg: 99.5560 aux.loss_ce: 0.0097 aux.acc_seg: 98.9267 +04/17 15:07:24 - mmengine - INFO - Iter(train) [ 23700/160000] base_lr: 8.5994e-05 lr: 3.1794e-07 eta: 1 day, 13:50:43 time: 0.9991 data_time: 0.0044 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0079 decode.acc_seg: 99.6264 aux.loss_ce: 0.0081 aux.acc_seg: 98.8258 +04/17 15:08:14 - mmengine - INFO - Iter(train) [ 23750/160000] base_lr: 8.5963e-05 lr: 3.1782e-07 eta: 1 day, 13:49:53 time: 0.9989 data_time: 0.0046 memory: 8462 loss: 0.0199 decode.loss_ce: 0.0105 decode.acc_seg: 99.6782 aux.loss_ce: 0.0094 aux.acc_seg: 98.9418 +04/17 15:09:04 - mmengine - INFO - Iter(train) [ 23800/160000] base_lr: 8.5931e-05 lr: 3.1771e-07 eta: 1 day, 13:49:03 time: 0.9997 data_time: 0.0044 memory: 8462 loss: 0.0199 decode.loss_ce: 0.0103 decode.acc_seg: 99.7223 aux.loss_ce: 0.0096 aux.acc_seg: 99.2645 +04/17 15:09:54 - mmengine - INFO - Iter(train) [ 23850/160000] base_lr: 8.5900e-05 lr: 3.1759e-07 eta: 1 day, 13:48:13 time: 1.0001 data_time: 0.0049 memory: 8462 loss: 0.0216 decode.loss_ce: 0.0113 decode.acc_seg: 99.6099 aux.loss_ce: 0.0103 aux.acc_seg: 98.5582 +04/17 15:10:44 - mmengine - INFO - Iter(train) [ 23900/160000] base_lr: 8.5868e-05 lr: 3.1747e-07 eta: 1 day, 13:47:23 time: 0.9998 data_time: 0.0048 memory: 8462 loss: 0.0192 decode.loss_ce: 0.0097 decode.acc_seg: 99.6197 aux.loss_ce: 0.0095 aux.acc_seg: 99.0349 +04/17 15:11:34 - mmengine - INFO - Iter(train) [ 23950/160000] base_lr: 8.5837e-05 lr: 3.1736e-07 eta: 1 day, 13:46:33 time: 0.9994 data_time: 0.0050 memory: 8462 loss: 0.0224 decode.loss_ce: 0.0117 decode.acc_seg: 99.5115 aux.loss_ce: 0.0107 aux.acc_seg: 98.7293 +04/17 15:12:24 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 15:12:24 - mmengine - INFO - Iter(train) [ 24000/160000] base_lr: 8.5805e-05 lr: 3.1724e-07 eta: 1 day, 13:45:43 time: 0.9991 data_time: 0.0049 memory: 8462 loss: 0.0198 decode.loss_ce: 0.0102 decode.acc_seg: 99.6090 aux.loss_ce: 0.0096 aux.acc_seg: 99.2201 +04/17 15:13:14 - mmengine - INFO - Iter(train) [ 24050/160000] base_lr: 8.5773e-05 lr: 3.1712e-07 eta: 1 day, 13:44:53 time: 1.0017 data_time: 0.0044 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0087 decode.acc_seg: 99.5626 aux.loss_ce: 0.0091 aux.acc_seg: 98.9571 +04/17 15:14:04 - mmengine - INFO - Iter(train) [ 24100/160000] base_lr: 8.5742e-05 lr: 3.1701e-07 eta: 1 day, 13:44:03 time: 1.0002 data_time: 0.0046 memory: 8462 loss: 0.0191 decode.loss_ce: 0.0099 decode.acc_seg: 99.3309 aux.loss_ce: 0.0092 aux.acc_seg: 98.7541 +04/17 15:14:54 - mmengine - INFO - Iter(train) [ 24150/160000] base_lr: 8.5710e-05 lr: 3.1689e-07 eta: 1 day, 13:43:13 time: 0.9984 data_time: 0.0045 memory: 8462 loss: 0.0236 decode.loss_ce: 0.0123 decode.acc_seg: 99.6435 aux.loss_ce: 0.0113 aux.acc_seg: 98.7827 +04/17 15:15:44 - mmengine - INFO - Iter(train) [ 24200/160000] base_lr: 8.5679e-05 lr: 3.1677e-07 eta: 1 day, 13:42:23 time: 1.0002 data_time: 0.0046 memory: 8462 loss: 0.0204 decode.loss_ce: 0.0103 decode.acc_seg: 99.6401 aux.loss_ce: 0.0101 aux.acc_seg: 98.8926 +04/17 15:16:34 - mmengine - INFO - Iter(train) [ 24250/160000] base_lr: 8.5647e-05 lr: 3.1666e-07 eta: 1 day, 13:41:33 time: 0.9988 data_time: 0.0044 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0092 decode.acc_seg: 99.5783 aux.loss_ce: 0.0099 aux.acc_seg: 99.0211 +04/17 15:17:24 - mmengine - INFO - Iter(train) [ 24300/160000] base_lr: 8.5616e-05 lr: 3.1654e-07 eta: 1 day, 13:40:43 time: 0.9982 data_time: 0.0044 memory: 8462 loss: 0.0164 decode.loss_ce: 0.0090 decode.acc_seg: 99.5420 aux.loss_ce: 0.0075 aux.acc_seg: 99.0318 +04/17 15:18:14 - mmengine - INFO - Iter(train) [ 24350/160000] base_lr: 8.5584e-05 lr: 3.1642e-07 eta: 1 day, 13:39:53 time: 0.9995 data_time: 0.0046 memory: 8462 loss: 0.0191 decode.loss_ce: 0.0098 decode.acc_seg: 99.6916 aux.loss_ce: 0.0093 aux.acc_seg: 99.1362 +04/17 15:19:04 - mmengine - INFO - Iter(train) [ 24400/160000] base_lr: 8.5553e-05 lr: 3.1631e-07 eta: 1 day, 13:39:03 time: 0.9997 data_time: 0.0047 memory: 8462 loss: 0.0196 decode.loss_ce: 0.0098 decode.acc_seg: 99.6367 aux.loss_ce: 0.0098 aux.acc_seg: 99.1823 +04/17 15:19:54 - mmengine - INFO - Iter(train) [ 24450/160000] base_lr: 8.5521e-05 lr: 3.1619e-07 eta: 1 day, 13:38:14 time: 1.0001 data_time: 0.0046 memory: 8462 loss: 0.0196 decode.loss_ce: 0.0104 decode.acc_seg: 99.5043 aux.loss_ce: 0.0092 aux.acc_seg: 98.9340 +04/17 15:20:44 - mmengine - INFO - Iter(train) [ 24500/160000] base_lr: 8.5489e-05 lr: 3.1607e-07 eta: 1 day, 13:37:23 time: 0.9998 data_time: 0.0045 memory: 8462 loss: 0.0194 decode.loss_ce: 0.0098 decode.acc_seg: 99.3469 aux.loss_ce: 0.0096 aux.acc_seg: 98.7484 +04/17 15:21:34 - mmengine - INFO - Iter(train) [ 24550/160000] base_lr: 8.5458e-05 lr: 3.1596e-07 eta: 1 day, 13:36:34 time: 1.0002 data_time: 0.0045 memory: 8462 loss: 0.0201 decode.loss_ce: 0.0102 decode.acc_seg: 99.5737 aux.loss_ce: 0.0099 aux.acc_seg: 98.8726 +04/17 15:22:24 - mmengine - INFO - Iter(train) [ 24600/160000] base_lr: 8.5426e-05 lr: 3.1584e-07 eta: 1 day, 13:35:44 time: 1.0013 data_time: 0.0045 memory: 8462 loss: 0.0187 decode.loss_ce: 0.0093 decode.acc_seg: 99.5886 aux.loss_ce: 0.0093 aux.acc_seg: 98.9346 +04/17 15:23:14 - mmengine - INFO - Iter(train) [ 24650/160000] base_lr: 8.5395e-05 lr: 3.1572e-07 eta: 1 day, 13:34:54 time: 0.9982 data_time: 0.0043 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0089 decode.acc_seg: 99.7517 aux.loss_ce: 0.0087 aux.acc_seg: 99.3788 +04/17 15:24:04 - mmengine - INFO - Iter(train) [ 24700/160000] base_lr: 8.5363e-05 lr: 3.1561e-07 eta: 1 day, 13:34:04 time: 0.9993 data_time: 0.0046 memory: 8462 loss: 0.0185 decode.loss_ce: 0.0090 decode.acc_seg: 99.5827 aux.loss_ce: 0.0095 aux.acc_seg: 98.7396 +04/17 15:24:54 - mmengine - INFO - Iter(train) [ 24750/160000] base_lr: 8.5332e-05 lr: 3.1549e-07 eta: 1 day, 13:33:14 time: 0.9988 data_time: 0.0046 memory: 8462 loss: 0.0229 decode.loss_ce: 0.0124 decode.acc_seg: 99.6279 aux.loss_ce: 0.0105 aux.acc_seg: 98.9002 +04/17 15:25:44 - mmengine - INFO - Iter(train) [ 24800/160000] base_lr: 8.5300e-05 lr: 3.1537e-07 eta: 1 day, 13:32:24 time: 0.9988 data_time: 0.0045 memory: 8462 loss: 0.0200 decode.loss_ce: 0.0100 decode.acc_seg: 99.5020 aux.loss_ce: 0.0099 aux.acc_seg: 98.9683 +04/17 15:26:34 - mmengine - INFO - Iter(train) [ 24850/160000] base_lr: 8.5269e-05 lr: 3.1526e-07 eta: 1 day, 13:31:34 time: 1.0006 data_time: 0.0050 memory: 8462 loss: 0.0202 decode.loss_ce: 0.0103 decode.acc_seg: 99.4736 aux.loss_ce: 0.0099 aux.acc_seg: 98.9622 +04/17 15:27:24 - mmengine - INFO - Iter(train) [ 24900/160000] base_lr: 8.5237e-05 lr: 3.1514e-07 eta: 1 day, 13:30:44 time: 0.9998 data_time: 0.0044 memory: 8462 loss: 0.0201 decode.loss_ce: 0.0099 decode.acc_seg: 99.5302 aux.loss_ce: 0.0102 aux.acc_seg: 98.9445 +04/17 15:28:14 - mmengine - INFO - Iter(train) [ 24950/160000] base_lr: 8.5206e-05 lr: 3.1502e-07 eta: 1 day, 13:29:54 time: 0.9981 data_time: 0.0046 memory: 8462 loss: 0.0183 decode.loss_ce: 0.0095 decode.acc_seg: 99.5508 aux.loss_ce: 0.0089 aux.acc_seg: 98.8819 +04/17 15:29:04 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 15:29:04 - mmengine - INFO - Iter(train) [ 25000/160000] base_lr: 8.5174e-05 lr: 3.1491e-07 eta: 1 day, 13:29:04 time: 0.9985 data_time: 0.0044 memory: 8462 loss: 0.0216 decode.loss_ce: 0.0109 decode.acc_seg: 99.5039 aux.loss_ce: 0.0106 aux.acc_seg: 98.5380 +04/17 15:29:54 - mmengine - INFO - Iter(train) [ 25050/160000] base_lr: 8.5142e-05 lr: 3.1479e-07 eta: 1 day, 13:28:14 time: 1.0018 data_time: 0.0046 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0091 decode.acc_seg: 99.5274 aux.loss_ce: 0.0087 aux.acc_seg: 98.6612 +04/17 15:30:44 - mmengine - INFO - Iter(train) [ 25100/160000] base_lr: 8.5111e-05 lr: 3.1467e-07 eta: 1 day, 13:27:24 time: 0.9999 data_time: 0.0042 memory: 8462 loss: 0.0213 decode.loss_ce: 0.0108 decode.acc_seg: 99.4074 aux.loss_ce: 0.0105 aux.acc_seg: 98.5809 +04/17 15:31:34 - mmengine - INFO - Iter(train) [ 25150/160000] base_lr: 8.5079e-05 lr: 3.1456e-07 eta: 1 day, 13:26:34 time: 0.9998 data_time: 0.0046 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0084 decode.acc_seg: 99.6000 aux.loss_ce: 0.0086 aux.acc_seg: 98.8234 +04/17 15:32:24 - mmengine - INFO - Iter(train) [ 25200/160000] base_lr: 8.5048e-05 lr: 3.1444e-07 eta: 1 day, 13:25:44 time: 0.9993 data_time: 0.0043 memory: 8462 loss: 0.0209 decode.loss_ce: 0.0106 decode.acc_seg: 99.3948 aux.loss_ce: 0.0103 aux.acc_seg: 98.6862 +04/17 15:33:14 - mmengine - INFO - Iter(train) [ 25250/160000] base_lr: 8.5016e-05 lr: 3.1432e-07 eta: 1 day, 13:24:54 time: 0.9995 data_time: 0.0048 memory: 8462 loss: 0.0207 decode.loss_ce: 0.0102 decode.acc_seg: 99.6029 aux.loss_ce: 0.0104 aux.acc_seg: 98.9231 +04/17 15:34:04 - mmengine - INFO - Iter(train) [ 25300/160000] base_lr: 8.4985e-05 lr: 3.1421e-07 eta: 1 day, 13:24:04 time: 1.0018 data_time: 0.0043 memory: 8462 loss: 0.0187 decode.loss_ce: 0.0096 decode.acc_seg: 99.5554 aux.loss_ce: 0.0091 aux.acc_seg: 98.9813 +04/17 15:34:54 - mmengine - INFO - Iter(train) [ 25350/160000] base_lr: 8.4953e-05 lr: 3.1409e-07 eta: 1 day, 13:23:14 time: 0.9996 data_time: 0.0043 memory: 8462 loss: 0.0197 decode.loss_ce: 0.0104 decode.acc_seg: 99.7545 aux.loss_ce: 0.0093 aux.acc_seg: 99.3500 +04/17 15:35:44 - mmengine - INFO - Iter(train) [ 25400/160000] base_lr: 8.4922e-05 lr: 3.1397e-07 eta: 1 day, 13:22:24 time: 1.0003 data_time: 0.0056 memory: 8462 loss: 0.0189 decode.loss_ce: 0.0096 decode.acc_seg: 99.7293 aux.loss_ce: 0.0092 aux.acc_seg: 99.2653 +04/17 15:36:34 - mmengine - INFO - Iter(train) [ 25450/160000] base_lr: 8.4890e-05 lr: 3.1386e-07 eta: 1 day, 13:21:34 time: 0.9994 data_time: 0.0042 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0091 decode.acc_seg: 99.5956 aux.loss_ce: 0.0086 aux.acc_seg: 99.1735 +04/17 15:37:24 - mmengine - INFO - Iter(train) [ 25500/160000] base_lr: 8.4859e-05 lr: 3.1374e-07 eta: 1 day, 13:20:45 time: 0.9998 data_time: 0.0046 memory: 8462 loss: 0.0199 decode.loss_ce: 0.0101 decode.acc_seg: 99.7757 aux.loss_ce: 0.0098 aux.acc_seg: 99.2811 +04/17 15:38:14 - mmengine - INFO - Iter(train) [ 25550/160000] base_lr: 8.4827e-05 lr: 3.1362e-07 eta: 1 day, 13:19:55 time: 0.9994 data_time: 0.0044 memory: 8462 loss: 0.0201 decode.loss_ce: 0.0105 decode.acc_seg: 99.6950 aux.loss_ce: 0.0096 aux.acc_seg: 99.2168 +04/17 15:39:04 - mmengine - INFO - Iter(train) [ 25600/160000] base_lr: 8.4795e-05 lr: 3.1351e-07 eta: 1 day, 13:19:05 time: 0.9999 data_time: 0.0043 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0091 decode.acc_seg: 99.5350 aux.loss_ce: 0.0084 aux.acc_seg: 99.0034 +04/17 15:39:54 - mmengine - INFO - Iter(train) [ 25650/160000] base_lr: 8.4764e-05 lr: 3.1339e-07 eta: 1 day, 13:18:15 time: 0.9996 data_time: 0.0043 memory: 8462 loss: 0.0182 decode.loss_ce: 0.0090 decode.acc_seg: 99.5932 aux.loss_ce: 0.0091 aux.acc_seg: 98.7549 +04/17 15:40:44 - mmengine - INFO - Iter(train) [ 25700/160000] base_lr: 8.4732e-05 lr: 3.1327e-07 eta: 1 day, 13:17:25 time: 1.0003 data_time: 0.0045 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0090 decode.acc_seg: 99.7034 aux.loss_ce: 0.0087 aux.acc_seg: 99.1829 +04/17 15:41:34 - mmengine - INFO - Iter(train) [ 25750/160000] base_lr: 8.4701e-05 lr: 3.1316e-07 eta: 1 day, 13:16:35 time: 1.0003 data_time: 0.0053 memory: 8462 loss: 0.0195 decode.loss_ce: 0.0100 decode.acc_seg: 99.6637 aux.loss_ce: 0.0095 aux.acc_seg: 99.0105 +04/17 15:42:24 - mmengine - INFO - Iter(train) [ 25800/160000] base_lr: 8.4669e-05 lr: 3.1304e-07 eta: 1 day, 13:15:45 time: 1.0000 data_time: 0.0048 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0085 decode.acc_seg: 99.7545 aux.loss_ce: 0.0085 aux.acc_seg: 99.1781 +04/17 15:43:14 - mmengine - INFO - Iter(train) [ 25850/160000] base_lr: 8.4638e-05 lr: 3.1292e-07 eta: 1 day, 13:14:55 time: 1.0000 data_time: 0.0044 memory: 8462 loss: 0.0217 decode.loss_ce: 0.0107 decode.acc_seg: 99.2132 aux.loss_ce: 0.0110 aux.acc_seg: 97.7404 +04/17 15:44:04 - mmengine - INFO - Iter(train) [ 25900/160000] base_lr: 8.4606e-05 lr: 3.1281e-07 eta: 1 day, 13:14:05 time: 0.9998 data_time: 0.0043 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0100 decode.acc_seg: 99.7927 aux.loss_ce: 0.0090 aux.acc_seg: 99.4165 +04/17 15:44:54 - mmengine - INFO - Iter(train) [ 25950/160000] base_lr: 8.4575e-05 lr: 3.1269e-07 eta: 1 day, 13:13:15 time: 1.0007 data_time: 0.0047 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0091 decode.acc_seg: 99.4757 aux.loss_ce: 0.0088 aux.acc_seg: 98.9252 +04/17 15:45:44 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 15:45:44 - mmengine - INFO - Iter(train) [ 26000/160000] base_lr: 8.4543e-05 lr: 3.1257e-07 eta: 1 day, 13:12:25 time: 0.9984 data_time: 0.0046 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0089 decode.acc_seg: 99.5583 aux.loss_ce: 0.0088 aux.acc_seg: 99.0234 +04/17 15:46:34 - mmengine - INFO - Iter(train) [ 26050/160000] base_lr: 8.4512e-05 lr: 3.1246e-07 eta: 1 day, 13:11:35 time: 1.0002 data_time: 0.0043 memory: 8462 loss: 0.0199 decode.loss_ce: 0.0104 decode.acc_seg: 99.3654 aux.loss_ce: 0.0095 aux.acc_seg: 98.8035 +04/17 15:47:24 - mmengine - INFO - Iter(train) [ 26100/160000] base_lr: 8.4480e-05 lr: 3.1234e-07 eta: 1 day, 13:10:45 time: 1.0006 data_time: 0.0042 memory: 8462 loss: 0.0200 decode.loss_ce: 0.0095 decode.acc_seg: 99.3250 aux.loss_ce: 0.0105 aux.acc_seg: 98.2485 +04/17 15:48:14 - mmengine - INFO - Iter(train) [ 26150/160000] base_lr: 8.4448e-05 lr: 3.1222e-07 eta: 1 day, 13:09:55 time: 1.0006 data_time: 0.0045 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0093 decode.acc_seg: 99.6895 aux.loss_ce: 0.0086 aux.acc_seg: 99.3401 +04/17 15:49:04 - mmengine - INFO - Iter(train) [ 26200/160000] base_lr: 8.4417e-05 lr: 3.1211e-07 eta: 1 day, 13:09:05 time: 0.9983 data_time: 0.0046 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0078 decode.acc_seg: 99.5541 aux.loss_ce: 0.0083 aux.acc_seg: 98.9290 +04/17 15:49:54 - mmengine - INFO - Iter(train) [ 26250/160000] base_lr: 8.4385e-05 lr: 3.1199e-07 eta: 1 day, 13:08:15 time: 1.0002 data_time: 0.0046 memory: 8462 loss: 0.0171 decode.loss_ce: 0.0088 decode.acc_seg: 99.5800 aux.loss_ce: 0.0084 aux.acc_seg: 99.3288 +04/17 15:50:44 - mmengine - INFO - Iter(train) [ 26300/160000] base_lr: 8.4354e-05 lr: 3.1187e-07 eta: 1 day, 13:07:25 time: 0.9991 data_time: 0.0048 memory: 8462 loss: 0.0169 decode.loss_ce: 0.0084 decode.acc_seg: 99.5434 aux.loss_ce: 0.0086 aux.acc_seg: 99.2088 +04/17 15:51:34 - mmengine - INFO - Iter(train) [ 26350/160000] base_lr: 8.4322e-05 lr: 3.1176e-07 eta: 1 day, 13:06:35 time: 0.9997 data_time: 0.0044 memory: 8462 loss: 0.0215 decode.loss_ce: 0.0101 decode.acc_seg: 99.6191 aux.loss_ce: 0.0114 aux.acc_seg: 99.0993 +04/17 15:52:24 - mmengine - INFO - Iter(train) [ 26400/160000] base_lr: 8.4291e-05 lr: 3.1164e-07 eta: 1 day, 13:05:45 time: 0.9989 data_time: 0.0047 memory: 8462 loss: 0.0202 decode.loss_ce: 0.0098 decode.acc_seg: 99.5804 aux.loss_ce: 0.0103 aux.acc_seg: 98.5632 +04/17 15:53:14 - mmengine - INFO - Iter(train) [ 26450/160000] base_lr: 8.4259e-05 lr: 3.1152e-07 eta: 1 day, 13:04:55 time: 0.9988 data_time: 0.0045 memory: 8462 loss: 0.0187 decode.loss_ce: 0.0096 decode.acc_seg: 99.6672 aux.loss_ce: 0.0091 aux.acc_seg: 99.0217 +04/17 15:54:04 - mmengine - INFO - Iter(train) [ 26500/160000] base_lr: 8.4228e-05 lr: 3.1141e-07 eta: 1 day, 13:04:06 time: 1.0003 data_time: 0.0049 memory: 8462 loss: 0.0220 decode.loss_ce: 0.0103 decode.acc_seg: 99.5424 aux.loss_ce: 0.0117 aux.acc_seg: 98.9143 +04/17 15:54:54 - mmengine - INFO - Iter(train) [ 26550/160000] base_lr: 8.4196e-05 lr: 3.1129e-07 eta: 1 day, 13:03:16 time: 1.0008 data_time: 0.0050 memory: 8462 loss: 0.0196 decode.loss_ce: 0.0094 decode.acc_seg: 99.3586 aux.loss_ce: 0.0102 aux.acc_seg: 98.1108 +04/17 15:55:44 - mmengine - INFO - Iter(train) [ 26600/160000] base_lr: 8.4165e-05 lr: 3.1117e-07 eta: 1 day, 13:02:26 time: 1.0004 data_time: 0.0048 memory: 8462 loss: 0.0171 decode.loss_ce: 0.0088 decode.acc_seg: 99.6861 aux.loss_ce: 0.0083 aux.acc_seg: 99.3816 +04/17 15:56:34 - mmengine - INFO - Iter(train) [ 26650/160000] base_lr: 8.4133e-05 lr: 3.1106e-07 eta: 1 day, 13:01:36 time: 0.9998 data_time: 0.0044 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0097 decode.acc_seg: 99.6843 aux.loss_ce: 0.0091 aux.acc_seg: 98.8846 +04/17 15:57:24 - mmengine - INFO - Iter(train) [ 26700/160000] base_lr: 8.4101e-05 lr: 3.1094e-07 eta: 1 day, 13:00:46 time: 0.9990 data_time: 0.0043 memory: 8462 loss: 0.0193 decode.loss_ce: 0.0096 decode.acc_seg: 99.5705 aux.loss_ce: 0.0097 aux.acc_seg: 98.9902 +04/17 15:58:14 - mmengine - INFO - Iter(train) [ 26750/160000] base_lr: 8.4070e-05 lr: 3.1082e-07 eta: 1 day, 12:59:56 time: 1.0011 data_time: 0.0043 memory: 8462 loss: 0.0183 decode.loss_ce: 0.0092 decode.acc_seg: 99.7662 aux.loss_ce: 0.0090 aux.acc_seg: 99.4053 +04/17 15:59:04 - mmengine - INFO - Iter(train) [ 26800/160000] base_lr: 8.4038e-05 lr: 3.1071e-07 eta: 1 day, 12:59:06 time: 1.0011 data_time: 0.0044 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0091 decode.acc_seg: 99.5886 aux.loss_ce: 0.0089 aux.acc_seg: 99.1833 +04/17 15:59:54 - mmengine - INFO - Iter(train) [ 26850/160000] base_lr: 8.4007e-05 lr: 3.1059e-07 eta: 1 day, 12:58:16 time: 0.9997 data_time: 0.0043 memory: 8462 loss: 0.0202 decode.loss_ce: 0.0100 decode.acc_seg: 99.6778 aux.loss_ce: 0.0103 aux.acc_seg: 99.0946 +04/17 16:00:44 - mmengine - INFO - Iter(train) [ 26900/160000] base_lr: 8.3975e-05 lr: 3.1047e-07 eta: 1 day, 12:57:27 time: 1.0005 data_time: 0.0048 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0086 decode.acc_seg: 99.6935 aux.loss_ce: 0.0088 aux.acc_seg: 99.0488 +04/17 16:01:34 - mmengine - INFO - Iter(train) [ 26950/160000] base_lr: 8.3944e-05 lr: 3.1036e-07 eta: 1 day, 12:56:37 time: 1.0004 data_time: 0.0046 memory: 8462 loss: 0.0214 decode.loss_ce: 0.0113 decode.acc_seg: 99.6801 aux.loss_ce: 0.0101 aux.acc_seg: 98.8554 +04/17 16:02:24 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 16:02:24 - mmengine - INFO - Iter(train) [ 27000/160000] base_lr: 8.3912e-05 lr: 3.1024e-07 eta: 1 day, 12:55:47 time: 1.0002 data_time: 0.0042 memory: 8462 loss: 0.0183 decode.loss_ce: 0.0090 decode.acc_seg: 99.6613 aux.loss_ce: 0.0093 aux.acc_seg: 99.0273 +04/17 16:03:14 - mmengine - INFO - Iter(train) [ 27050/160000] base_lr: 8.3881e-05 lr: 3.1012e-07 eta: 1 day, 12:54:57 time: 0.9988 data_time: 0.0046 memory: 8462 loss: 0.0194 decode.loss_ce: 0.0095 decode.acc_seg: 99.5941 aux.loss_ce: 0.0098 aux.acc_seg: 98.9958 +04/17 16:04:04 - mmengine - INFO - Iter(train) [ 27100/160000] base_lr: 8.3849e-05 lr: 3.1001e-07 eta: 1 day, 12:54:07 time: 1.0011 data_time: 0.0045 memory: 8462 loss: 0.0169 decode.loss_ce: 0.0085 decode.acc_seg: 99.6609 aux.loss_ce: 0.0084 aux.acc_seg: 99.1459 +04/17 16:04:54 - mmengine - INFO - Iter(train) [ 27150/160000] base_lr: 8.3818e-05 lr: 3.0989e-07 eta: 1 day, 12:53:17 time: 0.9974 data_time: 0.0046 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0095 decode.acc_seg: 99.3881 aux.loss_ce: 0.0093 aux.acc_seg: 98.5313 +04/17 16:05:44 - mmengine - INFO - Iter(train) [ 27200/160000] base_lr: 8.3786e-05 lr: 3.0977e-07 eta: 1 day, 12:52:27 time: 0.9983 data_time: 0.0045 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0088 decode.acc_seg: 99.6912 aux.loss_ce: 0.0084 aux.acc_seg: 99.0955 +04/17 16:06:34 - mmengine - INFO - Iter(train) [ 27250/160000] base_lr: 8.3754e-05 lr: 3.0966e-07 eta: 1 day, 12:51:38 time: 0.9988 data_time: 0.0043 memory: 8462 loss: 0.0194 decode.loss_ce: 0.0101 decode.acc_seg: 99.5813 aux.loss_ce: 0.0093 aux.acc_seg: 98.8226 +04/17 16:07:24 - mmengine - INFO - Iter(train) [ 27300/160000] base_lr: 8.3723e-05 lr: 3.0954e-07 eta: 1 day, 12:50:48 time: 1.0001 data_time: 0.0047 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0098 decode.acc_seg: 99.5985 aux.loss_ce: 0.0092 aux.acc_seg: 98.8377 +04/17 16:08:14 - mmengine - INFO - Iter(train) [ 27350/160000] base_lr: 8.3691e-05 lr: 3.0942e-07 eta: 1 day, 12:49:58 time: 0.9998 data_time: 0.0051 memory: 8462 loss: 0.0199 decode.loss_ce: 0.0100 decode.acc_seg: 99.6555 aux.loss_ce: 0.0099 aux.acc_seg: 98.7318 +04/17 16:09:04 - mmengine - INFO - Iter(train) [ 27400/160000] base_lr: 8.3660e-05 lr: 3.0931e-07 eta: 1 day, 12:49:08 time: 0.9989 data_time: 0.0041 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0084 decode.acc_seg: 99.7126 aux.loss_ce: 0.0085 aux.acc_seg: 99.1102 +04/17 16:09:54 - mmengine - INFO - Iter(train) [ 27450/160000] base_lr: 8.3628e-05 lr: 3.0919e-07 eta: 1 day, 12:48:18 time: 1.0001 data_time: 0.0043 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0084 decode.acc_seg: 99.6378 aux.loss_ce: 0.0082 aux.acc_seg: 98.9553 +04/17 16:10:44 - mmengine - INFO - Iter(train) [ 27500/160000] base_lr: 8.3597e-05 lr: 3.0907e-07 eta: 1 day, 12:47:28 time: 0.9997 data_time: 0.0044 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0093 decode.acc_seg: 99.6639 aux.loss_ce: 0.0086 aux.acc_seg: 99.1831 +04/17 16:11:34 - mmengine - INFO - Iter(train) [ 27550/160000] base_lr: 8.3565e-05 lr: 3.0896e-07 eta: 1 day, 12:46:38 time: 0.9997 data_time: 0.0050 memory: 8462 loss: 0.0198 decode.loss_ce: 0.0103 decode.acc_seg: 99.5102 aux.loss_ce: 0.0095 aux.acc_seg: 98.8453 +04/17 16:12:24 - mmengine - INFO - Iter(train) [ 27600/160000] base_lr: 8.3534e-05 lr: 3.0884e-07 eta: 1 day, 12:45:48 time: 0.9996 data_time: 0.0045 memory: 8462 loss: 0.0200 decode.loss_ce: 0.0097 decode.acc_seg: 99.6437 aux.loss_ce: 0.0103 aux.acc_seg: 98.9790 +04/17 16:13:14 - mmengine - INFO - Iter(train) [ 27650/160000] base_lr: 8.3502e-05 lr: 3.0872e-07 eta: 1 day, 12:44:58 time: 0.9998 data_time: 0.0048 memory: 8462 loss: 0.0207 decode.loss_ce: 0.0109 decode.acc_seg: 99.1842 aux.loss_ce: 0.0098 aux.acc_seg: 98.6662 +04/17 16:14:04 - mmengine - INFO - Iter(train) [ 27700/160000] base_lr: 8.3471e-05 lr: 3.0861e-07 eta: 1 day, 12:44:09 time: 1.0016 data_time: 0.0046 memory: 8462 loss: 0.0185 decode.loss_ce: 0.0091 decode.acc_seg: 99.5279 aux.loss_ce: 0.0093 aux.acc_seg: 98.8819 +04/17 16:14:54 - mmengine - INFO - Iter(train) [ 27750/160000] base_lr: 8.3439e-05 lr: 3.0849e-07 eta: 1 day, 12:43:19 time: 1.0006 data_time: 0.0050 memory: 8462 loss: 0.0165 decode.loss_ce: 0.0081 decode.acc_seg: 99.5584 aux.loss_ce: 0.0084 aux.acc_seg: 99.1430 +04/17 16:15:44 - mmengine - INFO - Iter(train) [ 27800/160000] base_lr: 8.3407e-05 lr: 3.0837e-07 eta: 1 day, 12:42:29 time: 1.0014 data_time: 0.0045 memory: 8462 loss: 0.0201 decode.loss_ce: 0.0102 decode.acc_seg: 99.7395 aux.loss_ce: 0.0099 aux.acc_seg: 99.1167 +04/17 16:16:34 - mmengine - INFO - Iter(train) [ 27850/160000] base_lr: 8.3376e-05 lr: 3.0826e-07 eta: 1 day, 12:41:39 time: 1.0003 data_time: 0.0048 memory: 8462 loss: 0.0189 decode.loss_ce: 0.0093 decode.acc_seg: 99.6017 aux.loss_ce: 0.0096 aux.acc_seg: 98.9437 +04/17 16:17:24 - mmengine - INFO - Iter(train) [ 27900/160000] base_lr: 8.3344e-05 lr: 3.0814e-07 eta: 1 day, 12:40:49 time: 1.0001 data_time: 0.0044 memory: 8462 loss: 0.0180 decode.loss_ce: 0.0093 decode.acc_seg: 99.5377 aux.loss_ce: 0.0087 aux.acc_seg: 98.9826 +04/17 16:18:14 - mmengine - INFO - Iter(train) [ 27950/160000] base_lr: 8.3313e-05 lr: 3.0802e-07 eta: 1 day, 12:39:59 time: 1.0013 data_time: 0.0044 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0089 decode.acc_seg: 99.6466 aux.loss_ce: 0.0099 aux.acc_seg: 99.0154 +04/17 16:19:04 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 16:19:04 - mmengine - INFO - Iter(train) [ 28000/160000] base_lr: 8.3281e-05 lr: 3.0791e-07 eta: 1 day, 12:39:09 time: 0.9994 data_time: 0.0043 memory: 8462 loss: 0.0187 decode.loss_ce: 0.0094 decode.acc_seg: 99.5798 aux.loss_ce: 0.0093 aux.acc_seg: 98.9143 +04/17 16:19:54 - mmengine - INFO - Iter(train) [ 28050/160000] base_lr: 8.3250e-05 lr: 3.0779e-07 eta: 1 day, 12:38:19 time: 0.9998 data_time: 0.0042 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0086 decode.acc_seg: 99.4314 aux.loss_ce: 0.0087 aux.acc_seg: 98.5657 +04/17 16:20:44 - mmengine - INFO - Iter(train) [ 28100/160000] base_lr: 8.3218e-05 lr: 3.0767e-07 eta: 1 day, 12:37:29 time: 1.0006 data_time: 0.0045 memory: 8462 loss: 0.0164 decode.loss_ce: 0.0083 decode.acc_seg: 99.4890 aux.loss_ce: 0.0080 aux.acc_seg: 98.7194 +04/17 16:21:34 - mmengine - INFO - Iter(train) [ 28150/160000] base_lr: 8.3187e-05 lr: 3.0756e-07 eta: 1 day, 12:36:40 time: 1.0013 data_time: 0.0045 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0078 decode.acc_seg: 99.6967 aux.loss_ce: 0.0080 aux.acc_seg: 99.5455 +04/17 16:22:24 - mmengine - INFO - Iter(train) [ 28200/160000] base_lr: 8.3155e-05 lr: 3.0744e-07 eta: 1 day, 12:35:50 time: 0.9999 data_time: 0.0048 memory: 8462 loss: 0.0180 decode.loss_ce: 0.0088 decode.acc_seg: 99.4530 aux.loss_ce: 0.0092 aux.acc_seg: 98.8672 +04/17 16:23:14 - mmengine - INFO - Iter(train) [ 28250/160000] base_lr: 8.3124e-05 lr: 3.0732e-07 eta: 1 day, 12:35:00 time: 1.0004 data_time: 0.0048 memory: 8462 loss: 0.0213 decode.loss_ce: 0.0109 decode.acc_seg: 99.3597 aux.loss_ce: 0.0105 aux.acc_seg: 98.2992 +04/17 16:24:04 - mmengine - INFO - Iter(train) [ 28300/160000] base_lr: 8.3092e-05 lr: 3.0721e-07 eta: 1 day, 12:34:10 time: 0.9988 data_time: 0.0045 memory: 8462 loss: 0.0213 decode.loss_ce: 0.0109 decode.acc_seg: 99.7747 aux.loss_ce: 0.0104 aux.acc_seg: 99.4843 +04/17 16:24:54 - mmengine - INFO - Iter(train) [ 28350/160000] base_lr: 8.3060e-05 lr: 3.0709e-07 eta: 1 day, 12:33:20 time: 0.9990 data_time: 0.0047 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0096 decode.acc_seg: 99.6977 aux.loss_ce: 0.0092 aux.acc_seg: 98.8693 +04/17 16:25:44 - mmengine - INFO - Iter(train) [ 28400/160000] base_lr: 8.3029e-05 lr: 3.0697e-07 eta: 1 day, 12:32:30 time: 1.0008 data_time: 0.0045 memory: 8462 loss: 0.0182 decode.loss_ce: 0.0092 decode.acc_seg: 99.7694 aux.loss_ce: 0.0090 aux.acc_seg: 99.2901 +04/17 16:26:34 - mmengine - INFO - Iter(train) [ 28450/160000] base_lr: 8.2997e-05 lr: 3.0686e-07 eta: 1 day, 12:31:40 time: 1.0002 data_time: 0.0048 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0089 decode.acc_seg: 99.5073 aux.loss_ce: 0.0089 aux.acc_seg: 98.7480 +04/17 16:27:24 - mmengine - INFO - Iter(train) [ 28500/160000] base_lr: 8.2966e-05 lr: 3.0674e-07 eta: 1 day, 12:30:51 time: 1.0001 data_time: 0.0046 memory: 8462 loss: 0.0193 decode.loss_ce: 0.0098 decode.acc_seg: 99.4968 aux.loss_ce: 0.0095 aux.acc_seg: 99.0419 +04/17 16:28:14 - mmengine - INFO - Iter(train) [ 28550/160000] base_lr: 8.2934e-05 lr: 3.0663e-07 eta: 1 day, 12:30:01 time: 1.0004 data_time: 0.0045 memory: 8462 loss: 0.0191 decode.loss_ce: 0.0091 decode.acc_seg: 99.6040 aux.loss_ce: 0.0100 aux.acc_seg: 98.9010 +04/17 16:29:04 - mmengine - INFO - Iter(train) [ 28600/160000] base_lr: 8.2903e-05 lr: 3.0651e-07 eta: 1 day, 12:29:11 time: 1.0007 data_time: 0.0045 memory: 8462 loss: 0.0204 decode.loss_ce: 0.0100 decode.acc_seg: 99.6107 aux.loss_ce: 0.0104 aux.acc_seg: 98.8239 +04/17 16:29:54 - mmengine - INFO - Iter(train) [ 28650/160000] base_lr: 8.2871e-05 lr: 3.0639e-07 eta: 1 day, 12:28:21 time: 0.9990 data_time: 0.0045 memory: 8462 loss: 0.0163 decode.loss_ce: 0.0082 decode.acc_seg: 99.7017 aux.loss_ce: 0.0081 aux.acc_seg: 99.3147 +04/17 16:30:44 - mmengine - INFO - Iter(train) [ 28700/160000] base_lr: 8.2840e-05 lr: 3.0628e-07 eta: 1 day, 12:27:31 time: 1.0018 data_time: 0.0045 memory: 8462 loss: 0.0191 decode.loss_ce: 0.0098 decode.acc_seg: 99.6914 aux.loss_ce: 0.0093 aux.acc_seg: 99.1346 +04/17 16:31:34 - mmengine - INFO - Iter(train) [ 28750/160000] base_lr: 8.2808e-05 lr: 3.0616e-07 eta: 1 day, 12:26:41 time: 1.0001 data_time: 0.0045 memory: 8462 loss: 0.0186 decode.loss_ce: 0.0093 decode.acc_seg: 99.6403 aux.loss_ce: 0.0093 aux.acc_seg: 99.2172 +04/17 16:32:24 - mmengine - INFO - Iter(train) [ 28800/160000] base_lr: 8.2777e-05 lr: 3.0604e-07 eta: 1 day, 12:25:51 time: 1.0006 data_time: 0.0051 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0086 decode.acc_seg: 99.6933 aux.loss_ce: 0.0085 aux.acc_seg: 99.2430 +04/17 16:33:14 - mmengine - INFO - Iter(train) [ 28850/160000] base_lr: 8.2745e-05 lr: 3.0593e-07 eta: 1 day, 12:25:02 time: 1.0003 data_time: 0.0049 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0070 decode.acc_seg: 99.7858 aux.loss_ce: 0.0081 aux.acc_seg: 99.3725 +04/17 16:34:04 - mmengine - INFO - Iter(train) [ 28900/160000] base_lr: 8.2713e-05 lr: 3.0581e-07 eta: 1 day, 12:24:12 time: 0.9995 data_time: 0.0047 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0087 decode.acc_seg: 99.5235 aux.loss_ce: 0.0087 aux.acc_seg: 99.2687 +04/17 16:34:54 - mmengine - INFO - Iter(train) [ 28950/160000] base_lr: 8.2682e-05 lr: 3.0569e-07 eta: 1 day, 12:23:22 time: 1.0003 data_time: 0.0051 memory: 8462 loss: 0.0207 decode.loss_ce: 0.0106 decode.acc_seg: 99.6569 aux.loss_ce: 0.0102 aux.acc_seg: 99.1526 +04/17 16:35:44 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 16:35:44 - mmengine - INFO - Iter(train) [ 29000/160000] base_lr: 8.2650e-05 lr: 3.0558e-07 eta: 1 day, 12:22:32 time: 1.0016 data_time: 0.0044 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0084 decode.acc_seg: 99.5972 aux.loss_ce: 0.0085 aux.acc_seg: 99.0278 +04/17 16:36:34 - mmengine - INFO - Iter(train) [ 29050/160000] base_lr: 8.2619e-05 lr: 3.0546e-07 eta: 1 day, 12:21:42 time: 1.0009 data_time: 0.0047 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0086 decode.acc_seg: 99.5779 aux.loss_ce: 0.0092 aux.acc_seg: 99.0356 +04/17 16:37:24 - mmengine - INFO - Iter(train) [ 29100/160000] base_lr: 8.2587e-05 lr: 3.0534e-07 eta: 1 day, 12:20:52 time: 1.0013 data_time: 0.0050 memory: 8462 loss: 0.0185 decode.loss_ce: 0.0094 decode.acc_seg: 99.5415 aux.loss_ce: 0.0091 aux.acc_seg: 99.1438 +04/17 16:38:14 - mmengine - INFO - Iter(train) [ 29150/160000] base_lr: 8.2556e-05 lr: 3.0523e-07 eta: 1 day, 12:20:02 time: 1.0008 data_time: 0.0043 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0093 decode.acc_seg: 99.7322 aux.loss_ce: 0.0095 aux.acc_seg: 99.3235 +04/17 16:39:04 - mmengine - INFO - Iter(train) [ 29200/160000] base_lr: 8.2524e-05 lr: 3.0511e-07 eta: 1 day, 12:19:13 time: 1.0004 data_time: 0.0043 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0076 decode.acc_seg: 99.7124 aux.loss_ce: 0.0080 aux.acc_seg: 99.0248 +04/17 16:39:54 - mmengine - INFO - Iter(train) [ 29250/160000] base_lr: 8.2493e-05 lr: 3.0499e-07 eta: 1 day, 12:18:23 time: 1.0018 data_time: 0.0044 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0090 decode.acc_seg: 99.5857 aux.loss_ce: 0.0088 aux.acc_seg: 99.1066 +04/17 16:40:44 - mmengine - INFO - Iter(train) [ 29300/160000] base_lr: 8.2461e-05 lr: 3.0488e-07 eta: 1 day, 12:17:33 time: 1.0012 data_time: 0.0046 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0086 decode.acc_seg: 99.6841 aux.loss_ce: 0.0089 aux.acc_seg: 99.0004 +04/17 16:41:34 - mmengine - INFO - Iter(train) [ 29350/160000] base_lr: 8.2430e-05 lr: 3.0476e-07 eta: 1 day, 12:16:43 time: 1.0000 data_time: 0.0046 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0090 decode.acc_seg: 99.5378 aux.loss_ce: 0.0092 aux.acc_seg: 98.6506 +04/17 16:42:24 - mmengine - INFO - Iter(train) [ 29400/160000] base_lr: 8.2398e-05 lr: 3.0464e-07 eta: 1 day, 12:15:53 time: 1.0002 data_time: 0.0045 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0090 decode.acc_seg: 99.7160 aux.loss_ce: 0.0092 aux.acc_seg: 99.1240 +04/17 16:43:14 - mmengine - INFO - Iter(train) [ 29450/160000] base_lr: 8.2366e-05 lr: 3.0453e-07 eta: 1 day, 12:15:03 time: 0.9994 data_time: 0.0048 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0098 decode.acc_seg: 99.4558 aux.loss_ce: 0.0090 aux.acc_seg: 98.5292 +04/17 16:44:04 - mmengine - INFO - Iter(train) [ 29500/160000] base_lr: 8.2335e-05 lr: 3.0441e-07 eta: 1 day, 12:14:13 time: 0.9997 data_time: 0.0046 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0084 decode.acc_seg: 99.6563 aux.loss_ce: 0.0093 aux.acc_seg: 99.0377 +04/17 16:44:54 - mmengine - INFO - Iter(train) [ 29550/160000] base_lr: 8.2303e-05 lr: 3.0429e-07 eta: 1 day, 12:13:24 time: 1.0001 data_time: 0.0042 memory: 8462 loss: 0.0164 decode.loss_ce: 0.0082 decode.acc_seg: 99.6290 aux.loss_ce: 0.0083 aux.acc_seg: 98.9460 +04/17 16:45:44 - mmengine - INFO - Iter(train) [ 29600/160000] base_lr: 8.2272e-05 lr: 3.0418e-07 eta: 1 day, 12:12:34 time: 0.9990 data_time: 0.0045 memory: 8462 loss: 0.0198 decode.loss_ce: 0.0096 decode.acc_seg: 99.5298 aux.loss_ce: 0.0101 aux.acc_seg: 98.7080 +04/17 16:46:34 - mmengine - INFO - Iter(train) [ 29650/160000] base_lr: 8.2240e-05 lr: 3.0406e-07 eta: 1 day, 12:11:44 time: 1.0004 data_time: 0.0044 memory: 8462 loss: 0.0183 decode.loss_ce: 0.0092 decode.acc_seg: 99.6342 aux.loss_ce: 0.0092 aux.acc_seg: 98.8804 +04/17 16:47:24 - mmengine - INFO - Iter(train) [ 29700/160000] base_lr: 8.2209e-05 lr: 3.0394e-07 eta: 1 day, 12:10:54 time: 1.0008 data_time: 0.0046 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0090 decode.acc_seg: 99.6902 aux.loss_ce: 0.0082 aux.acc_seg: 99.1652 +04/17 16:48:14 - mmengine - INFO - Iter(train) [ 29750/160000] base_lr: 8.2177e-05 lr: 3.0383e-07 eta: 1 day, 12:10:04 time: 0.9999 data_time: 0.0045 memory: 8462 loss: 0.0171 decode.loss_ce: 0.0082 decode.acc_seg: 99.6937 aux.loss_ce: 0.0089 aux.acc_seg: 99.2622 +04/17 16:49:04 - mmengine - INFO - Iter(train) [ 29800/160000] base_lr: 8.2146e-05 lr: 3.0371e-07 eta: 1 day, 12:09:14 time: 0.9988 data_time: 0.0046 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0089 decode.acc_seg: 99.6332 aux.loss_ce: 0.0088 aux.acc_seg: 98.9525 +04/17 16:49:54 - mmengine - INFO - Iter(train) [ 29850/160000] base_lr: 8.2114e-05 lr: 3.0359e-07 eta: 1 day, 12:08:24 time: 1.0006 data_time: 0.0047 memory: 8462 loss: 0.0190 decode.loss_ce: 0.0094 decode.acc_seg: 99.7751 aux.loss_ce: 0.0096 aux.acc_seg: 99.3799 +04/17 16:50:44 - mmengine - INFO - Iter(train) [ 29900/160000] base_lr: 8.2083e-05 lr: 3.0348e-07 eta: 1 day, 12:07:34 time: 1.0004 data_time: 0.0047 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0077 decode.acc_seg: 99.7232 aux.loss_ce: 0.0084 aux.acc_seg: 99.4005 +04/17 16:51:34 - mmengine - INFO - Iter(train) [ 29950/160000] base_lr: 8.2051e-05 lr: 3.0336e-07 eta: 1 day, 12:06:45 time: 0.9991 data_time: 0.0042 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0084 decode.acc_seg: 99.7128 aux.loss_ce: 0.0083 aux.acc_seg: 99.5285 +04/17 16:52:24 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 16:52:24 - mmengine - INFO - Iter(train) [ 30000/160000] base_lr: 8.2019e-05 lr: 3.0324e-07 eta: 1 day, 12:05:55 time: 1.0000 data_time: 0.0048 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0087 decode.acc_seg: 99.6927 aux.loss_ce: 0.0086 aux.acc_seg: 99.1001 +04/17 16:52:24 - mmengine - INFO - Saving checkpoint at 30000 iterations +04/17 16:52:34 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:17 time: 0.1160 data_time: 0.0015 memory: 4004 +04/17 16:52:40 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:11 time: 0.1161 data_time: 0.0016 memory: 4004 +04/17 16:52:46 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:05 time: 0.1158 data_time: 0.0013 memory: 4004 +04/17 16:52:51 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.1155 data_time: 0.0013 memory: 4004 +04/17 16:52:52 - mmengine - INFO - per class results: +04/17 16:52:52 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 99.17 | 99.57 | 99.58 | 99.6 | 99.57 | +| contrast | 81.89 | 90.44 | 90.05 | 89.66 | 90.44 | ++------------+-------+-------+--------+-----------+--------+ +04/17 16:52:52 - mmengine - INFO - Iter(val) [200/200] aAcc: 99.2000 mIoU: 90.5300 mAcc: 95.0000 mFscore: 94.8200 mPrecision: 94.6300 mRecall: 95.0000 data_time: 0.0018 time: 0.1163 +04/17 16:53:42 - mmengine - INFO - Iter(train) [ 30050/160000] base_lr: 8.1988e-05 lr: 3.0313e-07 eta: 1 day, 12:05:05 time: 0.9991 data_time: 0.0046 memory: 8462 loss: 0.0184 decode.loss_ce: 0.0093 decode.acc_seg: 99.6229 aux.loss_ce: 0.0092 aux.acc_seg: 98.9380 +04/17 16:54:32 - mmengine - INFO - Iter(train) [ 30100/160000] base_lr: 8.1956e-05 lr: 3.0301e-07 eta: 1 day, 12:04:15 time: 0.9996 data_time: 0.0051 memory: 8462 loss: 0.0183 decode.loss_ce: 0.0094 decode.acc_seg: 99.5686 aux.loss_ce: 0.0089 aux.acc_seg: 98.9799 +04/17 16:55:22 - mmengine - INFO - Iter(train) [ 30150/160000] base_lr: 8.1925e-05 lr: 3.0289e-07 eta: 1 day, 12:03:25 time: 0.9991 data_time: 0.0043 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0087 decode.acc_seg: 99.5405 aux.loss_ce: 0.0086 aux.acc_seg: 98.7236 +04/17 16:56:12 - mmengine - INFO - Iter(train) [ 30200/160000] base_lr: 8.1893e-05 lr: 3.0278e-07 eta: 1 day, 12:02:35 time: 1.0004 data_time: 0.0050 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0088 decode.acc_seg: 99.7475 aux.loss_ce: 0.0087 aux.acc_seg: 99.2769 +04/17 16:57:02 - mmengine - INFO - Iter(train) [ 30250/160000] base_lr: 8.1862e-05 lr: 3.0266e-07 eta: 1 day, 12:01:46 time: 0.9994 data_time: 0.0045 memory: 8462 loss: 0.0183 decode.loss_ce: 0.0089 decode.acc_seg: 99.4925 aux.loss_ce: 0.0094 aux.acc_seg: 98.6217 +04/17 16:57:52 - mmengine - INFO - Iter(train) [ 30300/160000] base_lr: 8.1830e-05 lr: 3.0254e-07 eta: 1 day, 12:00:56 time: 0.9999 data_time: 0.0049 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0087 decode.acc_seg: 99.6794 aux.loss_ce: 0.0085 aux.acc_seg: 99.3116 +04/17 16:58:42 - mmengine - INFO - Iter(train) [ 30350/160000] base_lr: 8.1799e-05 lr: 3.0243e-07 eta: 1 day, 12:00:06 time: 1.0018 data_time: 0.0048 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0086 decode.acc_seg: 99.7377 aux.loss_ce: 0.0087 aux.acc_seg: 99.4810 +04/17 16:59:32 - mmengine - INFO - Iter(train) [ 30400/160000] base_lr: 8.1767e-05 lr: 3.0231e-07 eta: 1 day, 11:59:16 time: 0.9999 data_time: 0.0043 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0082 decode.acc_seg: 99.5684 aux.loss_ce: 0.0086 aux.acc_seg: 98.9161 +04/17 17:00:22 - mmengine - INFO - Iter(train) [ 30450/160000] base_lr: 8.1736e-05 lr: 3.0219e-07 eta: 1 day, 11:58:26 time: 1.0013 data_time: 0.0044 memory: 8462 loss: 0.0183 decode.loss_ce: 0.0092 decode.acc_seg: 99.7215 aux.loss_ce: 0.0092 aux.acc_seg: 99.3488 +04/17 17:01:12 - mmengine - INFO - Iter(train) [ 30500/160000] base_lr: 8.1704e-05 lr: 3.0208e-07 eta: 1 day, 11:57:36 time: 1.0001 data_time: 0.0043 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0091 decode.acc_seg: 99.6208 aux.loss_ce: 0.0088 aux.acc_seg: 98.4980 +04/17 17:02:02 - mmengine - INFO - Iter(train) [ 30550/160000] base_lr: 8.1672e-05 lr: 3.0196e-07 eta: 1 day, 11:56:46 time: 0.9996 data_time: 0.0046 memory: 8462 loss: 0.0146 decode.loss_ce: 0.0070 decode.acc_seg: 99.7467 aux.loss_ce: 0.0076 aux.acc_seg: 99.0271 +04/17 17:02:52 - mmengine - INFO - Iter(train) [ 30600/160000] base_lr: 8.1641e-05 lr: 3.0184e-07 eta: 1 day, 11:55:57 time: 0.9997 data_time: 0.0043 memory: 8462 loss: 0.0229 decode.loss_ce: 0.0121 decode.acc_seg: 99.5672 aux.loss_ce: 0.0108 aux.acc_seg: 98.9258 +04/17 17:03:42 - mmengine - INFO - Iter(train) [ 30650/160000] base_lr: 8.1609e-05 lr: 3.0173e-07 eta: 1 day, 11:55:07 time: 0.9990 data_time: 0.0044 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0086 decode.acc_seg: 99.5903 aux.loss_ce: 0.0092 aux.acc_seg: 98.6874 +04/17 17:04:32 - mmengine - INFO - Iter(train) [ 30700/160000] base_lr: 8.1578e-05 lr: 3.0161e-07 eta: 1 day, 11:54:17 time: 1.0001 data_time: 0.0046 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0085 decode.acc_seg: 99.6534 aux.loss_ce: 0.0083 aux.acc_seg: 99.3263 +04/17 17:05:22 - mmengine - INFO - Iter(train) [ 30750/160000] base_lr: 8.1546e-05 lr: 3.0149e-07 eta: 1 day, 11:53:27 time: 1.0014 data_time: 0.0044 memory: 8462 loss: 0.0162 decode.loss_ce: 0.0080 decode.acc_seg: 99.6407 aux.loss_ce: 0.0082 aux.acc_seg: 98.9790 +04/17 17:06:12 - mmengine - INFO - Iter(train) [ 30800/160000] base_lr: 8.1515e-05 lr: 3.0138e-07 eta: 1 day, 11:52:37 time: 0.9999 data_time: 0.0047 memory: 8462 loss: 0.0204 decode.loss_ce: 0.0099 decode.acc_seg: 99.6201 aux.loss_ce: 0.0105 aux.acc_seg: 99.0280 +04/17 17:07:02 - mmengine - INFO - Iter(train) [ 30850/160000] base_lr: 8.1483e-05 lr: 3.0126e-07 eta: 1 day, 11:51:47 time: 1.0010 data_time: 0.0048 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0082 decode.acc_seg: 99.6662 aux.loss_ce: 0.0084 aux.acc_seg: 99.2002 +04/17 17:07:52 - mmengine - INFO - Iter(train) [ 30900/160000] base_lr: 8.1452e-05 lr: 3.0114e-07 eta: 1 day, 11:50:57 time: 1.0002 data_time: 0.0044 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0089 decode.acc_seg: 99.5846 aux.loss_ce: 0.0087 aux.acc_seg: 99.1982 +04/17 17:08:42 - mmengine - INFO - Iter(train) [ 30950/160000] base_lr: 8.1420e-05 lr: 3.0103e-07 eta: 1 day, 11:50:07 time: 1.0004 data_time: 0.0046 memory: 8462 loss: 0.0163 decode.loss_ce: 0.0082 decode.acc_seg: 99.6296 aux.loss_ce: 0.0081 aux.acc_seg: 99.0252 +04/17 17:09:32 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 17:09:32 - mmengine - INFO - Iter(train) [ 31000/160000] base_lr: 8.1389e-05 lr: 3.0091e-07 eta: 1 day, 11:49:17 time: 0.9979 data_time: 0.0043 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0093 decode.acc_seg: 99.6519 aux.loss_ce: 0.0086 aux.acc_seg: 99.0602 +04/17 17:10:22 - mmengine - INFO - Iter(train) [ 31050/160000] base_lr: 8.1357e-05 lr: 3.0079e-07 eta: 1 day, 11:48:28 time: 1.0004 data_time: 0.0044 memory: 8462 loss: 0.0163 decode.loss_ce: 0.0080 decode.acc_seg: 99.6960 aux.loss_ce: 0.0083 aux.acc_seg: 99.1999 +04/17 17:11:12 - mmengine - INFO - Iter(train) [ 31100/160000] base_lr: 8.1325e-05 lr: 3.0068e-07 eta: 1 day, 11:47:38 time: 1.0006 data_time: 0.0047 memory: 8462 loss: 0.0196 decode.loss_ce: 0.0094 decode.acc_seg: 99.5914 aux.loss_ce: 0.0101 aux.acc_seg: 98.2475 +04/17 17:12:02 - mmengine - INFO - Iter(train) [ 31150/160000] base_lr: 8.1294e-05 lr: 3.0056e-07 eta: 1 day, 11:46:48 time: 1.0006 data_time: 0.0047 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0077 decode.acc_seg: 99.7179 aux.loss_ce: 0.0084 aux.acc_seg: 99.2775 +04/17 17:12:52 - mmengine - INFO - Iter(train) [ 31200/160000] base_lr: 8.1262e-05 lr: 3.0044e-07 eta: 1 day, 11:45:58 time: 1.0006 data_time: 0.0046 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0089 decode.acc_seg: 99.5459 aux.loss_ce: 0.0086 aux.acc_seg: 98.5708 +04/17 17:13:42 - mmengine - INFO - Iter(train) [ 31250/160000] base_lr: 8.1231e-05 lr: 3.0033e-07 eta: 1 day, 11:45:08 time: 1.0007 data_time: 0.0043 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0086 decode.acc_seg: 99.5615 aux.loss_ce: 0.0088 aux.acc_seg: 98.9775 +04/17 17:14:32 - mmengine - INFO - Iter(train) [ 31300/160000] base_lr: 8.1199e-05 lr: 3.0021e-07 eta: 1 day, 11:44:18 time: 0.9993 data_time: 0.0046 memory: 8462 loss: 0.0164 decode.loss_ce: 0.0081 decode.acc_seg: 99.6435 aux.loss_ce: 0.0083 aux.acc_seg: 99.1495 +04/17 17:15:22 - mmengine - INFO - Iter(train) [ 31350/160000] base_lr: 8.1168e-05 lr: 3.0009e-07 eta: 1 day, 11:43:28 time: 1.0012 data_time: 0.0050 memory: 8462 loss: 0.0205 decode.loss_ce: 0.0106 decode.acc_seg: 99.4677 aux.loss_ce: 0.0099 aux.acc_seg: 98.7720 +04/17 17:16:12 - mmengine - INFO - Iter(train) [ 31400/160000] base_lr: 8.1136e-05 lr: 2.9998e-07 eta: 1 day, 11:42:38 time: 1.0002 data_time: 0.0044 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0084 decode.acc_seg: 99.6706 aux.loss_ce: 0.0094 aux.acc_seg: 98.9902 +04/17 17:17:02 - mmengine - INFO - Iter(train) [ 31450/160000] base_lr: 8.1105e-05 lr: 2.9986e-07 eta: 1 day, 11:41:49 time: 1.0006 data_time: 0.0047 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0087 decode.acc_seg: 99.6580 aux.loss_ce: 0.0087 aux.acc_seg: 99.0391 +04/17 17:17:52 - mmengine - INFO - Iter(train) [ 31500/160000] base_lr: 8.1073e-05 lr: 2.9974e-07 eta: 1 day, 11:40:59 time: 1.0016 data_time: 0.0048 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0087 decode.acc_seg: 99.5750 aux.loss_ce: 0.0094 aux.acc_seg: 98.9014 +04/17 17:18:42 - mmengine - INFO - Iter(train) [ 31550/160000] base_lr: 8.1042e-05 lr: 2.9963e-07 eta: 1 day, 11:40:09 time: 0.9995 data_time: 0.0043 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0077 decode.acc_seg: 99.6256 aux.loss_ce: 0.0081 aux.acc_seg: 99.2733 +04/17 17:19:32 - mmengine - INFO - Iter(train) [ 31600/160000] base_lr: 8.1010e-05 lr: 2.9951e-07 eta: 1 day, 11:39:19 time: 1.0005 data_time: 0.0043 memory: 8462 loss: 0.0182 decode.loss_ce: 0.0088 decode.acc_seg: 99.6853 aux.loss_ce: 0.0094 aux.acc_seg: 99.0355 +04/17 17:20:22 - mmengine - INFO - Iter(train) [ 31650/160000] base_lr: 8.0978e-05 lr: 2.9939e-07 eta: 1 day, 11:38:29 time: 1.0009 data_time: 0.0042 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0089 decode.acc_seg: 99.7139 aux.loss_ce: 0.0087 aux.acc_seg: 99.0431 +04/17 17:21:12 - mmengine - INFO - Iter(train) [ 31700/160000] base_lr: 8.0947e-05 lr: 2.9928e-07 eta: 1 day, 11:37:39 time: 1.0006 data_time: 0.0045 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0087 decode.acc_seg: 99.5094 aux.loss_ce: 0.0090 aux.acc_seg: 98.4970 +04/17 17:22:02 - mmengine - INFO - Iter(train) [ 31750/160000] base_lr: 8.0915e-05 lr: 2.9916e-07 eta: 1 day, 11:36:50 time: 1.0004 data_time: 0.0043 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0093 decode.acc_seg: 99.6027 aux.loss_ce: 0.0088 aux.acc_seg: 99.1896 +04/17 17:22:52 - mmengine - INFO - Iter(train) [ 31800/160000] base_lr: 8.0884e-05 lr: 2.9904e-07 eta: 1 day, 11:36:00 time: 1.0007 data_time: 0.0044 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0083 decode.acc_seg: 99.6967 aux.loss_ce: 0.0095 aux.acc_seg: 99.1680 +04/17 17:23:42 - mmengine - INFO - Iter(train) [ 31850/160000] base_lr: 8.0852e-05 lr: 2.9893e-07 eta: 1 day, 11:35:10 time: 1.0009 data_time: 0.0046 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0094 decode.acc_seg: 99.6681 aux.loss_ce: 0.0093 aux.acc_seg: 99.1896 +04/17 17:24:32 - mmengine - INFO - Iter(train) [ 31900/160000] base_lr: 8.0821e-05 lr: 2.9881e-07 eta: 1 day, 11:34:20 time: 1.0004 data_time: 0.0045 memory: 8462 loss: 0.0165 decode.loss_ce: 0.0083 decode.acc_seg: 99.6717 aux.loss_ce: 0.0083 aux.acc_seg: 99.2178 +04/17 17:25:22 - mmengine - INFO - Iter(train) [ 31950/160000] base_lr: 8.0789e-05 lr: 2.9869e-07 eta: 1 day, 11:33:30 time: 1.0007 data_time: 0.0046 memory: 8462 loss: 0.0186 decode.loss_ce: 0.0093 decode.acc_seg: 99.6681 aux.loss_ce: 0.0093 aux.acc_seg: 99.0532 +04/17 17:26:12 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 17:26:12 - mmengine - INFO - Iter(train) [ 32000/160000] base_lr: 8.0758e-05 lr: 2.9858e-07 eta: 1 day, 11:32:40 time: 1.0000 data_time: 0.0049 memory: 8462 loss: 0.0195 decode.loss_ce: 0.0096 decode.acc_seg: 99.5377 aux.loss_ce: 0.0099 aux.acc_seg: 98.8344 +04/17 17:27:02 - mmengine - INFO - Iter(train) [ 32050/160000] base_lr: 8.0726e-05 lr: 2.9846e-07 eta: 1 day, 11:31:51 time: 1.0012 data_time: 0.0048 memory: 8462 loss: 0.0214 decode.loss_ce: 0.0103 decode.acc_seg: 99.5535 aux.loss_ce: 0.0112 aux.acc_seg: 98.8245 +04/17 17:27:52 - mmengine - INFO - Iter(train) [ 32100/160000] base_lr: 8.0695e-05 lr: 2.9834e-07 eta: 1 day, 11:31:01 time: 1.0007 data_time: 0.0048 memory: 8462 loss: 0.0187 decode.loss_ce: 0.0091 decode.acc_seg: 99.5520 aux.loss_ce: 0.0096 aux.acc_seg: 98.9317 +04/17 17:28:42 - mmengine - INFO - Iter(train) [ 32150/160000] base_lr: 8.0663e-05 lr: 2.9823e-07 eta: 1 day, 11:30:11 time: 0.9988 data_time: 0.0045 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0085 decode.acc_seg: 99.6653 aux.loss_ce: 0.0082 aux.acc_seg: 98.9891 +04/17 17:29:32 - mmengine - INFO - Iter(train) [ 32200/160000] base_lr: 8.0631e-05 lr: 2.9811e-07 eta: 1 day, 11:29:21 time: 1.0006 data_time: 0.0047 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0074 decode.acc_seg: 99.6592 aux.loss_ce: 0.0084 aux.acc_seg: 98.9397 +04/17 17:30:22 - mmengine - INFO - Iter(train) [ 32250/160000] base_lr: 8.0600e-05 lr: 2.9799e-07 eta: 1 day, 11:28:31 time: 1.0011 data_time: 0.0045 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0079 decode.acc_seg: 99.7044 aux.loss_ce: 0.0087 aux.acc_seg: 99.1220 +04/17 17:31:12 - mmengine - INFO - Iter(train) [ 32300/160000] base_lr: 8.0568e-05 lr: 2.9788e-07 eta: 1 day, 11:27:42 time: 0.9999 data_time: 0.0046 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0083 decode.acc_seg: 99.5981 aux.loss_ce: 0.0089 aux.acc_seg: 99.0562 +04/17 17:32:02 - mmengine - INFO - Iter(train) [ 32350/160000] base_lr: 8.0537e-05 lr: 2.9776e-07 eta: 1 day, 11:26:52 time: 0.9999 data_time: 0.0047 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0072 decode.acc_seg: 99.7723 aux.loss_ce: 0.0081 aux.acc_seg: 99.3151 +04/17 17:32:52 - mmengine - INFO - Iter(train) [ 32400/160000] base_lr: 8.0505e-05 lr: 2.9764e-07 eta: 1 day, 11:26:02 time: 1.0009 data_time: 0.0047 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0078 decode.acc_seg: 99.6283 aux.loss_ce: 0.0080 aux.acc_seg: 99.1644 +04/17 17:33:42 - mmengine - INFO - Iter(train) [ 32450/160000] base_lr: 8.0474e-05 lr: 2.9753e-07 eta: 1 day, 11:25:12 time: 0.9998 data_time: 0.0045 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0083 decode.acc_seg: 99.5544 aux.loss_ce: 0.0087 aux.acc_seg: 98.7215 +04/17 17:34:32 - mmengine - INFO - Iter(train) [ 32500/160000] base_lr: 8.0442e-05 lr: 2.9741e-07 eta: 1 day, 11:24:22 time: 1.0011 data_time: 0.0049 memory: 8462 loss: 0.0163 decode.loss_ce: 0.0082 decode.acc_seg: 99.6866 aux.loss_ce: 0.0082 aux.acc_seg: 99.3580 +04/17 17:35:22 - mmengine - INFO - Iter(train) [ 32550/160000] base_lr: 8.0411e-05 lr: 2.9729e-07 eta: 1 day, 11:23:32 time: 1.0015 data_time: 0.0049 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0083 decode.acc_seg: 99.6582 aux.loss_ce: 0.0091 aux.acc_seg: 99.2222 +04/17 17:36:12 - mmengine - INFO - Iter(train) [ 32600/160000] base_lr: 8.0379e-05 lr: 2.9718e-07 eta: 1 day, 11:22:43 time: 1.0007 data_time: 0.0048 memory: 8462 loss: 0.0197 decode.loss_ce: 0.0102 decode.acc_seg: 99.6645 aux.loss_ce: 0.0095 aux.acc_seg: 98.9851 +04/17 17:37:03 - mmengine - INFO - Iter(train) [ 32650/160000] base_lr: 8.0348e-05 lr: 2.9706e-07 eta: 1 day, 11:21:53 time: 1.0004 data_time: 0.0047 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0081 decode.acc_seg: 99.8064 aux.loss_ce: 0.0086 aux.acc_seg: 99.3704 +04/17 17:37:52 - mmengine - INFO - Iter(train) [ 32700/160000] base_lr: 8.0316e-05 lr: 2.9694e-07 eta: 1 day, 11:21:03 time: 1.0001 data_time: 0.0044 memory: 8462 loss: 0.0182 decode.loss_ce: 0.0092 decode.acc_seg: 99.6901 aux.loss_ce: 0.0090 aux.acc_seg: 99.2725 +04/17 17:38:43 - mmengine - INFO - Iter(train) [ 32750/160000] base_lr: 8.0284e-05 lr: 2.9683e-07 eta: 1 day, 11:20:13 time: 1.0008 data_time: 0.0051 memory: 8462 loss: 0.0151 decode.loss_ce: 0.0077 decode.acc_seg: 99.5844 aux.loss_ce: 0.0074 aux.acc_seg: 98.8190 +04/17 17:39:33 - mmengine - INFO - Iter(train) [ 32800/160000] base_lr: 8.0253e-05 lr: 2.9671e-07 eta: 1 day, 11:19:23 time: 1.0012 data_time: 0.0043 memory: 8462 loss: 0.0198 decode.loss_ce: 0.0101 decode.acc_seg: 99.6080 aux.loss_ce: 0.0098 aux.acc_seg: 99.1676 +04/17 17:40:23 - mmengine - INFO - Iter(train) [ 32850/160000] base_lr: 8.0221e-05 lr: 2.9659e-07 eta: 1 day, 11:18:34 time: 0.9995 data_time: 0.0046 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0079 decode.acc_seg: 99.6521 aux.loss_ce: 0.0082 aux.acc_seg: 98.9561 +04/17 17:41:13 - mmengine - INFO - Iter(train) [ 32900/160000] base_lr: 8.0190e-05 lr: 2.9648e-07 eta: 1 day, 11:17:44 time: 1.0006 data_time: 0.0048 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0077 decode.acc_seg: 99.7070 aux.loss_ce: 0.0081 aux.acc_seg: 99.1634 +04/17 17:42:03 - mmengine - INFO - Iter(train) [ 32950/160000] base_lr: 8.0158e-05 lr: 2.9636e-07 eta: 1 day, 11:16:54 time: 1.0020 data_time: 0.0045 memory: 8462 loss: 0.0136 decode.loss_ce: 0.0064 decode.acc_seg: 99.7488 aux.loss_ce: 0.0072 aux.acc_seg: 99.3393 +04/17 17:42:53 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 17:42:53 - mmengine - INFO - Iter(train) [ 33000/160000] base_lr: 8.0127e-05 lr: 2.9624e-07 eta: 1 day, 11:16:04 time: 0.9996 data_time: 0.0045 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0079 decode.acc_seg: 99.7297 aux.loss_ce: 0.0088 aux.acc_seg: 99.3464 +04/17 17:43:43 - mmengine - INFO - Iter(train) [ 33050/160000] base_lr: 8.0095e-05 lr: 2.9613e-07 eta: 1 day, 11:15:14 time: 1.0004 data_time: 0.0044 memory: 8462 loss: 0.0186 decode.loss_ce: 0.0090 decode.acc_seg: 99.6449 aux.loss_ce: 0.0096 aux.acc_seg: 99.0391 +04/17 17:44:33 - mmengine - INFO - Iter(train) [ 33100/160000] base_lr: 8.0064e-05 lr: 2.9601e-07 eta: 1 day, 11:14:25 time: 1.0002 data_time: 0.0044 memory: 8462 loss: 0.0189 decode.loss_ce: 0.0091 decode.acc_seg: 99.4652 aux.loss_ce: 0.0098 aux.acc_seg: 98.3170 +04/17 17:45:23 - mmengine - INFO - Iter(train) [ 33150/160000] base_lr: 8.0032e-05 lr: 2.9589e-07 eta: 1 day, 11:13:35 time: 1.0021 data_time: 0.0044 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0073 decode.acc_seg: 99.7093 aux.loss_ce: 0.0081 aux.acc_seg: 99.1882 +04/17 17:46:13 - mmengine - INFO - Iter(train) [ 33200/160000] base_lr: 8.0001e-05 lr: 2.9578e-07 eta: 1 day, 11:12:45 time: 1.0014 data_time: 0.0044 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0077 decode.acc_seg: 99.7759 aux.loss_ce: 0.0083 aux.acc_seg: 99.1198 +04/17 17:47:03 - mmengine - INFO - Iter(train) [ 33250/160000] base_lr: 7.9969e-05 lr: 2.9566e-07 eta: 1 day, 11:11:55 time: 1.0003 data_time: 0.0047 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0079 decode.acc_seg: 99.7992 aux.loss_ce: 0.0073 aux.acc_seg: 99.5865 +04/17 17:47:53 - mmengine - INFO - Iter(train) [ 33300/160000] base_lr: 7.9937e-05 lr: 2.9555e-07 eta: 1 day, 11:11:06 time: 1.0017 data_time: 0.0054 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0086 decode.acc_seg: 99.6231 aux.loss_ce: 0.0089 aux.acc_seg: 99.0973 +04/17 17:48:43 - mmengine - INFO - Iter(train) [ 33350/160000] base_lr: 7.9906e-05 lr: 2.9543e-07 eta: 1 day, 11:10:16 time: 1.0011 data_time: 0.0044 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0073 decode.acc_seg: 99.5970 aux.loss_ce: 0.0083 aux.acc_seg: 99.0637 +04/17 17:49:33 - mmengine - INFO - Iter(train) [ 33400/160000] base_lr: 7.9874e-05 lr: 2.9531e-07 eta: 1 day, 11:09:26 time: 1.0000 data_time: 0.0045 memory: 8462 loss: 0.0171 decode.loss_ce: 0.0086 decode.acc_seg: 99.6271 aux.loss_ce: 0.0085 aux.acc_seg: 98.9088 +04/17 17:50:23 - mmengine - INFO - Iter(train) [ 33450/160000] base_lr: 7.9843e-05 lr: 2.9520e-07 eta: 1 day, 11:08:36 time: 1.0015 data_time: 0.0043 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0083 decode.acc_seg: 99.6145 aux.loss_ce: 0.0085 aux.acc_seg: 99.1199 +04/17 17:51:13 - mmengine - INFO - Iter(train) [ 33500/160000] base_lr: 7.9811e-05 lr: 2.9508e-07 eta: 1 day, 11:07:46 time: 1.0003 data_time: 0.0050 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0082 decode.acc_seg: 99.7141 aux.loss_ce: 0.0098 aux.acc_seg: 98.9956 +04/17 17:52:03 - mmengine - INFO - Iter(train) [ 33550/160000] base_lr: 7.9780e-05 lr: 2.9496e-07 eta: 1 day, 11:06:56 time: 1.0000 data_time: 0.0047 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0077 decode.acc_seg: 99.5331 aux.loss_ce: 0.0078 aux.acc_seg: 99.1173 +04/17 17:52:53 - mmengine - INFO - Iter(train) [ 33600/160000] base_lr: 7.9748e-05 lr: 2.9485e-07 eta: 1 day, 11:06:07 time: 1.0011 data_time: 0.0046 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0089 decode.acc_seg: 99.6397 aux.loss_ce: 0.0087 aux.acc_seg: 99.3166 +04/17 17:53:43 - mmengine - INFO - Iter(train) [ 33650/160000] base_lr: 7.9717e-05 lr: 2.9473e-07 eta: 1 day, 11:05:17 time: 1.0014 data_time: 0.0045 memory: 8462 loss: 0.0184 decode.loss_ce: 0.0092 decode.acc_seg: 99.5447 aux.loss_ce: 0.0092 aux.acc_seg: 98.8676 +04/17 17:54:33 - mmengine - INFO - Iter(train) [ 33700/160000] base_lr: 7.9685e-05 lr: 2.9461e-07 eta: 1 day, 11:04:27 time: 1.0010 data_time: 0.0041 memory: 8462 loss: 0.0164 decode.loss_ce: 0.0079 decode.acc_seg: 99.6780 aux.loss_ce: 0.0085 aux.acc_seg: 98.8388 +04/17 17:55:23 - mmengine - INFO - Iter(train) [ 33750/160000] base_lr: 7.9653e-05 lr: 2.9450e-07 eta: 1 day, 11:03:37 time: 1.0008 data_time: 0.0049 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0085 decode.acc_seg: 99.7696 aux.loss_ce: 0.0092 aux.acc_seg: 99.3156 +04/17 17:56:13 - mmengine - INFO - Iter(train) [ 33800/160000] base_lr: 7.9622e-05 lr: 2.9438e-07 eta: 1 day, 11:02:48 time: 1.0008 data_time: 0.0048 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0086 decode.acc_seg: 99.5960 aux.loss_ce: 0.0091 aux.acc_seg: 98.7976 +04/17 17:57:03 - mmengine - INFO - Iter(train) [ 33850/160000] base_lr: 7.9590e-05 lr: 2.9426e-07 eta: 1 day, 11:01:58 time: 1.0015 data_time: 0.0043 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0076 decode.acc_seg: 99.6857 aux.loss_ce: 0.0082 aux.acc_seg: 99.0356 +04/17 17:57:54 - mmengine - INFO - Iter(train) [ 33900/160000] base_lr: 7.9559e-05 lr: 2.9415e-07 eta: 1 day, 11:01:08 time: 1.0011 data_time: 0.0050 memory: 8462 loss: 0.0191 decode.loss_ce: 0.0098 decode.acc_seg: 99.7295 aux.loss_ce: 0.0093 aux.acc_seg: 99.0496 +04/17 17:58:44 - mmengine - INFO - Iter(train) [ 33950/160000] base_lr: 7.9527e-05 lr: 2.9403e-07 eta: 1 day, 11:00:18 time: 1.0003 data_time: 0.0041 memory: 8462 loss: 0.0185 decode.loss_ce: 0.0091 decode.acc_seg: 99.5611 aux.loss_ce: 0.0093 aux.acc_seg: 98.7150 +04/17 17:59:34 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 17:59:34 - mmengine - INFO - Iter(train) [ 34000/160000] base_lr: 7.9496e-05 lr: 2.9391e-07 eta: 1 day, 10:59:28 time: 1.0011 data_time: 0.0044 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0079 decode.acc_seg: 99.5853 aux.loss_ce: 0.0082 aux.acc_seg: 98.8203 +04/17 18:00:24 - mmengine - INFO - Iter(train) [ 34050/160000] base_lr: 7.9464e-05 lr: 2.9380e-07 eta: 1 day, 10:58:38 time: 1.0006 data_time: 0.0044 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0084 decode.acc_seg: 99.6889 aux.loss_ce: 0.0095 aux.acc_seg: 99.3422 +04/17 18:01:14 - mmengine - INFO - Iter(train) [ 34100/160000] base_lr: 7.9433e-05 lr: 2.9368e-07 eta: 1 day, 10:57:49 time: 1.0003 data_time: 0.0047 memory: 8462 loss: 0.0163 decode.loss_ce: 0.0077 decode.acc_seg: 99.5663 aux.loss_ce: 0.0086 aux.acc_seg: 98.7095 +04/17 18:02:04 - mmengine - INFO - Iter(train) [ 34150/160000] base_lr: 7.9401e-05 lr: 2.9356e-07 eta: 1 day, 10:56:59 time: 1.0005 data_time: 0.0044 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0078 decode.acc_seg: 99.5920 aux.loss_ce: 0.0082 aux.acc_seg: 99.0913 +04/17 18:02:54 - mmengine - INFO - Iter(train) [ 34200/160000] base_lr: 7.9370e-05 lr: 2.9345e-07 eta: 1 day, 10:56:09 time: 0.9992 data_time: 0.0045 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0077 decode.acc_seg: 99.5230 aux.loss_ce: 0.0081 aux.acc_seg: 98.9002 +04/17 18:03:44 - mmengine - INFO - Iter(train) [ 34250/160000] base_lr: 7.9338e-05 lr: 2.9333e-07 eta: 1 day, 10:55:19 time: 1.0003 data_time: 0.0044 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0078 decode.acc_seg: 99.7005 aux.loss_ce: 0.0083 aux.acc_seg: 99.4404 +04/17 18:04:34 - mmengine - INFO - Iter(train) [ 34300/160000] base_lr: 7.9306e-05 lr: 2.9321e-07 eta: 1 day, 10:54:29 time: 1.0001 data_time: 0.0046 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0073 decode.acc_seg: 99.7816 aux.loss_ce: 0.0077 aux.acc_seg: 99.3183 +04/17 18:05:24 - mmengine - INFO - Iter(train) [ 34350/160000] base_lr: 7.9275e-05 lr: 2.9310e-07 eta: 1 day, 10:53:39 time: 0.9992 data_time: 0.0046 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0084 decode.acc_seg: 99.7252 aux.loss_ce: 0.0095 aux.acc_seg: 99.2121 +04/17 18:06:14 - mmengine - INFO - Iter(train) [ 34400/160000] base_lr: 7.9243e-05 lr: 2.9298e-07 eta: 1 day, 10:52:50 time: 1.0001 data_time: 0.0047 memory: 8462 loss: 0.0191 decode.loss_ce: 0.0094 decode.acc_seg: 99.7040 aux.loss_ce: 0.0097 aux.acc_seg: 99.1589 +04/17 18:07:04 - mmengine - INFO - Iter(train) [ 34450/160000] base_lr: 7.9212e-05 lr: 2.9286e-07 eta: 1 day, 10:52:00 time: 1.0013 data_time: 0.0046 memory: 8462 loss: 0.0193 decode.loss_ce: 0.0094 decode.acc_seg: 99.3542 aux.loss_ce: 0.0099 aux.acc_seg: 98.7295 +04/17 18:07:54 - mmengine - INFO - Iter(train) [ 34500/160000] base_lr: 7.9180e-05 lr: 2.9275e-07 eta: 1 day, 10:51:10 time: 0.9988 data_time: 0.0043 memory: 8462 loss: 0.0180 decode.loss_ce: 0.0089 decode.acc_seg: 99.7618 aux.loss_ce: 0.0091 aux.acc_seg: 99.5354 +04/17 18:08:44 - mmengine - INFO - Iter(train) [ 34550/160000] base_lr: 7.9149e-05 lr: 2.9263e-07 eta: 1 day, 10:50:20 time: 1.0006 data_time: 0.0043 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0081 decode.acc_seg: 99.4892 aux.loss_ce: 0.0091 aux.acc_seg: 98.6187 +04/17 18:09:34 - mmengine - INFO - Iter(train) [ 34600/160000] base_lr: 7.9117e-05 lr: 2.9251e-07 eta: 1 day, 10:49:30 time: 1.0007 data_time: 0.0047 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0081 decode.acc_seg: 99.6304 aux.loss_ce: 0.0079 aux.acc_seg: 98.9979 +04/17 18:10:24 - mmengine - INFO - Iter(train) [ 34650/160000] base_lr: 7.9086e-05 lr: 2.9240e-07 eta: 1 day, 10:48:41 time: 1.0000 data_time: 0.0047 memory: 8462 loss: 0.0186 decode.loss_ce: 0.0088 decode.acc_seg: 99.6246 aux.loss_ce: 0.0098 aux.acc_seg: 98.9098 +04/17 18:11:14 - mmengine - INFO - Iter(train) [ 34700/160000] base_lr: 7.9054e-05 lr: 2.9228e-07 eta: 1 day, 10:47:51 time: 1.0005 data_time: 0.0050 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0073 decode.acc_seg: 99.7663 aux.loss_ce: 0.0085 aux.acc_seg: 99.4064 +04/17 18:12:04 - mmengine - INFO - Iter(train) [ 34750/160000] base_lr: 7.9023e-05 lr: 2.9216e-07 eta: 1 day, 10:47:01 time: 1.0008 data_time: 0.0048 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0084 decode.acc_seg: 99.7515 aux.loss_ce: 0.0092 aux.acc_seg: 99.3605 +04/17 18:12:54 - mmengine - INFO - Iter(train) [ 34800/160000] base_lr: 7.8991e-05 lr: 2.9205e-07 eta: 1 day, 10:46:11 time: 1.0012 data_time: 0.0046 memory: 8462 loss: 0.0191 decode.loss_ce: 0.0094 decode.acc_seg: 99.4804 aux.loss_ce: 0.0097 aux.acc_seg: 99.0023 +04/17 18:13:44 - mmengine - INFO - Iter(train) [ 34850/160000] base_lr: 7.8959e-05 lr: 2.9193e-07 eta: 1 day, 10:45:22 time: 1.0021 data_time: 0.0047 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0083 decode.acc_seg: 99.5070 aux.loss_ce: 0.0083 aux.acc_seg: 98.9429 +04/17 18:14:34 - mmengine - INFO - Iter(train) [ 34900/160000] base_lr: 7.8928e-05 lr: 2.9181e-07 eta: 1 day, 10:44:32 time: 1.0006 data_time: 0.0049 memory: 8462 loss: 0.0144 decode.loss_ce: 0.0069 decode.acc_seg: 99.7284 aux.loss_ce: 0.0076 aux.acc_seg: 99.1909 +04/17 18:15:24 - mmengine - INFO - Iter(train) [ 34950/160000] base_lr: 7.8896e-05 lr: 2.9170e-07 eta: 1 day, 10:43:42 time: 1.0003 data_time: 0.0044 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0081 decode.acc_seg: 99.5810 aux.loss_ce: 0.0089 aux.acc_seg: 99.0734 +04/17 18:16:14 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 18:16:14 - mmengine - INFO - Iter(train) [ 35000/160000] base_lr: 7.8865e-05 lr: 2.9158e-07 eta: 1 day, 10:42:52 time: 1.0006 data_time: 0.0050 memory: 8462 loss: 0.0159 decode.loss_ce: 0.0074 decode.acc_seg: 99.7099 aux.loss_ce: 0.0085 aux.acc_seg: 99.0732 +04/17 18:17:05 - mmengine - INFO - Iter(train) [ 35050/160000] base_lr: 7.8833e-05 lr: 2.9146e-07 eta: 1 day, 10:42:03 time: 1.0005 data_time: 0.0044 memory: 8462 loss: 0.0179 decode.loss_ce: 0.0087 decode.acc_seg: 99.5447 aux.loss_ce: 0.0092 aux.acc_seg: 99.0398 +04/17 18:17:55 - mmengine - INFO - Iter(train) [ 35100/160000] base_lr: 7.8802e-05 lr: 2.9135e-07 eta: 1 day, 10:41:13 time: 1.0006 data_time: 0.0048 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0090 decode.acc_seg: 99.4404 aux.loss_ce: 0.0085 aux.acc_seg: 98.6664 +04/17 18:18:45 - mmengine - INFO - Iter(train) [ 35150/160000] base_lr: 7.8770e-05 lr: 2.9123e-07 eta: 1 day, 10:40:23 time: 1.0017 data_time: 0.0050 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0074 decode.acc_seg: 99.7715 aux.loss_ce: 0.0073 aux.acc_seg: 99.4728 +04/17 18:19:35 - mmengine - INFO - Iter(train) [ 35200/160000] base_lr: 7.8739e-05 lr: 2.9111e-07 eta: 1 day, 10:39:33 time: 1.0000 data_time: 0.0045 memory: 8462 loss: 0.0183 decode.loss_ce: 0.0093 decode.acc_seg: 99.7454 aux.loss_ce: 0.0090 aux.acc_seg: 99.4465 +04/17 18:20:25 - mmengine - INFO - Iter(train) [ 35250/160000] base_lr: 7.8707e-05 lr: 2.9100e-07 eta: 1 day, 10:38:43 time: 1.0004 data_time: 0.0045 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0080 decode.acc_seg: 99.6777 aux.loss_ce: 0.0087 aux.acc_seg: 99.0917 +04/17 18:21:15 - mmengine - INFO - Iter(train) [ 35300/160000] base_lr: 7.8676e-05 lr: 2.9088e-07 eta: 1 day, 10:37:54 time: 1.0014 data_time: 0.0042 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0085 decode.acc_seg: 99.7107 aux.loss_ce: 0.0088 aux.acc_seg: 99.1545 +04/17 18:22:05 - mmengine - INFO - Iter(train) [ 35350/160000] base_lr: 7.8644e-05 lr: 2.9076e-07 eta: 1 day, 10:37:04 time: 0.9998 data_time: 0.0045 memory: 8462 loss: 0.0162 decode.loss_ce: 0.0077 decode.acc_seg: 99.6267 aux.loss_ce: 0.0085 aux.acc_seg: 98.8953 +04/17 18:22:55 - mmengine - INFO - Iter(train) [ 35400/160000] base_lr: 7.8612e-05 lr: 2.9065e-07 eta: 1 day, 10:36:14 time: 0.9996 data_time: 0.0042 memory: 8462 loss: 0.0145 decode.loss_ce: 0.0068 decode.acc_seg: 99.6824 aux.loss_ce: 0.0077 aux.acc_seg: 99.0627 +04/17 18:23:45 - mmengine - INFO - Iter(train) [ 35450/160000] base_lr: 7.8581e-05 lr: 2.9053e-07 eta: 1 day, 10:35:24 time: 1.0002 data_time: 0.0044 memory: 8462 loss: 0.0169 decode.loss_ce: 0.0082 decode.acc_seg: 99.5533 aux.loss_ce: 0.0087 aux.acc_seg: 98.6879 +04/17 18:24:35 - mmengine - INFO - Iter(train) [ 35500/160000] base_lr: 7.8549e-05 lr: 2.9041e-07 eta: 1 day, 10:34:34 time: 1.0011 data_time: 0.0046 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0070 decode.acc_seg: 99.7019 aux.loss_ce: 0.0079 aux.acc_seg: 99.1934 +04/17 18:25:25 - mmengine - INFO - Iter(train) [ 35550/160000] base_lr: 7.8518e-05 lr: 2.9030e-07 eta: 1 day, 10:33:44 time: 1.0006 data_time: 0.0046 memory: 8462 loss: 0.0183 decode.loss_ce: 0.0092 decode.acc_seg: 99.6517 aux.loss_ce: 0.0091 aux.acc_seg: 99.1537 +04/17 18:26:15 - mmengine - INFO - Iter(train) [ 35600/160000] base_lr: 7.8486e-05 lr: 2.9018e-07 eta: 1 day, 10:32:55 time: 1.0007 data_time: 0.0047 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0083 decode.acc_seg: 99.6302 aux.loss_ce: 0.0092 aux.acc_seg: 99.0995 +04/17 18:27:05 - mmengine - INFO - Iter(train) [ 35650/160000] base_lr: 7.8455e-05 lr: 2.9006e-07 eta: 1 day, 10:32:05 time: 0.9998 data_time: 0.0045 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0086 decode.acc_seg: 99.5853 aux.loss_ce: 0.0087 aux.acc_seg: 99.2022 +04/17 18:27:55 - mmengine - INFO - Iter(train) [ 35700/160000] base_lr: 7.8423e-05 lr: 2.8995e-07 eta: 1 day, 10:31:15 time: 0.9993 data_time: 0.0046 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0081 decode.acc_seg: 99.5628 aux.loss_ce: 0.0090 aux.acc_seg: 98.5685 +04/17 18:28:45 - mmengine - INFO - Iter(train) [ 35750/160000] base_lr: 7.8392e-05 lr: 2.8983e-07 eta: 1 day, 10:30:25 time: 1.0009 data_time: 0.0052 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0075 decode.acc_seg: 99.7187 aux.loss_ce: 0.0080 aux.acc_seg: 99.4799 +04/17 18:29:35 - mmengine - INFO - Iter(train) [ 35800/160000] base_lr: 7.8360e-05 lr: 2.8971e-07 eta: 1 day, 10:29:35 time: 1.0014 data_time: 0.0046 memory: 8462 loss: 0.0163 decode.loss_ce: 0.0076 decode.acc_seg: 99.6855 aux.loss_ce: 0.0087 aux.acc_seg: 99.1003 +04/17 18:30:25 - mmengine - INFO - Iter(train) [ 35850/160000] base_lr: 7.8329e-05 lr: 2.8960e-07 eta: 1 day, 10:28:45 time: 1.0012 data_time: 0.0047 memory: 8462 loss: 0.0136 decode.loss_ce: 0.0067 decode.acc_seg: 99.7375 aux.loss_ce: 0.0069 aux.acc_seg: 99.3561 +04/17 18:31:15 - mmengine - INFO - Iter(train) [ 35900/160000] base_lr: 7.8297e-05 lr: 2.8948e-07 eta: 1 day, 10:27:56 time: 1.0005 data_time: 0.0046 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0078 decode.acc_seg: 99.5060 aux.loss_ce: 0.0088 aux.acc_seg: 99.0116 +04/17 18:32:05 - mmengine - INFO - Iter(train) [ 35950/160000] base_lr: 7.8265e-05 lr: 2.8936e-07 eta: 1 day, 10:27:06 time: 1.0015 data_time: 0.0043 memory: 8462 loss: 0.0159 decode.loss_ce: 0.0077 decode.acc_seg: 99.7074 aux.loss_ce: 0.0082 aux.acc_seg: 99.2037 +04/17 18:32:55 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 18:32:55 - mmengine - INFO - Iter(train) [ 36000/160000] base_lr: 7.8234e-05 lr: 2.8925e-07 eta: 1 day, 10:26:16 time: 1.0010 data_time: 0.0049 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0086 decode.acc_seg: 99.5695 aux.loss_ce: 0.0088 aux.acc_seg: 98.9298 +04/17 18:33:45 - mmengine - INFO - Iter(train) [ 36050/160000] base_lr: 7.8202e-05 lr: 2.8913e-07 eta: 1 day, 10:25:26 time: 1.0012 data_time: 0.0044 memory: 8462 loss: 0.0151 decode.loss_ce: 0.0070 decode.acc_seg: 99.6040 aux.loss_ce: 0.0082 aux.acc_seg: 98.9573 +04/17 18:34:35 - mmengine - INFO - Iter(train) [ 36100/160000] base_lr: 7.8171e-05 lr: 2.8901e-07 eta: 1 day, 10:24:37 time: 1.0012 data_time: 0.0046 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0075 decode.acc_seg: 99.8489 aux.loss_ce: 0.0078 aux.acc_seg: 99.6672 +04/17 18:35:26 - mmengine - INFO - Iter(train) [ 36150/160000] base_lr: 7.8139e-05 lr: 2.8890e-07 eta: 1 day, 10:23:47 time: 1.0021 data_time: 0.0046 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0082 decode.acc_seg: 99.6416 aux.loss_ce: 0.0091 aux.acc_seg: 99.0370 +04/17 18:36:16 - mmengine - INFO - Iter(train) [ 36200/160000] base_lr: 7.8108e-05 lr: 2.8878e-07 eta: 1 day, 10:22:57 time: 1.0005 data_time: 0.0046 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0077 decode.acc_seg: 99.7070 aux.loss_ce: 0.0089 aux.acc_seg: 99.0919 +04/17 18:37:06 - mmengine - INFO - Iter(train) [ 36250/160000] base_lr: 7.8076e-05 lr: 2.8866e-07 eta: 1 day, 10:22:07 time: 1.0010 data_time: 0.0051 memory: 8462 loss: 0.0141 decode.loss_ce: 0.0067 decode.acc_seg: 99.6822 aux.loss_ce: 0.0074 aux.acc_seg: 98.9012 +04/17 18:37:56 - mmengine - INFO - Iter(train) [ 36300/160000] base_lr: 7.8045e-05 lr: 2.8855e-07 eta: 1 day, 10:21:17 time: 1.0011 data_time: 0.0045 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0086 decode.acc_seg: 99.6410 aux.loss_ce: 0.0093 aux.acc_seg: 98.8859 +04/17 18:38:46 - mmengine - INFO - Iter(train) [ 36350/160000] base_lr: 7.8013e-05 lr: 2.8843e-07 eta: 1 day, 10:20:28 time: 1.0006 data_time: 0.0046 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0083 decode.acc_seg: 99.6387 aux.loss_ce: 0.0089 aux.acc_seg: 98.8016 +04/17 18:39:36 - mmengine - INFO - Iter(train) [ 36400/160000] base_lr: 7.7982e-05 lr: 2.8831e-07 eta: 1 day, 10:19:38 time: 1.0013 data_time: 0.0045 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0078 decode.acc_seg: 99.6830 aux.loss_ce: 0.0077 aux.acc_seg: 99.3126 +04/17 18:40:26 - mmengine - INFO - Iter(train) [ 36450/160000] base_lr: 7.7950e-05 lr: 2.8820e-07 eta: 1 day, 10:18:48 time: 1.0016 data_time: 0.0048 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0082 decode.acc_seg: 99.6511 aux.loss_ce: 0.0093 aux.acc_seg: 98.9363 +04/17 18:41:16 - mmengine - INFO - Iter(train) [ 36500/160000] base_lr: 7.7918e-05 lr: 2.8808e-07 eta: 1 day, 10:17:58 time: 0.9997 data_time: 0.0045 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0085 decode.acc_seg: 99.7471 aux.loss_ce: 0.0093 aux.acc_seg: 99.2018 +04/17 18:42:06 - mmengine - INFO - Iter(train) [ 36550/160000] base_lr: 7.7887e-05 lr: 2.8796e-07 eta: 1 day, 10:17:08 time: 1.0006 data_time: 0.0046 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0073 decode.acc_seg: 99.5436 aux.loss_ce: 0.0085 aux.acc_seg: 98.8192 +04/17 18:42:56 - mmengine - INFO - Iter(train) [ 36600/160000] base_lr: 7.7855e-05 lr: 2.8785e-07 eta: 1 day, 10:16:18 time: 1.0000 data_time: 0.0043 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0078 decode.acc_seg: 99.7873 aux.loss_ce: 0.0083 aux.acc_seg: 99.2937 +04/17 18:43:46 - mmengine - INFO - Iter(train) [ 36650/160000] base_lr: 7.7824e-05 lr: 2.8773e-07 eta: 1 day, 10:15:28 time: 1.0014 data_time: 0.0049 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0075 decode.acc_seg: 99.7637 aux.loss_ce: 0.0077 aux.acc_seg: 99.2682 +04/17 18:44:36 - mmengine - INFO - Iter(train) [ 36700/160000] base_lr: 7.7792e-05 lr: 2.8761e-07 eta: 1 day, 10:14:39 time: 1.0011 data_time: 0.0043 memory: 8462 loss: 0.0180 decode.loss_ce: 0.0090 decode.acc_seg: 99.5399 aux.loss_ce: 0.0090 aux.acc_seg: 99.2033 +04/17 18:45:26 - mmengine - INFO - Iter(train) [ 36750/160000] base_lr: 7.7761e-05 lr: 2.8750e-07 eta: 1 day, 10:13:49 time: 0.9999 data_time: 0.0047 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0084 decode.acc_seg: 99.6468 aux.loss_ce: 0.0090 aux.acc_seg: 98.8775 +04/17 18:46:16 - mmengine - INFO - Iter(train) [ 36800/160000] base_lr: 7.7729e-05 lr: 2.8738e-07 eta: 1 day, 10:12:59 time: 1.0014 data_time: 0.0041 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0074 decode.acc_seg: 99.7473 aux.loss_ce: 0.0083 aux.acc_seg: 99.3780 +04/17 18:47:06 - mmengine - INFO - Iter(train) [ 36850/160000] base_lr: 7.7698e-05 lr: 2.8726e-07 eta: 1 day, 10:12:09 time: 1.0005 data_time: 0.0042 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0082 decode.acc_seg: 99.5848 aux.loss_ce: 0.0096 aux.acc_seg: 98.5756 +04/17 18:47:56 - mmengine - INFO - Iter(train) [ 36900/160000] base_lr: 7.7666e-05 lr: 2.8715e-07 eta: 1 day, 10:11:19 time: 1.0010 data_time: 0.0051 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0076 decode.acc_seg: 99.7629 aux.loss_ce: 0.0081 aux.acc_seg: 99.1388 +04/17 18:48:46 - mmengine - INFO - Iter(train) [ 36950/160000] base_lr: 7.7635e-05 lr: 2.8703e-07 eta: 1 day, 10:10:29 time: 1.0018 data_time: 0.0045 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0070 decode.acc_seg: 99.7015 aux.loss_ce: 0.0070 aux.acc_seg: 99.1478 +04/17 18:49:36 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 18:49:36 - mmengine - INFO - Iter(train) [ 37000/160000] base_lr: 7.7603e-05 lr: 2.8691e-07 eta: 1 day, 10:09:40 time: 1.0006 data_time: 0.0046 memory: 8462 loss: 0.0162 decode.loss_ce: 0.0078 decode.acc_seg: 99.7145 aux.loss_ce: 0.0085 aux.acc_seg: 98.6742 +04/17 18:50:26 - mmengine - INFO - Iter(train) [ 37050/160000] base_lr: 7.7571e-05 lr: 2.8680e-07 eta: 1 day, 10:08:50 time: 1.0004 data_time: 0.0043 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0074 decode.acc_seg: 99.6902 aux.loss_ce: 0.0082 aux.acc_seg: 99.0879 +04/17 18:51:16 - mmengine - INFO - Iter(train) [ 37100/160000] base_lr: 7.7540e-05 lr: 2.8668e-07 eta: 1 day, 10:08:00 time: 1.0007 data_time: 0.0045 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0070 decode.acc_seg: 99.7778 aux.loss_ce: 0.0084 aux.acc_seg: 99.3973 +04/17 18:52:06 - mmengine - INFO - Iter(train) [ 37150/160000] base_lr: 7.7508e-05 lr: 2.8656e-07 eta: 1 day, 10:07:10 time: 1.0002 data_time: 0.0046 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0075 decode.acc_seg: 99.6952 aux.loss_ce: 0.0079 aux.acc_seg: 99.2620 +04/17 18:52:56 - mmengine - INFO - Iter(train) [ 37200/160000] base_lr: 7.7477e-05 lr: 2.8645e-07 eta: 1 day, 10:06:20 time: 0.9997 data_time: 0.0044 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0071 decode.acc_seg: 99.7860 aux.loss_ce: 0.0082 aux.acc_seg: 99.2750 +04/17 18:53:46 - mmengine - INFO - Iter(train) [ 37250/160000] base_lr: 7.7445e-05 lr: 2.8633e-07 eta: 1 day, 10:05:30 time: 0.9995 data_time: 0.0043 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0087 decode.acc_seg: 99.5674 aux.loss_ce: 0.0091 aux.acc_seg: 98.4161 +04/17 18:54:36 - mmengine - INFO - Iter(train) [ 37300/160000] base_lr: 7.7414e-05 lr: 2.8621e-07 eta: 1 day, 10:04:40 time: 1.0016 data_time: 0.0048 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0076 decode.acc_seg: 99.6614 aux.loss_ce: 0.0085 aux.acc_seg: 98.8615 +04/17 18:55:26 - mmengine - INFO - Iter(train) [ 37350/160000] base_lr: 7.7382e-05 lr: 2.8610e-07 eta: 1 day, 10:03:51 time: 1.0001 data_time: 0.0045 memory: 8462 loss: 0.0144 decode.loss_ce: 0.0071 decode.acc_seg: 99.6620 aux.loss_ce: 0.0073 aux.acc_seg: 99.1179 +04/17 18:56:17 - mmengine - INFO - Iter(train) [ 37400/160000] base_lr: 7.7351e-05 lr: 2.8598e-07 eta: 1 day, 10:03:01 time: 1.0004 data_time: 0.0047 memory: 8462 loss: 0.0164 decode.loss_ce: 0.0077 decode.acc_seg: 99.6948 aux.loss_ce: 0.0087 aux.acc_seg: 99.0623 +04/17 18:57:07 - mmengine - INFO - Iter(train) [ 37450/160000] base_lr: 7.7319e-05 lr: 2.8586e-07 eta: 1 day, 10:02:11 time: 0.9997 data_time: 0.0044 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0080 decode.acc_seg: 99.6958 aux.loss_ce: 0.0087 aux.acc_seg: 99.1169 +04/17 18:57:57 - mmengine - INFO - Iter(train) [ 37500/160000] base_lr: 7.7288e-05 lr: 2.8575e-07 eta: 1 day, 10:01:21 time: 0.9990 data_time: 0.0042 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0069 decode.acc_seg: 99.7894 aux.loss_ce: 0.0079 aux.acc_seg: 99.3212 +04/17 18:58:47 - mmengine - INFO - Iter(train) [ 37550/160000] base_lr: 7.7256e-05 lr: 2.8563e-07 eta: 1 day, 10:00:31 time: 1.0024 data_time: 0.0059 memory: 8462 loss: 0.0196 decode.loss_ce: 0.0102 decode.acc_seg: 99.4537 aux.loss_ce: 0.0094 aux.acc_seg: 99.0900 +04/17 18:59:37 - mmengine - INFO - Iter(train) [ 37600/160000] base_lr: 7.7224e-05 lr: 2.8551e-07 eta: 1 day, 9:59:42 time: 1.0004 data_time: 0.0043 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0081 decode.acc_seg: 99.5842 aux.loss_ce: 0.0087 aux.acc_seg: 98.7751 +04/17 19:00:27 - mmengine - INFO - Iter(train) [ 37650/160000] base_lr: 7.7193e-05 lr: 2.8540e-07 eta: 1 day, 9:58:52 time: 1.0025 data_time: 0.0047 memory: 8462 loss: 0.0133 decode.loss_ce: 0.0064 decode.acc_seg: 99.8184 aux.loss_ce: 0.0070 aux.acc_seg: 99.5562 +04/17 19:01:17 - mmengine - INFO - Iter(train) [ 37700/160000] base_lr: 7.7161e-05 lr: 2.8528e-07 eta: 1 day, 9:58:02 time: 1.0024 data_time: 0.0043 memory: 8462 loss: 0.0145 decode.loss_ce: 0.0072 decode.acc_seg: 99.4846 aux.loss_ce: 0.0073 aux.acc_seg: 98.8129 +04/17 19:02:07 - mmengine - INFO - Iter(train) [ 37750/160000] base_lr: 7.7130e-05 lr: 2.8516e-07 eta: 1 day, 9:57:12 time: 1.0009 data_time: 0.0042 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0073 decode.acc_seg: 99.6445 aux.loss_ce: 0.0082 aux.acc_seg: 99.0534 +04/17 19:02:57 - mmengine - INFO - Iter(train) [ 37800/160000] base_lr: 7.7098e-05 lr: 2.8505e-07 eta: 1 day, 9:56:23 time: 1.0005 data_time: 0.0045 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0078 decode.acc_seg: 99.7906 aux.loss_ce: 0.0078 aux.acc_seg: 99.3464 +04/17 19:03:47 - mmengine - INFO - Iter(train) [ 37850/160000] base_lr: 7.7067e-05 lr: 2.8493e-07 eta: 1 day, 9:55:33 time: 1.0015 data_time: 0.0043 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0088 decode.acc_seg: 99.6288 aux.loss_ce: 0.0089 aux.acc_seg: 99.0606 +04/17 19:04:37 - mmengine - INFO - Iter(train) [ 37900/160000] base_lr: 7.7035e-05 lr: 2.8481e-07 eta: 1 day, 9:54:43 time: 1.0019 data_time: 0.0048 memory: 8462 loss: 0.0199 decode.loss_ce: 0.0099 decode.acc_seg: 99.6737 aux.loss_ce: 0.0100 aux.acc_seg: 98.9614 +04/17 19:05:27 - mmengine - INFO - Iter(train) [ 37950/160000] base_lr: 7.7004e-05 lr: 2.8470e-07 eta: 1 day, 9:53:53 time: 1.0014 data_time: 0.0047 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0077 decode.acc_seg: 99.6334 aux.loss_ce: 0.0080 aux.acc_seg: 99.1543 +04/17 19:06:17 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 19:06:17 - mmengine - INFO - Iter(train) [ 38000/160000] base_lr: 7.6972e-05 lr: 2.8458e-07 eta: 1 day, 9:53:04 time: 1.0017 data_time: 0.0045 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0075 decode.acc_seg: 99.7446 aux.loss_ce: 0.0080 aux.acc_seg: 99.4513 +04/17 19:07:07 - mmengine - INFO - Iter(train) [ 38050/160000] base_lr: 7.6941e-05 lr: 2.8447e-07 eta: 1 day, 9:52:14 time: 1.0026 data_time: 0.0044 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0074 decode.acc_seg: 99.6555 aux.loss_ce: 0.0080 aux.acc_seg: 99.0978 +04/17 19:07:57 - mmengine - INFO - Iter(train) [ 38100/160000] base_lr: 7.6909e-05 lr: 2.8435e-07 eta: 1 day, 9:51:24 time: 1.0015 data_time: 0.0049 memory: 8462 loss: 0.0159 decode.loss_ce: 0.0078 decode.acc_seg: 99.6794 aux.loss_ce: 0.0081 aux.acc_seg: 99.0656 +04/17 19:08:48 - mmengine - INFO - Iter(train) [ 38150/160000] base_lr: 7.6877e-05 lr: 2.8423e-07 eta: 1 day, 9:50:34 time: 1.0009 data_time: 0.0048 memory: 8462 loss: 0.0163 decode.loss_ce: 0.0079 decode.acc_seg: 99.7627 aux.loss_ce: 0.0085 aux.acc_seg: 99.2317 +04/17 19:09:38 - mmengine - INFO - Iter(train) [ 38200/160000] base_lr: 7.6846e-05 lr: 2.8412e-07 eta: 1 day, 9:49:45 time: 1.0015 data_time: 0.0044 memory: 8462 loss: 0.0146 decode.loss_ce: 0.0073 decode.acc_seg: 99.7185 aux.loss_ce: 0.0073 aux.acc_seg: 99.0358 +04/17 19:10:28 - mmengine - INFO - Iter(train) [ 38250/160000] base_lr: 7.6814e-05 lr: 2.8400e-07 eta: 1 day, 9:48:55 time: 1.0012 data_time: 0.0044 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0079 decode.acc_seg: 99.7562 aux.loss_ce: 0.0093 aux.acc_seg: 99.3286 +04/17 19:11:18 - mmengine - INFO - Iter(train) [ 38300/160000] base_lr: 7.6783e-05 lr: 2.8388e-07 eta: 1 day, 9:48:05 time: 1.0018 data_time: 0.0052 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0074 decode.acc_seg: 99.7210 aux.loss_ce: 0.0080 aux.acc_seg: 99.2395 +04/17 19:12:08 - mmengine - INFO - Iter(train) [ 38350/160000] base_lr: 7.6751e-05 lr: 2.8377e-07 eta: 1 day, 9:47:15 time: 1.0011 data_time: 0.0043 memory: 8462 loss: 0.0189 decode.loss_ce: 0.0093 decode.acc_seg: 99.5348 aux.loss_ce: 0.0096 aux.acc_seg: 98.8853 +04/17 19:12:58 - mmengine - INFO - Iter(train) [ 38400/160000] base_lr: 7.6720e-05 lr: 2.8365e-07 eta: 1 day, 9:46:25 time: 1.0017 data_time: 0.0046 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0080 decode.acc_seg: 99.7086 aux.loss_ce: 0.0080 aux.acc_seg: 99.3008 +04/17 19:13:48 - mmengine - INFO - Iter(train) [ 38450/160000] base_lr: 7.6688e-05 lr: 2.8353e-07 eta: 1 day, 9:45:36 time: 1.0002 data_time: 0.0046 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0084 decode.acc_seg: 99.4303 aux.loss_ce: 0.0089 aux.acc_seg: 98.4764 +04/17 19:14:38 - mmengine - INFO - Iter(train) [ 38500/160000] base_lr: 7.6657e-05 lr: 2.8342e-07 eta: 1 day, 9:44:46 time: 1.0006 data_time: 0.0044 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0073 decode.acc_seg: 99.7578 aux.loss_ce: 0.0077 aux.acc_seg: 99.1474 +04/17 19:15:28 - mmengine - INFO - Iter(train) [ 38550/160000] base_lr: 7.6625e-05 lr: 2.8330e-07 eta: 1 day, 9:43:56 time: 1.0016 data_time: 0.0049 memory: 8462 loss: 0.0137 decode.loss_ce: 0.0066 decode.acc_seg: 99.7684 aux.loss_ce: 0.0071 aux.acc_seg: 99.2023 +04/17 19:16:18 - mmengine - INFO - Iter(train) [ 38600/160000] base_lr: 7.6594e-05 lr: 2.8318e-07 eta: 1 day, 9:43:06 time: 1.0022 data_time: 0.0044 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0074 decode.acc_seg: 99.2529 aux.loss_ce: 0.0075 aux.acc_seg: 98.7282 +04/17 19:17:08 - mmengine - INFO - Iter(train) [ 38650/160000] base_lr: 7.6562e-05 lr: 2.8307e-07 eta: 1 day, 9:42:16 time: 1.0006 data_time: 0.0046 memory: 8462 loss: 0.0164 decode.loss_ce: 0.0077 decode.acc_seg: 99.5695 aux.loss_ce: 0.0087 aux.acc_seg: 98.8285 +04/17 19:17:58 - mmengine - INFO - Iter(train) [ 38700/160000] base_lr: 7.6530e-05 lr: 2.8295e-07 eta: 1 day, 9:41:27 time: 1.0010 data_time: 0.0044 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0071 decode.acc_seg: 99.7311 aux.loss_ce: 0.0079 aux.acc_seg: 99.1293 +04/17 19:18:48 - mmengine - INFO - Iter(train) [ 38750/160000] base_lr: 7.6499e-05 lr: 2.8283e-07 eta: 1 day, 9:40:37 time: 1.0011 data_time: 0.0047 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0067 decode.acc_seg: 99.7602 aux.loss_ce: 0.0075 aux.acc_seg: 99.3006 +04/17 19:19:38 - mmengine - INFO - Iter(train) [ 38800/160000] base_lr: 7.6467e-05 lr: 2.8272e-07 eta: 1 day, 9:39:47 time: 0.9997 data_time: 0.0046 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0089 decode.acc_seg: 99.7873 aux.loss_ce: 0.0086 aux.acc_seg: 99.3906 +04/17 19:20:28 - mmengine - INFO - Iter(train) [ 38850/160000] base_lr: 7.6436e-05 lr: 2.8260e-07 eta: 1 day, 9:38:57 time: 1.0017 data_time: 0.0048 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0081 decode.acc_seg: 99.6780 aux.loss_ce: 0.0087 aux.acc_seg: 99.0749 +04/17 19:21:18 - mmengine - INFO - Iter(train) [ 38900/160000] base_lr: 7.6404e-05 lr: 2.8248e-07 eta: 1 day, 9:38:07 time: 1.0007 data_time: 0.0046 memory: 8462 loss: 0.0187 decode.loss_ce: 0.0090 decode.acc_seg: 99.6393 aux.loss_ce: 0.0096 aux.acc_seg: 99.0166 +04/17 19:22:08 - mmengine - INFO - Iter(train) [ 38950/160000] base_lr: 7.6373e-05 lr: 2.8237e-07 eta: 1 day, 9:37:17 time: 1.0005 data_time: 0.0045 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0076 decode.acc_seg: 99.7993 aux.loss_ce: 0.0089 aux.acc_seg: 99.1705 +04/17 19:22:58 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 19:22:58 - mmengine - INFO - Iter(train) [ 39000/160000] base_lr: 7.6341e-05 lr: 2.8225e-07 eta: 1 day, 9:36:27 time: 0.9992 data_time: 0.0044 memory: 8462 loss: 0.0184 decode.loss_ce: 0.0089 decode.acc_seg: 99.6943 aux.loss_ce: 0.0095 aux.acc_seg: 98.9830 +04/17 19:23:48 - mmengine - INFO - Iter(train) [ 39050/160000] base_lr: 7.6310e-05 lr: 2.8213e-07 eta: 1 day, 9:35:38 time: 1.0006 data_time: 0.0044 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0087 decode.acc_seg: 99.6820 aux.loss_ce: 0.0088 aux.acc_seg: 99.0828 +04/17 19:24:39 - mmengine - INFO - Iter(train) [ 39100/160000] base_lr: 7.6278e-05 lr: 2.8202e-07 eta: 1 day, 9:34:48 time: 1.0006 data_time: 0.0046 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0073 decode.acc_seg: 99.7107 aux.loss_ce: 0.0075 aux.acc_seg: 99.1867 +04/17 19:25:29 - mmengine - INFO - Iter(train) [ 39150/160000] base_lr: 7.6247e-05 lr: 2.8190e-07 eta: 1 day, 9:33:58 time: 0.9999 data_time: 0.0042 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0087 decode.acc_seg: 99.7093 aux.loss_ce: 0.0084 aux.acc_seg: 99.2432 +04/17 19:26:19 - mmengine - INFO - Iter(train) [ 39200/160000] base_lr: 7.6215e-05 lr: 2.8178e-07 eta: 1 day, 9:33:08 time: 1.0016 data_time: 0.0045 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0070 decode.acc_seg: 99.6433 aux.loss_ce: 0.0077 aux.acc_seg: 99.0658 +04/17 19:27:09 - mmengine - INFO - Iter(train) [ 39250/160000] base_lr: 7.6183e-05 lr: 2.8167e-07 eta: 1 day, 9:32:18 time: 1.0018 data_time: 0.0041 memory: 8462 loss: 0.0180 decode.loss_ce: 0.0090 decode.acc_seg: 99.6525 aux.loss_ce: 0.0090 aux.acc_seg: 99.0257 +04/17 19:27:59 - mmengine - INFO - Iter(train) [ 39300/160000] base_lr: 7.6152e-05 lr: 2.8155e-07 eta: 1 day, 9:31:28 time: 1.0014 data_time: 0.0045 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0074 decode.acc_seg: 99.8295 aux.loss_ce: 0.0080 aux.acc_seg: 99.4982 +04/17 19:28:49 - mmengine - INFO - Iter(train) [ 39350/160000] base_lr: 7.6120e-05 lr: 2.8143e-07 eta: 1 day, 9:30:38 time: 1.0003 data_time: 0.0044 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0078 decode.acc_seg: 99.5817 aux.loss_ce: 0.0094 aux.acc_seg: 98.6925 +04/17 19:29:39 - mmengine - INFO - Iter(train) [ 39400/160000] base_lr: 7.6089e-05 lr: 2.8132e-07 eta: 1 day, 9:29:49 time: 1.0016 data_time: 0.0041 memory: 8462 loss: 0.0163 decode.loss_ce: 0.0078 decode.acc_seg: 99.7734 aux.loss_ce: 0.0085 aux.acc_seg: 99.3156 +04/17 19:30:29 - mmengine - INFO - Iter(train) [ 39450/160000] base_lr: 7.6057e-05 lr: 2.8120e-07 eta: 1 day, 9:28:59 time: 1.0000 data_time: 0.0046 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0088 decode.acc_seg: 99.7704 aux.loss_ce: 0.0089 aux.acc_seg: 99.4074 +04/17 19:31:19 - mmengine - INFO - Iter(train) [ 39500/160000] base_lr: 7.6026e-05 lr: 2.8108e-07 eta: 1 day, 9:28:09 time: 1.0006 data_time: 0.0045 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0072 decode.acc_seg: 99.7780 aux.loss_ce: 0.0074 aux.acc_seg: 99.3631 +04/17 19:32:09 - mmengine - INFO - Iter(train) [ 39550/160000] base_lr: 7.5994e-05 lr: 2.8097e-07 eta: 1 day, 9:27:19 time: 1.0022 data_time: 0.0047 memory: 8462 loss: 0.0146 decode.loss_ce: 0.0069 decode.acc_seg: 99.7141 aux.loss_ce: 0.0077 aux.acc_seg: 99.1346 +04/17 19:32:59 - mmengine - INFO - Iter(train) [ 39600/160000] base_lr: 7.5963e-05 lr: 2.8085e-07 eta: 1 day, 9:26:29 time: 1.0008 data_time: 0.0051 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0074 decode.acc_seg: 99.7297 aux.loss_ce: 0.0081 aux.acc_seg: 99.2731 +04/17 19:33:49 - mmengine - INFO - Iter(train) [ 39650/160000] base_lr: 7.5931e-05 lr: 2.8073e-07 eta: 1 day, 9:25:39 time: 1.0010 data_time: 0.0046 memory: 8462 loss: 0.0142 decode.loss_ce: 0.0071 decode.acc_seg: 99.4810 aux.loss_ce: 0.0071 aux.acc_seg: 98.8874 +04/17 19:34:39 - mmengine - INFO - Iter(train) [ 39700/160000] base_lr: 7.5900e-05 lr: 2.8062e-07 eta: 1 day, 9:24:50 time: 1.0007 data_time: 0.0043 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0078 decode.acc_seg: 99.7000 aux.loss_ce: 0.0090 aux.acc_seg: 99.2617 +04/17 19:35:29 - mmengine - INFO - Iter(train) [ 39750/160000] base_lr: 7.5868e-05 lr: 2.8050e-07 eta: 1 day, 9:24:00 time: 1.0003 data_time: 0.0045 memory: 8462 loss: 0.0162 decode.loss_ce: 0.0078 decode.acc_seg: 99.6181 aux.loss_ce: 0.0084 aux.acc_seg: 98.8501 +04/17 19:36:19 - mmengine - INFO - Iter(train) [ 39800/160000] base_lr: 7.5836e-05 lr: 2.8038e-07 eta: 1 day, 9:23:10 time: 1.0013 data_time: 0.0043 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0077 decode.acc_seg: 99.7023 aux.loss_ce: 0.0088 aux.acc_seg: 99.0917 +04/17 19:37:09 - mmengine - INFO - Iter(train) [ 39850/160000] base_lr: 7.5805e-05 lr: 2.8027e-07 eta: 1 day, 9:22:20 time: 1.0010 data_time: 0.0049 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0079 decode.acc_seg: 99.6706 aux.loss_ce: 0.0086 aux.acc_seg: 98.9368 +04/17 19:37:59 - mmengine - INFO - Iter(train) [ 39900/160000] base_lr: 7.5773e-05 lr: 2.8015e-07 eta: 1 day, 9:21:30 time: 1.0016 data_time: 0.0052 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0076 decode.acc_seg: 99.4333 aux.loss_ce: 0.0079 aux.acc_seg: 98.9017 +04/17 19:38:49 - mmengine - INFO - Iter(train) [ 39950/160000] base_lr: 7.5742e-05 lr: 2.8003e-07 eta: 1 day, 9:20:41 time: 1.0004 data_time: 0.0045 memory: 8462 loss: 0.0186 decode.loss_ce: 0.0091 decode.acc_seg: 99.6988 aux.loss_ce: 0.0096 aux.acc_seg: 99.2237 +04/17 19:39:40 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 19:39:40 - mmengine - INFO - Iter(train) [ 40000/160000] base_lr: 7.5710e-05 lr: 2.7992e-07 eta: 1 day, 9:19:51 time: 1.0016 data_time: 0.0046 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0072 decode.acc_seg: 99.6504 aux.loss_ce: 0.0081 aux.acc_seg: 99.0877 +04/17 19:39:40 - mmengine - INFO - Saving checkpoint at 40000 iterations +04/17 19:39:49 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:17 time: 0.1158 data_time: 0.0016 memory: 4004 +04/17 19:39:55 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:11 time: 0.1157 data_time: 0.0015 memory: 4004 +04/17 19:40:01 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:05 time: 0.1156 data_time: 0.0014 memory: 4004 +04/17 19:40:07 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.1155 data_time: 0.0012 memory: 4004 +04/17 19:40:07 - mmengine - INFO - per class results: +04/17 19:40:07 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 99.18 | 99.58 | 99.59 | 99.6 | 99.58 | +| contrast | 81.99 | 90.36 | 90.1 | 89.85 | 90.36 | ++------------+-------+-------+--------+-----------+--------+ +04/17 19:40:07 - mmengine - INFO - Iter(val) [200/200] aAcc: 99.2100 mIoU: 90.5800 mAcc: 94.9700 mFscore: 94.8500 mPrecision: 94.7200 mRecall: 94.9700 data_time: 0.0018 time: 0.1161 +04/17 19:40:57 - mmengine - INFO - Iter(train) [ 40050/160000] base_lr: 7.5679e-05 lr: 2.7980e-07 eta: 1 day, 9:19:02 time: 0.9999 data_time: 0.0051 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0085 decode.acc_seg: 99.7334 aux.loss_ce: 0.0089 aux.acc_seg: 99.3618 +04/17 19:41:47 - mmengine - INFO - Iter(train) [ 40100/160000] base_lr: 7.5647e-05 lr: 2.7968e-07 eta: 1 day, 9:18:12 time: 1.0000 data_time: 0.0047 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0078 decode.acc_seg: 99.6231 aux.loss_ce: 0.0079 aux.acc_seg: 99.2281 +04/17 19:42:37 - mmengine - INFO - Iter(train) [ 40150/160000] base_lr: 7.5616e-05 lr: 2.7957e-07 eta: 1 day, 9:17:22 time: 1.0004 data_time: 0.0049 memory: 8462 loss: 0.0197 decode.loss_ce: 0.0102 decode.acc_seg: 99.7002 aux.loss_ce: 0.0094 aux.acc_seg: 99.0236 +04/17 19:43:27 - mmengine - INFO - Iter(train) [ 40200/160000] base_lr: 7.5584e-05 lr: 2.7945e-07 eta: 1 day, 9:16:32 time: 1.0012 data_time: 0.0054 memory: 8462 loss: 0.0142 decode.loss_ce: 0.0066 decode.acc_seg: 99.8011 aux.loss_ce: 0.0076 aux.acc_seg: 99.4436 +04/17 19:44:17 - mmengine - INFO - Iter(train) [ 40250/160000] base_lr: 7.5553e-05 lr: 2.7933e-07 eta: 1 day, 9:15:43 time: 1.0022 data_time: 0.0048 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0076 decode.acc_seg: 99.6887 aux.loss_ce: 0.0077 aux.acc_seg: 99.2823 +04/17 19:45:07 - mmengine - INFO - Iter(train) [ 40300/160000] base_lr: 7.5521e-05 lr: 2.7922e-07 eta: 1 day, 9:14:53 time: 1.0010 data_time: 0.0048 memory: 8462 loss: 0.0141 decode.loss_ce: 0.0069 decode.acc_seg: 99.7808 aux.loss_ce: 0.0072 aux.acc_seg: 99.4278 +04/17 19:45:57 - mmengine - INFO - Iter(train) [ 40350/160000] base_lr: 7.5489e-05 lr: 2.7910e-07 eta: 1 day, 9:14:03 time: 1.0020 data_time: 0.0047 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0068 decode.acc_seg: 99.7711 aux.loss_ce: 0.0072 aux.acc_seg: 99.3299 +04/17 19:46:48 - mmengine - INFO - Iter(train) [ 40400/160000] base_lr: 7.5458e-05 lr: 2.7898e-07 eta: 1 day, 9:13:13 time: 1.0018 data_time: 0.0045 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0076 decode.acc_seg: 99.6933 aux.loss_ce: 0.0077 aux.acc_seg: 99.2861 +04/17 19:47:38 - mmengine - INFO - Iter(train) [ 40450/160000] base_lr: 7.5426e-05 lr: 2.7887e-07 eta: 1 day, 9:12:23 time: 1.0012 data_time: 0.0046 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0074 decode.acc_seg: 99.6700 aux.loss_ce: 0.0086 aux.acc_seg: 99.1879 +04/17 19:48:28 - mmengine - INFO - Iter(train) [ 40500/160000] base_lr: 7.5395e-05 lr: 2.7875e-07 eta: 1 day, 9:11:34 time: 1.0008 data_time: 0.0045 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0085 decode.acc_seg: 99.6632 aux.loss_ce: 0.0086 aux.acc_seg: 98.9334 +04/17 19:49:18 - mmengine - INFO - Iter(train) [ 40550/160000] base_lr: 7.5363e-05 lr: 2.7863e-07 eta: 1 day, 9:10:44 time: 1.0011 data_time: 0.0048 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0077 decode.acc_seg: 99.7723 aux.loss_ce: 0.0081 aux.acc_seg: 99.3284 +04/17 19:50:08 - mmengine - INFO - Iter(train) [ 40600/160000] base_lr: 7.5332e-05 lr: 2.7852e-07 eta: 1 day, 9:09:54 time: 1.0024 data_time: 0.0044 memory: 8462 loss: 0.0178 decode.loss_ce: 0.0085 decode.acc_seg: 99.6166 aux.loss_ce: 0.0094 aux.acc_seg: 98.9628 +04/17 19:50:58 - mmengine - INFO - Iter(train) [ 40650/160000] base_lr: 7.5300e-05 lr: 2.7840e-07 eta: 1 day, 9:09:04 time: 1.0010 data_time: 0.0042 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0081 decode.acc_seg: 99.6910 aux.loss_ce: 0.0086 aux.acc_seg: 99.0755 +04/17 19:51:48 - mmengine - INFO - Iter(train) [ 40700/160000] base_lr: 7.5269e-05 lr: 2.7828e-07 eta: 1 day, 9:08:14 time: 1.0011 data_time: 0.0045 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0071 decode.acc_seg: 99.5726 aux.loss_ce: 0.0078 aux.acc_seg: 98.9605 +04/17 19:52:38 - mmengine - INFO - Iter(train) [ 40750/160000] base_lr: 7.5237e-05 lr: 2.7817e-07 eta: 1 day, 9:07:25 time: 1.0007 data_time: 0.0045 memory: 8462 loss: 0.0159 decode.loss_ce: 0.0080 decode.acc_seg: 99.7034 aux.loss_ce: 0.0079 aux.acc_seg: 99.1114 +04/17 19:53:28 - mmengine - INFO - Iter(train) [ 40800/160000] base_lr: 7.5206e-05 lr: 2.7805e-07 eta: 1 day, 9:06:35 time: 1.0015 data_time: 0.0047 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0085 decode.acc_seg: 99.6832 aux.loss_ce: 0.0089 aux.acc_seg: 98.9595 +04/17 19:54:18 - mmengine - INFO - Iter(train) [ 40850/160000] base_lr: 7.5174e-05 lr: 2.7793e-07 eta: 1 day, 9:05:45 time: 1.0010 data_time: 0.0047 memory: 8462 loss: 0.0184 decode.loss_ce: 0.0090 decode.acc_seg: 99.6548 aux.loss_ce: 0.0094 aux.acc_seg: 99.0057 +04/17 19:55:08 - mmengine - INFO - Iter(train) [ 40900/160000] base_lr: 7.5142e-05 lr: 2.7782e-07 eta: 1 day, 9:04:55 time: 1.0018 data_time: 0.0046 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0067 decode.acc_seg: 99.7723 aux.loss_ce: 0.0080 aux.acc_seg: 99.2199 +04/17 19:55:58 - mmengine - INFO - Iter(train) [ 40950/160000] base_lr: 7.5111e-05 lr: 2.7770e-07 eta: 1 day, 9:04:05 time: 1.0011 data_time: 0.0048 memory: 8462 loss: 0.0185 decode.loss_ce: 0.0096 decode.acc_seg: 99.6653 aux.loss_ce: 0.0089 aux.acc_seg: 98.8714 +04/17 19:56:48 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 19:56:48 - mmengine - INFO - Iter(train) [ 41000/160000] base_lr: 7.5079e-05 lr: 2.7758e-07 eta: 1 day, 9:03:15 time: 1.0021 data_time: 0.0052 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0080 decode.acc_seg: 99.6866 aux.loss_ce: 0.0088 aux.acc_seg: 99.2544 +04/17 19:57:38 - mmengine - INFO - Iter(train) [ 41050/160000] base_lr: 7.5048e-05 lr: 2.7747e-07 eta: 1 day, 9:02:26 time: 1.0022 data_time: 0.0046 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0072 decode.acc_seg: 99.7316 aux.loss_ce: 0.0075 aux.acc_seg: 99.3258 +04/17 19:58:28 - mmengine - INFO - Iter(train) [ 41100/160000] base_lr: 7.5016e-05 lr: 2.7735e-07 eta: 1 day, 9:01:36 time: 1.0011 data_time: 0.0048 memory: 8462 loss: 0.0184 decode.loss_ce: 0.0085 decode.acc_seg: 99.5567 aux.loss_ce: 0.0099 aux.acc_seg: 98.6467 +04/17 19:59:18 - mmengine - INFO - Iter(train) [ 41150/160000] base_lr: 7.4985e-05 lr: 2.7723e-07 eta: 1 day, 9:00:46 time: 1.0007 data_time: 0.0048 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0071 decode.acc_seg: 99.5911 aux.loss_ce: 0.0077 aux.acc_seg: 98.8766 +04/17 20:00:09 - mmengine - INFO - Iter(train) [ 41200/160000] base_lr: 7.4953e-05 lr: 2.7712e-07 eta: 1 day, 8:59:56 time: 1.0023 data_time: 0.0050 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0065 decode.acc_seg: 99.7635 aux.loss_ce: 0.0078 aux.acc_seg: 99.2750 +04/17 20:00:59 - mmengine - INFO - Iter(train) [ 41250/160000] base_lr: 7.4922e-05 lr: 2.7700e-07 eta: 1 day, 8:59:06 time: 1.0000 data_time: 0.0045 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0090 decode.acc_seg: 99.6201 aux.loss_ce: 0.0088 aux.acc_seg: 98.7860 +04/17 20:01:49 - mmengine - INFO - Iter(train) [ 41300/160000] base_lr: 7.4890e-05 lr: 2.7688e-07 eta: 1 day, 8:58:16 time: 1.0025 data_time: 0.0049 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0083 decode.acc_seg: 99.7313 aux.loss_ce: 0.0084 aux.acc_seg: 99.1772 +04/17 20:02:39 - mmengine - INFO - Iter(train) [ 41350/160000] base_lr: 7.4859e-05 lr: 2.7677e-07 eta: 1 day, 8:57:27 time: 1.0009 data_time: 0.0046 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0079 decode.acc_seg: 99.5749 aux.loss_ce: 0.0093 aux.acc_seg: 98.8382 +04/17 20:03:29 - mmengine - INFO - Iter(train) [ 41400/160000] base_lr: 7.4827e-05 lr: 2.7665e-07 eta: 1 day, 8:56:37 time: 1.0015 data_time: 0.0044 memory: 8462 loss: 0.0204 decode.loss_ce: 0.0106 decode.acc_seg: 99.4713 aux.loss_ce: 0.0098 aux.acc_seg: 98.8861 +04/17 20:04:19 - mmengine - INFO - Iter(train) [ 41450/160000] base_lr: 7.4795e-05 lr: 2.7653e-07 eta: 1 day, 8:55:47 time: 1.0011 data_time: 0.0045 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0071 decode.acc_seg: 99.6305 aux.loss_ce: 0.0077 aux.acc_seg: 99.0713 +04/17 20:05:09 - mmengine - INFO - Iter(train) [ 41500/160000] base_lr: 7.4764e-05 lr: 2.7642e-07 eta: 1 day, 8:54:57 time: 1.0014 data_time: 0.0042 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0065 decode.acc_seg: 99.6655 aux.loss_ce: 0.0078 aux.acc_seg: 98.9206 +04/17 20:05:59 - mmengine - INFO - Iter(train) [ 41550/160000] base_lr: 7.4732e-05 lr: 2.7630e-07 eta: 1 day, 8:54:07 time: 1.0006 data_time: 0.0044 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0073 decode.acc_seg: 99.5380 aux.loss_ce: 0.0084 aux.acc_seg: 98.7995 +04/17 20:06:49 - mmengine - INFO - Iter(train) [ 41600/160000] base_lr: 7.4701e-05 lr: 2.7618e-07 eta: 1 day, 8:53:17 time: 1.0016 data_time: 0.0048 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0067 decode.acc_seg: 99.6614 aux.loss_ce: 0.0076 aux.acc_seg: 98.7581 +04/17 20:07:39 - mmengine - INFO - Iter(train) [ 41650/160000] base_lr: 7.4669e-05 lr: 2.7607e-07 eta: 1 day, 8:52:28 time: 1.0018 data_time: 0.0048 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0071 decode.acc_seg: 99.7755 aux.loss_ce: 0.0078 aux.acc_seg: 99.2527 +04/17 20:08:29 - mmengine - INFO - Iter(train) [ 41700/160000] base_lr: 7.4638e-05 lr: 2.7595e-07 eta: 1 day, 8:51:38 time: 1.0032 data_time: 0.0043 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0079 decode.acc_seg: 99.6704 aux.loss_ce: 0.0079 aux.acc_seg: 98.9634 +04/17 20:09:19 - mmengine - INFO - Iter(train) [ 41750/160000] base_lr: 7.4606e-05 lr: 2.7583e-07 eta: 1 day, 8:50:48 time: 1.0015 data_time: 0.0041 memory: 8462 loss: 0.0131 decode.loss_ce: 0.0062 decode.acc_seg: 99.8398 aux.loss_ce: 0.0068 aux.acc_seg: 99.5270 +04/17 20:10:09 - mmengine - INFO - Iter(train) [ 41800/160000] base_lr: 7.4575e-05 lr: 2.7572e-07 eta: 1 day, 8:49:58 time: 1.0016 data_time: 0.0046 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0082 decode.acc_seg: 99.6449 aux.loss_ce: 0.0090 aux.acc_seg: 98.7869 +04/17 20:10:59 - mmengine - INFO - Iter(train) [ 41850/160000] base_lr: 7.4543e-05 lr: 2.7560e-07 eta: 1 day, 8:49:08 time: 1.0008 data_time: 0.0046 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0075 decode.acc_seg: 99.7562 aux.loss_ce: 0.0078 aux.acc_seg: 99.2947 +04/17 20:11:49 - mmengine - INFO - Iter(train) [ 41900/160000] base_lr: 7.4512e-05 lr: 2.7548e-07 eta: 1 day, 8:48:19 time: 1.0013 data_time: 0.0049 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0072 decode.acc_seg: 99.7164 aux.loss_ce: 0.0075 aux.acc_seg: 99.2901 +04/17 20:12:39 - mmengine - INFO - Iter(train) [ 41950/160000] base_lr: 7.4480e-05 lr: 2.7537e-07 eta: 1 day, 8:47:29 time: 1.0010 data_time: 0.0047 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0077 decode.acc_seg: 99.6908 aux.loss_ce: 0.0084 aux.acc_seg: 99.1018 +04/17 20:13:30 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 20:13:30 - mmengine - INFO - Iter(train) [ 42000/160000] base_lr: 7.4448e-05 lr: 2.7525e-07 eta: 1 day, 8:46:39 time: 1.0012 data_time: 0.0046 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0083 decode.acc_seg: 99.6696 aux.loss_ce: 0.0085 aux.acc_seg: 99.3273 +04/17 20:14:20 - mmengine - INFO - Iter(train) [ 42050/160000] base_lr: 7.4417e-05 lr: 2.7513e-07 eta: 1 day, 8:45:49 time: 1.0016 data_time: 0.0045 memory: 8462 loss: 0.0181 decode.loss_ce: 0.0099 decode.acc_seg: 99.7437 aux.loss_ce: 0.0082 aux.acc_seg: 99.4400 +04/17 20:15:10 - mmengine - INFO - Iter(train) [ 42100/160000] base_lr: 7.4385e-05 lr: 2.7502e-07 eta: 1 day, 8:44:59 time: 1.0019 data_time: 0.0045 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0090 decode.acc_seg: 99.7183 aux.loss_ce: 0.0098 aux.acc_seg: 99.2462 +04/17 20:16:00 - mmengine - INFO - Iter(train) [ 42150/160000] base_lr: 7.4354e-05 lr: 2.7490e-07 eta: 1 day, 8:44:09 time: 1.0017 data_time: 0.0044 memory: 8462 loss: 0.0164 decode.loss_ce: 0.0077 decode.acc_seg: 99.7400 aux.loss_ce: 0.0087 aux.acc_seg: 99.2044 +04/17 20:16:50 - mmengine - INFO - Iter(train) [ 42200/160000] base_lr: 7.4322e-05 lr: 2.7478e-07 eta: 1 day, 8:43:19 time: 1.0005 data_time: 0.0045 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0080 decode.acc_seg: 99.5975 aux.loss_ce: 0.0090 aux.acc_seg: 99.0339 +04/17 20:17:40 - mmengine - INFO - Iter(train) [ 42250/160000] base_lr: 7.4291e-05 lr: 2.7467e-07 eta: 1 day, 8:42:30 time: 1.0018 data_time: 0.0054 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0081 decode.acc_seg: 99.6281 aux.loss_ce: 0.0090 aux.acc_seg: 98.9731 +04/17 20:18:30 - mmengine - INFO - Iter(train) [ 42300/160000] base_lr: 7.4259e-05 lr: 2.7455e-07 eta: 1 day, 8:41:40 time: 1.0004 data_time: 0.0046 memory: 8462 loss: 0.0188 decode.loss_ce: 0.0093 decode.acc_seg: 99.3917 aux.loss_ce: 0.0095 aux.acc_seg: 98.3717 +04/17 20:19:20 - mmengine - INFO - Iter(train) [ 42350/160000] base_lr: 7.4228e-05 lr: 2.7443e-07 eta: 1 day, 8:40:50 time: 1.0015 data_time: 0.0057 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0086 decode.acc_seg: 99.5617 aux.loss_ce: 0.0090 aux.acc_seg: 99.0385 +04/17 20:20:10 - mmengine - INFO - Iter(train) [ 42400/160000] base_lr: 7.4196e-05 lr: 2.7432e-07 eta: 1 day, 8:40:00 time: 1.0004 data_time: 0.0045 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0072 decode.acc_seg: 99.8081 aux.loss_ce: 0.0079 aux.acc_seg: 99.3773 +04/17 20:21:00 - mmengine - INFO - Iter(train) [ 42450/160000] base_lr: 7.4165e-05 lr: 2.7420e-07 eta: 1 day, 8:39:10 time: 0.9989 data_time: 0.0044 memory: 8462 loss: 0.0165 decode.loss_ce: 0.0079 decode.acc_seg: 99.6317 aux.loss_ce: 0.0085 aux.acc_seg: 98.9735 +04/17 20:21:50 - mmengine - INFO - Iter(train) [ 42500/160000] base_lr: 7.4133e-05 lr: 2.7408e-07 eta: 1 day, 8:38:20 time: 1.0010 data_time: 0.0047 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0070 decode.acc_seg: 99.6326 aux.loss_ce: 0.0080 aux.acc_seg: 98.9738 +04/17 20:22:40 - mmengine - INFO - Iter(train) [ 42550/160000] base_lr: 7.4101e-05 lr: 2.7397e-07 eta: 1 day, 8:37:30 time: 1.0008 data_time: 0.0047 memory: 8462 loss: 0.0144 decode.loss_ce: 0.0069 decode.acc_seg: 99.6315 aux.loss_ce: 0.0075 aux.acc_seg: 99.2037 +04/17 20:23:30 - mmengine - INFO - Iter(train) [ 42600/160000] base_lr: 7.4070e-05 lr: 2.7385e-07 eta: 1 day, 8:36:41 time: 1.0011 data_time: 0.0048 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0074 decode.acc_seg: 99.7305 aux.loss_ce: 0.0075 aux.acc_seg: 99.2727 +04/17 20:24:20 - mmengine - INFO - Iter(train) [ 42650/160000] base_lr: 7.4038e-05 lr: 2.7373e-07 eta: 1 day, 8:35:51 time: 1.0007 data_time: 0.0046 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0073 decode.acc_seg: 99.6355 aux.loss_ce: 0.0079 aux.acc_seg: 98.8504 +04/17 20:25:10 - mmengine - INFO - Iter(train) [ 42700/160000] base_lr: 7.4007e-05 lr: 2.7362e-07 eta: 1 day, 8:35:01 time: 1.0007 data_time: 0.0051 memory: 8462 loss: 0.0162 decode.loss_ce: 0.0077 decode.acc_seg: 99.6918 aux.loss_ce: 0.0085 aux.acc_seg: 98.7600 +04/17 20:26:00 - mmengine - INFO - Iter(train) [ 42750/160000] base_lr: 7.3975e-05 lr: 2.7350e-07 eta: 1 day, 8:34:11 time: 1.0010 data_time: 0.0048 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0085 decode.acc_seg: 99.6593 aux.loss_ce: 0.0087 aux.acc_seg: 98.9767 +04/17 20:26:50 - mmengine - INFO - Iter(train) [ 42800/160000] base_lr: 7.3944e-05 lr: 2.7339e-07 eta: 1 day, 8:33:21 time: 1.0008 data_time: 0.0045 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0077 decode.acc_seg: 99.6262 aux.loss_ce: 0.0078 aux.acc_seg: 99.1676 +04/17 20:27:40 - mmengine - INFO - Iter(train) [ 42850/160000] base_lr: 7.3912e-05 lr: 2.7327e-07 eta: 1 day, 8:32:31 time: 1.0015 data_time: 0.0049 memory: 8462 loss: 0.0182 decode.loss_ce: 0.0089 decode.acc_seg: 99.6262 aux.loss_ce: 0.0092 aux.acc_seg: 99.1257 +04/17 20:28:31 - mmengine - INFO - Iter(train) [ 42900/160000] base_lr: 7.3881e-05 lr: 2.7315e-07 eta: 1 day, 8:31:41 time: 1.0011 data_time: 0.0047 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0078 decode.acc_seg: 99.6033 aux.loss_ce: 0.0090 aux.acc_seg: 98.9317 +04/17 20:29:21 - mmengine - INFO - Iter(train) [ 42950/160000] base_lr: 7.3849e-05 lr: 2.7304e-07 eta: 1 day, 8:30:51 time: 1.0000 data_time: 0.0053 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0073 decode.acc_seg: 99.6120 aux.loss_ce: 0.0080 aux.acc_seg: 99.0517 +04/17 20:30:11 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 20:30:11 - mmengine - INFO - Iter(train) [ 43000/160000] base_lr: 7.3818e-05 lr: 2.7292e-07 eta: 1 day, 8:30:02 time: 1.0005 data_time: 0.0046 memory: 8462 loss: 0.0169 decode.loss_ce: 0.0080 decode.acc_seg: 99.7398 aux.loss_ce: 0.0089 aux.acc_seg: 99.1747 +04/17 20:31:01 - mmengine - INFO - Iter(train) [ 43050/160000] base_lr: 7.3786e-05 lr: 2.7280e-07 eta: 1 day, 8:29:12 time: 1.0021 data_time: 0.0047 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0073 decode.acc_seg: 99.6790 aux.loss_ce: 0.0084 aux.acc_seg: 98.9561 +04/17 20:31:51 - mmengine - INFO - Iter(train) [ 43100/160000] base_lr: 7.3754e-05 lr: 2.7269e-07 eta: 1 day, 8:28:22 time: 1.0011 data_time: 0.0053 memory: 8462 loss: 0.0145 decode.loss_ce: 0.0069 decode.acc_seg: 99.7549 aux.loss_ce: 0.0076 aux.acc_seg: 99.3456 +04/17 20:32:41 - mmengine - INFO - Iter(train) [ 43150/160000] base_lr: 7.3723e-05 lr: 2.7257e-07 eta: 1 day, 8:27:32 time: 1.0000 data_time: 0.0050 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0090 decode.acc_seg: 99.6460 aux.loss_ce: 0.0087 aux.acc_seg: 98.9811 +04/17 20:33:31 - mmengine - INFO - Iter(train) [ 43200/160000] base_lr: 7.3691e-05 lr: 2.7245e-07 eta: 1 day, 8:26:42 time: 1.0003 data_time: 0.0045 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0069 decode.acc_seg: 99.7410 aux.loss_ce: 0.0074 aux.acc_seg: 99.2250 +04/17 20:34:21 - mmengine - INFO - Iter(train) [ 43250/160000] base_lr: 7.3660e-05 lr: 2.7234e-07 eta: 1 day, 8:25:52 time: 1.0008 data_time: 0.0046 memory: 8462 loss: 0.0144 decode.loss_ce: 0.0069 decode.acc_seg: 99.6929 aux.loss_ce: 0.0075 aux.acc_seg: 99.1802 +04/17 20:35:11 - mmengine - INFO - Iter(train) [ 43300/160000] base_lr: 7.3628e-05 lr: 2.7222e-07 eta: 1 day, 8:25:02 time: 1.0023 data_time: 0.0052 memory: 8462 loss: 0.0175 decode.loss_ce: 0.0104 decode.acc_seg: 99.7032 aux.loss_ce: 0.0070 aux.acc_seg: 99.2306 +04/17 20:36:01 - mmengine - INFO - Iter(train) [ 43350/160000] base_lr: 7.3597e-05 lr: 2.7210e-07 eta: 1 day, 8:24:13 time: 1.0004 data_time: 0.0046 memory: 8462 loss: 0.0141 decode.loss_ce: 0.0067 decode.acc_seg: 99.7654 aux.loss_ce: 0.0074 aux.acc_seg: 99.4520 +04/17 20:36:51 - mmengine - INFO - Iter(train) [ 43400/160000] base_lr: 7.3565e-05 lr: 2.7199e-07 eta: 1 day, 8:23:23 time: 1.0011 data_time: 0.0045 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0079 decode.acc_seg: 99.6899 aux.loss_ce: 0.0078 aux.acc_seg: 99.1871 +04/17 20:37:41 - mmengine - INFO - Iter(train) [ 43450/160000] base_lr: 7.3534e-05 lr: 2.7187e-07 eta: 1 day, 8:22:33 time: 1.0005 data_time: 0.0044 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0071 decode.acc_seg: 99.6826 aux.loss_ce: 0.0076 aux.acc_seg: 98.9618 +04/17 20:38:31 - mmengine - INFO - Iter(train) [ 43500/160000] base_lr: 7.3502e-05 lr: 2.7175e-07 eta: 1 day, 8:21:43 time: 1.0016 data_time: 0.0045 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0069 decode.acc_seg: 99.7271 aux.loss_ce: 0.0078 aux.acc_seg: 99.2912 +04/17 20:39:21 - mmengine - INFO - Iter(train) [ 43550/160000] base_lr: 7.3470e-05 lr: 2.7164e-07 eta: 1 day, 8:20:53 time: 1.0016 data_time: 0.0043 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0085 decode.acc_seg: 99.5556 aux.loss_ce: 0.0092 aux.acc_seg: 98.8022 +04/17 20:40:11 - mmengine - INFO - Iter(train) [ 43600/160000] base_lr: 7.3439e-05 lr: 2.7152e-07 eta: 1 day, 8:20:03 time: 1.0012 data_time: 0.0048 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0071 decode.acc_seg: 99.7625 aux.loss_ce: 0.0079 aux.acc_seg: 99.1348 +04/17 20:41:01 - mmengine - INFO - Iter(train) [ 43650/160000] base_lr: 7.3407e-05 lr: 2.7140e-07 eta: 1 day, 8:19:13 time: 1.0017 data_time: 0.0049 memory: 8462 loss: 0.0145 decode.loss_ce: 0.0068 decode.acc_seg: 99.6870 aux.loss_ce: 0.0077 aux.acc_seg: 99.0499 +04/17 20:41:51 - mmengine - INFO - Iter(train) [ 43700/160000] base_lr: 7.3376e-05 lr: 2.7129e-07 eta: 1 day, 8:18:24 time: 1.0016 data_time: 0.0045 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0075 decode.acc_seg: 99.6363 aux.loss_ce: 0.0085 aux.acc_seg: 99.2643 +04/17 20:42:42 - mmengine - INFO - Iter(train) [ 43750/160000] base_lr: 7.3344e-05 lr: 2.7117e-07 eta: 1 day, 8:17:34 time: 1.0015 data_time: 0.0049 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0073 decode.acc_seg: 99.6826 aux.loss_ce: 0.0082 aux.acc_seg: 99.3416 +04/17 20:43:32 - mmengine - INFO - Iter(train) [ 43800/160000] base_lr: 7.3313e-05 lr: 2.7105e-07 eta: 1 day, 8:16:44 time: 1.0012 data_time: 0.0051 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0075 decode.acc_seg: 99.5754 aux.loss_ce: 0.0093 aux.acc_seg: 98.4991 +04/17 20:44:22 - mmengine - INFO - Iter(train) [ 43850/160000] base_lr: 7.3281e-05 lr: 2.7094e-07 eta: 1 day, 8:15:54 time: 1.0010 data_time: 0.0045 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0071 decode.acc_seg: 99.7011 aux.loss_ce: 0.0081 aux.acc_seg: 99.3639 +04/17 20:45:12 - mmengine - INFO - Iter(train) [ 43900/160000] base_lr: 7.3250e-05 lr: 2.7082e-07 eta: 1 day, 8:15:04 time: 1.0008 data_time: 0.0045 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0077 decode.acc_seg: 99.6813 aux.loss_ce: 0.0089 aux.acc_seg: 99.2746 +04/17 20:46:02 - mmengine - INFO - Iter(train) [ 43950/160000] base_lr: 7.3218e-05 lr: 2.7070e-07 eta: 1 day, 8:14:14 time: 1.0016 data_time: 0.0049 memory: 8462 loss: 0.0151 decode.loss_ce: 0.0073 decode.acc_seg: 99.5857 aux.loss_ce: 0.0078 aux.acc_seg: 99.0938 +04/17 20:46:52 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 20:46:52 - mmengine - INFO - Iter(train) [ 44000/160000] base_lr: 7.3187e-05 lr: 2.7059e-07 eta: 1 day, 8:13:24 time: 1.0006 data_time: 0.0047 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0079 decode.acc_seg: 99.6220 aux.loss_ce: 0.0082 aux.acc_seg: 98.9563 +04/17 20:47:42 - mmengine - INFO - Iter(train) [ 44050/160000] base_lr: 7.3155e-05 lr: 2.7047e-07 eta: 1 day, 8:12:34 time: 0.9998 data_time: 0.0048 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0075 decode.acc_seg: 99.6170 aux.loss_ce: 0.0083 aux.acc_seg: 99.1631 +04/17 20:48:32 - mmengine - INFO - Iter(train) [ 44100/160000] base_lr: 7.3123e-05 lr: 2.7035e-07 eta: 1 day, 8:11:45 time: 1.0008 data_time: 0.0046 memory: 8462 loss: 0.0146 decode.loss_ce: 0.0069 decode.acc_seg: 99.6635 aux.loss_ce: 0.0077 aux.acc_seg: 99.0484 +04/17 20:49:22 - mmengine - INFO - Iter(train) [ 44150/160000] base_lr: 7.3092e-05 lr: 2.7024e-07 eta: 1 day, 8:10:55 time: 1.0011 data_time: 0.0049 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0082 decode.acc_seg: 99.6815 aux.loss_ce: 0.0089 aux.acc_seg: 99.2743 +04/17 20:50:12 - mmengine - INFO - Iter(train) [ 44200/160000] base_lr: 7.3060e-05 lr: 2.7012e-07 eta: 1 day, 8:10:05 time: 1.0009 data_time: 0.0048 memory: 8462 loss: 0.0165 decode.loss_ce: 0.0079 decode.acc_seg: 99.7707 aux.loss_ce: 0.0086 aux.acc_seg: 99.3401 +04/17 20:51:02 - mmengine - INFO - Iter(train) [ 44250/160000] base_lr: 7.3029e-05 lr: 2.7000e-07 eta: 1 day, 8:09:15 time: 1.0007 data_time: 0.0045 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0071 decode.acc_seg: 99.6984 aux.loss_ce: 0.0081 aux.acc_seg: 99.1928 +04/17 20:51:52 - mmengine - INFO - Iter(train) [ 44300/160000] base_lr: 7.2997e-05 lr: 2.6989e-07 eta: 1 day, 8:08:25 time: 1.0013 data_time: 0.0048 memory: 8462 loss: 0.0202 decode.loss_ce: 0.0098 decode.acc_seg: 99.5548 aux.loss_ce: 0.0105 aux.acc_seg: 98.5489 +04/17 20:52:42 - mmengine - INFO - Iter(train) [ 44350/160000] base_lr: 7.2966e-05 lr: 2.6977e-07 eta: 1 day, 8:07:35 time: 1.0020 data_time: 0.0043 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0072 decode.acc_seg: 99.6668 aux.loss_ce: 0.0080 aux.acc_seg: 98.9964 +04/17 20:53:32 - mmengine - INFO - Iter(train) [ 44400/160000] base_lr: 7.2934e-05 lr: 2.6965e-07 eta: 1 day, 8:06:45 time: 1.0017 data_time: 0.0046 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0086 decode.acc_seg: 99.6059 aux.loss_ce: 0.0089 aux.acc_seg: 98.5731 +04/17 20:54:22 - mmengine - INFO - Iter(train) [ 44450/160000] base_lr: 7.2903e-05 lr: 2.6954e-07 eta: 1 day, 8:05:55 time: 1.0008 data_time: 0.0046 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0072 decode.acc_seg: 99.6885 aux.loss_ce: 0.0085 aux.acc_seg: 99.1718 +04/17 20:55:12 - mmengine - INFO - Iter(train) [ 44500/160000] base_lr: 7.2871e-05 lr: 2.6942e-07 eta: 1 day, 8:05:05 time: 1.0014 data_time: 0.0049 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0072 decode.acc_seg: 99.7606 aux.loss_ce: 0.0078 aux.acc_seg: 99.1333 +04/17 20:56:02 - mmengine - INFO - Iter(train) [ 44550/160000] base_lr: 7.2840e-05 lr: 2.6930e-07 eta: 1 day, 8:04:16 time: 1.0013 data_time: 0.0047 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0072 decode.acc_seg: 99.6046 aux.loss_ce: 0.0079 aux.acc_seg: 98.7156 +04/17 20:56:52 - mmengine - INFO - Iter(train) [ 44600/160000] base_lr: 7.2808e-05 lr: 2.6919e-07 eta: 1 day, 8:03:26 time: 1.0008 data_time: 0.0044 memory: 8462 loss: 0.0137 decode.loss_ce: 0.0063 decode.acc_seg: 99.7654 aux.loss_ce: 0.0074 aux.acc_seg: 99.2189 +04/17 20:57:43 - mmengine - INFO - Iter(train) [ 44650/160000] base_lr: 7.2776e-05 lr: 2.6907e-07 eta: 1 day, 8:02:36 time: 1.0019 data_time: 0.0046 memory: 8462 loss: 0.0176 decode.loss_ce: 0.0085 decode.acc_seg: 99.7700 aux.loss_ce: 0.0092 aux.acc_seg: 99.3835 +04/17 20:58:33 - mmengine - INFO - Iter(train) [ 44700/160000] base_lr: 7.2745e-05 lr: 2.6895e-07 eta: 1 day, 8:01:46 time: 1.0019 data_time: 0.0052 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0080 decode.acc_seg: 99.6042 aux.loss_ce: 0.0088 aux.acc_seg: 99.1199 +04/17 20:59:23 - mmengine - INFO - Iter(train) [ 44750/160000] base_lr: 7.2713e-05 lr: 2.6884e-07 eta: 1 day, 8:00:56 time: 1.0007 data_time: 0.0044 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0080 decode.acc_seg: 99.7177 aux.loss_ce: 0.0087 aux.acc_seg: 99.2779 +04/17 21:00:13 - mmengine - INFO - Iter(train) [ 44800/160000] base_lr: 7.2682e-05 lr: 2.6872e-07 eta: 1 day, 8:00:06 time: 1.0030 data_time: 0.0047 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0071 decode.acc_seg: 99.7200 aux.loss_ce: 0.0086 aux.acc_seg: 99.2632 +04/17 21:01:03 - mmengine - INFO - Iter(train) [ 44850/160000] base_lr: 7.2650e-05 lr: 2.6860e-07 eta: 1 day, 7:59:16 time: 1.0022 data_time: 0.0044 memory: 8462 loss: 0.0171 decode.loss_ce: 0.0080 decode.acc_seg: 99.7496 aux.loss_ce: 0.0091 aux.acc_seg: 99.2411 +04/17 21:01:53 - mmengine - INFO - Iter(train) [ 44900/160000] base_lr: 7.2619e-05 lr: 2.6849e-07 eta: 1 day, 7:58:27 time: 1.0002 data_time: 0.0044 memory: 8462 loss: 0.0163 decode.loss_ce: 0.0077 decode.acc_seg: 99.5272 aux.loss_ce: 0.0086 aux.acc_seg: 98.4587 +04/17 21:02:43 - mmengine - INFO - Iter(train) [ 44950/160000] base_lr: 7.2587e-05 lr: 2.6837e-07 eta: 1 day, 7:57:37 time: 1.0009 data_time: 0.0051 memory: 8462 loss: 0.0151 decode.loss_ce: 0.0070 decode.acc_seg: 99.7227 aux.loss_ce: 0.0081 aux.acc_seg: 99.2695 +04/17 21:03:33 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 21:03:33 - mmengine - INFO - Iter(train) [ 45000/160000] base_lr: 7.2556e-05 lr: 2.6825e-07 eta: 1 day, 7:56:47 time: 1.0002 data_time: 0.0048 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0069 decode.acc_seg: 99.7528 aux.loss_ce: 0.0079 aux.acc_seg: 99.4102 +04/17 21:04:23 - mmengine - INFO - Iter(train) [ 45050/160000] base_lr: 7.2524e-05 lr: 2.6814e-07 eta: 1 day, 7:55:57 time: 1.0014 data_time: 0.0044 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0076 decode.acc_seg: 99.7311 aux.loss_ce: 0.0092 aux.acc_seg: 98.9716 +04/17 21:05:13 - mmengine - INFO - Iter(train) [ 45100/160000] base_lr: 7.2493e-05 lr: 2.6802e-07 eta: 1 day, 7:55:07 time: 1.0025 data_time: 0.0046 memory: 8462 loss: 0.0144 decode.loss_ce: 0.0068 decode.acc_seg: 99.7829 aux.loss_ce: 0.0076 aux.acc_seg: 99.1774 +04/17 21:06:03 - mmengine - INFO - Iter(train) [ 45150/160000] base_lr: 7.2461e-05 lr: 2.6790e-07 eta: 1 day, 7:54:17 time: 1.0001 data_time: 0.0042 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0075 decode.acc_seg: 99.6052 aux.loss_ce: 0.0083 aux.acc_seg: 98.9243 +04/17 21:06:53 - mmengine - INFO - Iter(train) [ 45200/160000] base_lr: 7.2429e-05 lr: 2.6779e-07 eta: 1 day, 7:53:27 time: 1.0000 data_time: 0.0050 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0074 decode.acc_seg: 99.7332 aux.loss_ce: 0.0078 aux.acc_seg: 99.3267 +04/17 21:07:43 - mmengine - INFO - Iter(train) [ 45250/160000] base_lr: 7.2398e-05 lr: 2.6767e-07 eta: 1 day, 7:52:37 time: 1.0008 data_time: 0.0051 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0072 decode.acc_seg: 99.7673 aux.loss_ce: 0.0077 aux.acc_seg: 99.4791 +04/17 21:08:33 - mmengine - INFO - Iter(train) [ 45300/160000] base_lr: 7.2366e-05 lr: 2.6755e-07 eta: 1 day, 7:51:48 time: 1.0018 data_time: 0.0044 memory: 8462 loss: 0.0146 decode.loss_ce: 0.0069 decode.acc_seg: 99.7829 aux.loss_ce: 0.0077 aux.acc_seg: 99.3141 +04/17 21:09:23 - mmengine - INFO - Iter(train) [ 45350/160000] base_lr: 7.2335e-05 lr: 2.6744e-07 eta: 1 day, 7:50:58 time: 1.0025 data_time: 0.0053 memory: 8462 loss: 0.0151 decode.loss_ce: 0.0071 decode.acc_seg: 99.6622 aux.loss_ce: 0.0079 aux.acc_seg: 99.2636 +04/17 21:10:13 - mmengine - INFO - Iter(train) [ 45400/160000] base_lr: 7.2303e-05 lr: 2.6732e-07 eta: 1 day, 7:50:08 time: 0.9997 data_time: 0.0048 memory: 8462 loss: 0.0164 decode.loss_ce: 0.0081 decode.acc_seg: 99.6243 aux.loss_ce: 0.0083 aux.acc_seg: 99.2947 +04/17 21:11:03 - mmengine - INFO - Iter(train) [ 45450/160000] base_lr: 7.2272e-05 lr: 2.6720e-07 eta: 1 day, 7:49:18 time: 1.0011 data_time: 0.0048 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0079 decode.acc_seg: 99.6975 aux.loss_ce: 0.0088 aux.acc_seg: 99.0618 +04/17 21:11:54 - mmengine - INFO - Iter(train) [ 45500/160000] base_lr: 7.2240e-05 lr: 2.6709e-07 eta: 1 day, 7:48:28 time: 1.0007 data_time: 0.0045 memory: 8462 loss: 0.0151 decode.loss_ce: 0.0074 decode.acc_seg: 99.7581 aux.loss_ce: 0.0077 aux.acc_seg: 99.3103 +04/17 21:12:44 - mmengine - INFO - Iter(train) [ 45550/160000] base_lr: 7.2209e-05 lr: 2.6697e-07 eta: 1 day, 7:47:38 time: 1.0012 data_time: 0.0048 memory: 8462 loss: 0.0151 decode.loss_ce: 0.0068 decode.acc_seg: 99.7274 aux.loss_ce: 0.0083 aux.acc_seg: 99.1192 +04/17 21:13:34 - mmengine - INFO - Iter(train) [ 45600/160000] base_lr: 7.2177e-05 lr: 2.6685e-07 eta: 1 day, 7:46:48 time: 1.0006 data_time: 0.0049 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0071 decode.acc_seg: 99.7150 aux.loss_ce: 0.0081 aux.acc_seg: 99.1716 +04/17 21:14:24 - mmengine - INFO - Iter(train) [ 45650/160000] base_lr: 7.2146e-05 lr: 2.6674e-07 eta: 1 day, 7:45:59 time: 1.0009 data_time: 0.0045 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0073 decode.acc_seg: 99.6828 aux.loss_ce: 0.0081 aux.acc_seg: 99.1716 +04/17 21:15:14 - mmengine - INFO - Iter(train) [ 45700/160000] base_lr: 7.2114e-05 lr: 2.6662e-07 eta: 1 day, 7:45:09 time: 0.9997 data_time: 0.0046 memory: 8462 loss: 0.0180 decode.loss_ce: 0.0083 decode.acc_seg: 99.6675 aux.loss_ce: 0.0097 aux.acc_seg: 99.1091 +04/17 21:16:04 - mmengine - INFO - Iter(train) [ 45750/160000] base_lr: 7.2082e-05 lr: 2.6650e-07 eta: 1 day, 7:44:19 time: 1.0000 data_time: 0.0044 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0068 decode.acc_seg: 99.5943 aux.loss_ce: 0.0087 aux.acc_seg: 98.4968 +04/17 21:16:54 - mmengine - INFO - Iter(train) [ 45800/160000] base_lr: 7.2051e-05 lr: 2.6639e-07 eta: 1 day, 7:43:29 time: 1.0016 data_time: 0.0046 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0080 decode.acc_seg: 99.6555 aux.loss_ce: 0.0081 aux.acc_seg: 99.0829 +04/17 21:17:44 - mmengine - INFO - Iter(train) [ 45850/160000] base_lr: 7.2019e-05 lr: 2.6627e-07 eta: 1 day, 7:42:39 time: 1.0004 data_time: 0.0046 memory: 8462 loss: 0.0138 decode.loss_ce: 0.0063 decode.acc_seg: 99.8165 aux.loss_ce: 0.0075 aux.acc_seg: 99.5226 +04/17 21:18:34 - mmengine - INFO - Iter(train) [ 45900/160000] base_lr: 7.1988e-05 lr: 2.6615e-07 eta: 1 day, 7:41:49 time: 1.0015 data_time: 0.0044 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0071 decode.acc_seg: 99.7004 aux.loss_ce: 0.0081 aux.acc_seg: 99.2018 +04/17 21:19:24 - mmengine - INFO - Iter(train) [ 45950/160000] base_lr: 7.1956e-05 lr: 2.6604e-07 eta: 1 day, 7:40:59 time: 1.0010 data_time: 0.0056 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0065 decode.acc_seg: 99.7581 aux.loss_ce: 0.0079 aux.acc_seg: 99.2363 +04/17 21:20:14 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 21:20:14 - mmengine - INFO - Iter(train) [ 46000/160000] base_lr: 7.1925e-05 lr: 2.6592e-07 eta: 1 day, 7:40:09 time: 1.0018 data_time: 0.0045 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0077 decode.acc_seg: 99.5554 aux.loss_ce: 0.0091 aux.acc_seg: 98.8052 +04/17 21:21:04 - mmengine - INFO - Iter(train) [ 46050/160000] base_lr: 7.1893e-05 lr: 2.6580e-07 eta: 1 day, 7:39:19 time: 1.0008 data_time: 0.0043 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0082 decode.acc_seg: 99.7744 aux.loss_ce: 0.0090 aux.acc_seg: 99.3090 +04/17 21:21:54 - mmengine - INFO - Iter(train) [ 46100/160000] base_lr: 7.1862e-05 lr: 2.6569e-07 eta: 1 day, 7:38:29 time: 1.0013 data_time: 0.0048 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0068 decode.acc_seg: 99.7307 aux.loss_ce: 0.0075 aux.acc_seg: 99.2559 +04/17 21:22:44 - mmengine - INFO - Iter(train) [ 46150/160000] base_lr: 7.1830e-05 lr: 2.6557e-07 eta: 1 day, 7:37:39 time: 1.0002 data_time: 0.0046 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0075 decode.acc_seg: 99.7231 aux.loss_ce: 0.0085 aux.acc_seg: 99.2443 +04/17 21:23:34 - mmengine - INFO - Iter(train) [ 46200/160000] base_lr: 7.1799e-05 lr: 2.6545e-07 eta: 1 day, 7:36:49 time: 1.0006 data_time: 0.0046 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0068 decode.acc_seg: 99.7627 aux.loss_ce: 0.0079 aux.acc_seg: 99.3732 +04/17 21:24:24 - mmengine - INFO - Iter(train) [ 46250/160000] base_lr: 7.1767e-05 lr: 2.6534e-07 eta: 1 day, 7:35:59 time: 1.0017 data_time: 0.0050 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0076 decode.acc_seg: 99.7726 aux.loss_ce: 0.0079 aux.acc_seg: 99.2201 +04/17 21:25:14 - mmengine - INFO - Iter(train) [ 46300/160000] base_lr: 7.1735e-05 lr: 2.6522e-07 eta: 1 day, 7:35:09 time: 1.0022 data_time: 0.0050 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0071 decode.acc_seg: 99.6254 aux.loss_ce: 0.0078 aux.acc_seg: 98.9658 +04/17 21:26:04 - mmengine - INFO - Iter(train) [ 46350/160000] base_lr: 7.1704e-05 lr: 2.6510e-07 eta: 1 day, 7:34:19 time: 0.9999 data_time: 0.0048 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0073 decode.acc_seg: 99.6065 aux.loss_ce: 0.0078 aux.acc_seg: 99.1550 +04/17 21:26:54 - mmengine - INFO - Iter(train) [ 46400/160000] base_lr: 7.1672e-05 lr: 2.6499e-07 eta: 1 day, 7:33:30 time: 1.0006 data_time: 0.0045 memory: 8462 loss: 0.0136 decode.loss_ce: 0.0064 decode.acc_seg: 99.7606 aux.loss_ce: 0.0072 aux.acc_seg: 99.3614 +04/17 21:27:44 - mmengine - INFO - Iter(train) [ 46450/160000] base_lr: 7.1641e-05 lr: 2.6487e-07 eta: 1 day, 7:32:40 time: 1.0016 data_time: 0.0055 memory: 8462 loss: 0.0132 decode.loss_ce: 0.0063 decode.acc_seg: 99.7169 aux.loss_ce: 0.0070 aux.acc_seg: 99.2430 +04/17 21:28:34 - mmengine - INFO - Iter(train) [ 46500/160000] base_lr: 7.1609e-05 lr: 2.6475e-07 eta: 1 day, 7:31:50 time: 1.0007 data_time: 0.0046 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0075 decode.acc_seg: 99.7643 aux.loss_ce: 0.0082 aux.acc_seg: 99.2771 +04/17 21:29:25 - mmengine - INFO - Iter(train) [ 46550/160000] base_lr: 7.1578e-05 lr: 2.6464e-07 eta: 1 day, 7:31:00 time: 1.0010 data_time: 0.0045 memory: 8462 loss: 0.0144 decode.loss_ce: 0.0066 decode.acc_seg: 99.6943 aux.loss_ce: 0.0078 aux.acc_seg: 98.9288 +04/17 21:30:15 - mmengine - INFO - Iter(train) [ 46600/160000] base_lr: 7.1546e-05 lr: 2.6452e-07 eta: 1 day, 7:30:10 time: 1.0026 data_time: 0.0048 memory: 8462 loss: 0.0180 decode.loss_ce: 0.0082 decode.acc_seg: 99.6992 aux.loss_ce: 0.0099 aux.acc_seg: 99.0416 +04/17 21:31:05 - mmengine - INFO - Iter(train) [ 46650/160000] base_lr: 7.1515e-05 lr: 2.6440e-07 eta: 1 day, 7:29:20 time: 1.0021 data_time: 0.0048 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0070 decode.acc_seg: 99.6363 aux.loss_ce: 0.0083 aux.acc_seg: 98.8831 +04/17 21:31:55 - mmengine - INFO - Iter(train) [ 46700/160000] base_lr: 7.1483e-05 lr: 2.6429e-07 eta: 1 day, 7:28:30 time: 1.0022 data_time: 0.0044 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0074 decode.acc_seg: 99.7843 aux.loss_ce: 0.0082 aux.acc_seg: 99.4307 +04/17 21:32:45 - mmengine - INFO - Iter(train) [ 46750/160000] base_lr: 7.1452e-05 lr: 2.6417e-07 eta: 1 day, 7:27:41 time: 1.0001 data_time: 0.0046 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0071 decode.acc_seg: 99.7663 aux.loss_ce: 0.0078 aux.acc_seg: 99.2990 +04/17 21:33:35 - mmengine - INFO - Iter(train) [ 46800/160000] base_lr: 7.1420e-05 lr: 2.6405e-07 eta: 1 day, 7:26:51 time: 0.9995 data_time: 0.0044 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0071 decode.acc_seg: 99.8091 aux.loss_ce: 0.0085 aux.acc_seg: 99.3412 +04/17 21:34:25 - mmengine - INFO - Iter(train) [ 46850/160000] base_lr: 7.1388e-05 lr: 2.6394e-07 eta: 1 day, 7:26:01 time: 1.0011 data_time: 0.0044 memory: 8462 loss: 0.0139 decode.loss_ce: 0.0065 decode.acc_seg: 99.7366 aux.loss_ce: 0.0074 aux.acc_seg: 99.0967 +04/17 21:35:15 - mmengine - INFO - Iter(train) [ 46900/160000] base_lr: 7.1357e-05 lr: 2.6382e-07 eta: 1 day, 7:25:11 time: 1.0007 data_time: 0.0046 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0073 decode.acc_seg: 99.7536 aux.loss_ce: 0.0074 aux.acc_seg: 99.1590 +04/17 21:36:05 - mmengine - INFO - Iter(train) [ 46950/160000] base_lr: 7.1325e-05 lr: 2.6370e-07 eta: 1 day, 7:24:21 time: 0.9995 data_time: 0.0045 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0069 decode.acc_seg: 99.7025 aux.loss_ce: 0.0083 aux.acc_seg: 99.0747 +04/17 21:36:55 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 21:36:55 - mmengine - INFO - Iter(train) [ 47000/160000] base_lr: 7.1294e-05 lr: 2.6359e-07 eta: 1 day, 7:23:31 time: 1.0004 data_time: 0.0045 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0076 decode.acc_seg: 99.6403 aux.loss_ce: 0.0079 aux.acc_seg: 99.0517 +04/17 21:37:45 - mmengine - INFO - Iter(train) [ 47050/160000] base_lr: 7.1262e-05 lr: 2.6347e-07 eta: 1 day, 7:22:41 time: 1.0009 data_time: 0.0050 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0073 decode.acc_seg: 99.7501 aux.loss_ce: 0.0085 aux.acc_seg: 99.1697 +04/17 21:38:35 - mmengine - INFO - Iter(train) [ 47100/160000] base_lr: 7.1231e-05 lr: 2.6335e-07 eta: 1 day, 7:21:51 time: 1.0017 data_time: 0.0044 memory: 8462 loss: 0.0165 decode.loss_ce: 0.0077 decode.acc_seg: 99.6986 aux.loss_ce: 0.0087 aux.acc_seg: 99.1751 +04/17 21:39:25 - mmengine - INFO - Iter(train) [ 47150/160000] base_lr: 7.1199e-05 lr: 2.6324e-07 eta: 1 day, 7:21:01 time: 1.0005 data_time: 0.0046 memory: 8462 loss: 0.0141 decode.loss_ce: 0.0066 decode.acc_seg: 99.7280 aux.loss_ce: 0.0075 aux.acc_seg: 99.3244 +04/17 21:40:15 - mmengine - INFO - Iter(train) [ 47200/160000] base_lr: 7.1168e-05 lr: 2.6312e-07 eta: 1 day, 7:20:12 time: 1.0006 data_time: 0.0047 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0074 decode.acc_seg: 99.7494 aux.loss_ce: 0.0084 aux.acc_seg: 99.3801 +04/17 21:41:05 - mmengine - INFO - Iter(train) [ 47250/160000] base_lr: 7.1136e-05 lr: 2.6300e-07 eta: 1 day, 7:19:22 time: 1.0014 data_time: 0.0047 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0071 decode.acc_seg: 99.7694 aux.loss_ce: 0.0081 aux.acc_seg: 99.4297 +04/17 21:41:55 - mmengine - INFO - Iter(train) [ 47300/160000] base_lr: 7.1105e-05 lr: 2.6289e-07 eta: 1 day, 7:18:32 time: 1.0004 data_time: 0.0044 memory: 8462 loss: 0.0145 decode.loss_ce: 0.0067 decode.acc_seg: 99.7293 aux.loss_ce: 0.0078 aux.acc_seg: 99.2870 +04/17 21:42:45 - mmengine - INFO - Iter(train) [ 47350/160000] base_lr: 7.1073e-05 lr: 2.6277e-07 eta: 1 day, 7:17:42 time: 1.0014 data_time: 0.0047 memory: 8462 loss: 0.0159 decode.loss_ce: 0.0074 decode.acc_seg: 99.7446 aux.loss_ce: 0.0084 aux.acc_seg: 99.1817 +04/17 21:43:35 - mmengine - INFO - Iter(train) [ 47400/160000] base_lr: 7.1041e-05 lr: 2.6265e-07 eta: 1 day, 7:16:52 time: 1.0012 data_time: 0.0048 memory: 8462 loss: 0.0142 decode.loss_ce: 0.0067 decode.acc_seg: 99.7017 aux.loss_ce: 0.0076 aux.acc_seg: 98.8943 +04/17 21:44:25 - mmengine - INFO - Iter(train) [ 47450/160000] base_lr: 7.1010e-05 lr: 2.6254e-07 eta: 1 day, 7:16:02 time: 1.0012 data_time: 0.0050 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0075 decode.acc_seg: 99.7063 aux.loss_ce: 0.0084 aux.acc_seg: 98.9916 +04/17 21:45:16 - mmengine - INFO - Iter(train) [ 47500/160000] base_lr: 7.0978e-05 lr: 2.6242e-07 eta: 1 day, 7:15:12 time: 1.0005 data_time: 0.0044 memory: 8462 loss: 0.0151 decode.loss_ce: 0.0068 decode.acc_seg: 99.7053 aux.loss_ce: 0.0083 aux.acc_seg: 99.2805 +04/17 21:46:06 - mmengine - INFO - Iter(train) [ 47550/160000] base_lr: 7.0947e-05 lr: 2.6231e-07 eta: 1 day, 7:14:22 time: 0.9998 data_time: 0.0046 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0071 decode.acc_seg: 99.7715 aux.loss_ce: 0.0079 aux.acc_seg: 99.3822 +04/17 21:46:56 - mmengine - INFO - Iter(train) [ 47600/160000] base_lr: 7.0915e-05 lr: 2.6219e-07 eta: 1 day, 7:13:32 time: 0.9994 data_time: 0.0052 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0072 decode.acc_seg: 99.7133 aux.loss_ce: 0.0082 aux.acc_seg: 99.2496 +04/17 21:47:46 - mmengine - INFO - Iter(train) [ 47650/160000] base_lr: 7.0884e-05 lr: 2.6207e-07 eta: 1 day, 7:12:42 time: 1.0028 data_time: 0.0049 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0070 decode.acc_seg: 99.8072 aux.loss_ce: 0.0085 aux.acc_seg: 99.4287 +04/17 21:48:36 - mmengine - INFO - Iter(train) [ 47700/160000] base_lr: 7.0852e-05 lr: 2.6196e-07 eta: 1 day, 7:11:52 time: 1.0018 data_time: 0.0043 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0078 decode.acc_seg: 99.6267 aux.loss_ce: 0.0090 aux.acc_seg: 98.8432 +04/17 21:49:26 - mmengine - INFO - Iter(train) [ 47750/160000] base_lr: 7.0821e-05 lr: 2.6184e-07 eta: 1 day, 7:11:02 time: 1.0009 data_time: 0.0043 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0068 decode.acc_seg: 99.7084 aux.loss_ce: 0.0078 aux.acc_seg: 99.1014 +04/17 21:50:16 - mmengine - INFO - Iter(train) [ 47800/160000] base_lr: 7.0789e-05 lr: 2.6172e-07 eta: 1 day, 7:10:12 time: 1.0012 data_time: 0.0050 memory: 8462 loss: 0.0139 decode.loss_ce: 0.0065 decode.acc_seg: 99.6197 aux.loss_ce: 0.0073 aux.acc_seg: 98.8844 +04/17 21:51:06 - mmengine - INFO - Iter(train) [ 47850/160000] base_lr: 7.0758e-05 lr: 2.6161e-07 eta: 1 day, 7:09:23 time: 1.0020 data_time: 0.0044 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0068 decode.acc_seg: 99.6948 aux.loss_ce: 0.0080 aux.acc_seg: 99.1024 +04/17 21:51:56 - mmengine - INFO - Iter(train) [ 47900/160000] base_lr: 7.0726e-05 lr: 2.6149e-07 eta: 1 day, 7:08:33 time: 1.0014 data_time: 0.0049 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0075 decode.acc_seg: 99.8003 aux.loss_ce: 0.0087 aux.acc_seg: 99.3961 +04/17 21:52:46 - mmengine - INFO - Iter(train) [ 47950/160000] base_lr: 7.0694e-05 lr: 2.6137e-07 eta: 1 day, 7:07:43 time: 1.0014 data_time: 0.0045 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0074 decode.acc_seg: 99.5560 aux.loss_ce: 0.0084 aux.acc_seg: 98.7616 +04/17 21:53:36 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 21:53:36 - mmengine - INFO - Iter(train) [ 48000/160000] base_lr: 7.0663e-05 lr: 2.6126e-07 eta: 1 day, 7:06:53 time: 1.0011 data_time: 0.0052 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0074 decode.acc_seg: 99.7181 aux.loss_ce: 0.0086 aux.acc_seg: 99.3011 +04/17 21:54:26 - mmengine - INFO - Iter(train) [ 48050/160000] base_lr: 7.0631e-05 lr: 2.6114e-07 eta: 1 day, 7:06:03 time: 1.0004 data_time: 0.0050 memory: 8462 loss: 0.0159 decode.loss_ce: 0.0075 decode.acc_seg: 99.5834 aux.loss_ce: 0.0084 aux.acc_seg: 98.8811 +04/17 21:55:16 - mmengine - INFO - Iter(train) [ 48100/160000] base_lr: 7.0600e-05 lr: 2.6102e-07 eta: 1 day, 7:05:13 time: 0.9993 data_time: 0.0044 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0068 decode.acc_seg: 99.6038 aux.loss_ce: 0.0075 aux.acc_seg: 98.6641 +04/17 21:56:06 - mmengine - INFO - Iter(train) [ 48150/160000] base_lr: 7.0568e-05 lr: 2.6091e-07 eta: 1 day, 7:04:23 time: 1.0014 data_time: 0.0045 memory: 8462 loss: 0.0145 decode.loss_ce: 0.0068 decode.acc_seg: 99.7147 aux.loss_ce: 0.0077 aux.acc_seg: 99.1568 +04/17 21:56:56 - mmengine - INFO - Iter(train) [ 48200/160000] base_lr: 7.0537e-05 lr: 2.6079e-07 eta: 1 day, 7:03:33 time: 1.0011 data_time: 0.0043 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0070 decode.acc_seg: 99.7541 aux.loss_ce: 0.0073 aux.acc_seg: 99.3460 +04/17 21:57:46 - mmengine - INFO - Iter(train) [ 48250/160000] base_lr: 7.0505e-05 lr: 2.6067e-07 eta: 1 day, 7:02:44 time: 1.0001 data_time: 0.0048 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0070 decode.acc_seg: 99.6876 aux.loss_ce: 0.0080 aux.acc_seg: 98.9485 +04/17 21:58:36 - mmengine - INFO - Iter(train) [ 48300/160000] base_lr: 7.0474e-05 lr: 2.6056e-07 eta: 1 day, 7:01:54 time: 1.0016 data_time: 0.0050 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0073 decode.acc_seg: 99.6477 aux.loss_ce: 0.0083 aux.acc_seg: 98.9332 +04/17 21:59:27 - mmengine - INFO - Iter(train) [ 48350/160000] base_lr: 7.0442e-05 lr: 2.6044e-07 eta: 1 day, 7:01:04 time: 1.0017 data_time: 0.0044 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0069 decode.acc_seg: 99.7740 aux.loss_ce: 0.0072 aux.acc_seg: 99.5386 +04/17 22:00:17 - mmengine - INFO - Iter(train) [ 48400/160000] base_lr: 7.0411e-05 lr: 2.6032e-07 eta: 1 day, 7:00:14 time: 1.0016 data_time: 0.0047 memory: 8462 loss: 0.0138 decode.loss_ce: 0.0064 decode.acc_seg: 99.6397 aux.loss_ce: 0.0074 aux.acc_seg: 99.0906 +04/17 22:01:07 - mmengine - INFO - Iter(train) [ 48450/160000] base_lr: 7.0379e-05 lr: 2.6021e-07 eta: 1 day, 6:59:24 time: 1.0025 data_time: 0.0049 memory: 8462 loss: 0.0254 decode.loss_ce: 0.0140 decode.acc_seg: 99.8217 aux.loss_ce: 0.0114 aux.acc_seg: 99.3887 +04/17 22:01:57 - mmengine - INFO - Iter(train) [ 48500/160000] base_lr: 7.0347e-05 lr: 2.6009e-07 eta: 1 day, 6:58:34 time: 1.0004 data_time: 0.0046 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0083 decode.acc_seg: 99.7135 aux.loss_ce: 0.0083 aux.acc_seg: 99.0097 +04/17 22:02:47 - mmengine - INFO - Iter(train) [ 48550/160000] base_lr: 7.0316e-05 lr: 2.5997e-07 eta: 1 day, 6:57:44 time: 1.0025 data_time: 0.0048 memory: 8462 loss: 0.0165 decode.loss_ce: 0.0078 decode.acc_seg: 99.6487 aux.loss_ce: 0.0086 aux.acc_seg: 98.7770 +04/17 22:03:37 - mmengine - INFO - Iter(train) [ 48600/160000] base_lr: 7.0284e-05 lr: 2.5986e-07 eta: 1 day, 6:56:55 time: 1.0025 data_time: 0.0049 memory: 8462 loss: 0.0146 decode.loss_ce: 0.0067 decode.acc_seg: 99.7282 aux.loss_ce: 0.0078 aux.acc_seg: 99.1587 +04/17 22:04:27 - mmengine - INFO - Iter(train) [ 48650/160000] base_lr: 7.0253e-05 lr: 2.5974e-07 eta: 1 day, 6:56:05 time: 1.0011 data_time: 0.0048 memory: 8462 loss: 0.0146 decode.loss_ce: 0.0068 decode.acc_seg: 99.6964 aux.loss_ce: 0.0078 aux.acc_seg: 99.3540 +04/17 22:05:17 - mmengine - INFO - Iter(train) [ 48700/160000] base_lr: 7.0221e-05 lr: 2.5962e-07 eta: 1 day, 6:55:15 time: 1.0023 data_time: 0.0053 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0074 decode.acc_seg: 99.7150 aux.loss_ce: 0.0078 aux.acc_seg: 99.5104 +04/17 22:06:07 - mmengine - INFO - Iter(train) [ 48750/160000] base_lr: 7.0190e-05 lr: 2.5951e-07 eta: 1 day, 6:54:25 time: 1.0009 data_time: 0.0048 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0067 decode.acc_seg: 99.7252 aux.loss_ce: 0.0080 aux.acc_seg: 99.3021 +04/17 22:06:57 - mmengine - INFO - Iter(train) [ 48800/160000] base_lr: 7.0158e-05 lr: 2.5939e-07 eta: 1 day, 6:53:35 time: 1.0016 data_time: 0.0051 memory: 8462 loss: 0.0135 decode.loss_ce: 0.0063 decode.acc_seg: 99.6714 aux.loss_ce: 0.0072 aux.acc_seg: 99.1762 +04/17 22:07:47 - mmengine - INFO - Iter(train) [ 48850/160000] base_lr: 7.0127e-05 lr: 2.5927e-07 eta: 1 day, 6:52:45 time: 1.0010 data_time: 0.0051 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0073 decode.acc_seg: 99.5945 aux.loss_ce: 0.0078 aux.acc_seg: 99.1806 +04/17 22:08:37 - mmengine - INFO - Iter(train) [ 48900/160000] base_lr: 7.0095e-05 lr: 2.5916e-07 eta: 1 day, 6:51:55 time: 1.0009 data_time: 0.0044 memory: 8462 loss: 0.0121 decode.loss_ce: 0.0057 decode.acc_seg: 99.8064 aux.loss_ce: 0.0064 aux.acc_seg: 99.4448 +04/17 22:09:28 - mmengine - INFO - Iter(train) [ 48950/160000] base_lr: 7.0064e-05 lr: 2.5904e-07 eta: 1 day, 6:51:06 time: 1.0026 data_time: 0.0048 memory: 8462 loss: 0.0130 decode.loss_ce: 0.0064 decode.acc_seg: 99.6710 aux.loss_ce: 0.0066 aux.acc_seg: 99.3107 +04/17 22:10:18 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 22:10:18 - mmengine - INFO - Iter(train) [ 49000/160000] base_lr: 7.0032e-05 lr: 2.5892e-07 eta: 1 day, 6:50:16 time: 1.0002 data_time: 0.0046 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0070 decode.acc_seg: 99.7532 aux.loss_ce: 0.0077 aux.acc_seg: 99.3370 +04/17 22:11:08 - mmengine - INFO - Iter(train) [ 49050/160000] base_lr: 7.0000e-05 lr: 2.5881e-07 eta: 1 day, 6:49:26 time: 1.0013 data_time: 0.0046 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0074 decode.acc_seg: 99.5754 aux.loss_ce: 0.0087 aux.acc_seg: 98.7215 +04/17 22:11:58 - mmengine - INFO - Iter(train) [ 49100/160000] base_lr: 6.9969e-05 lr: 2.5869e-07 eta: 1 day, 6:48:36 time: 1.0005 data_time: 0.0044 memory: 8462 loss: 0.0174 decode.loss_ce: 0.0082 decode.acc_seg: 99.7231 aux.loss_ce: 0.0092 aux.acc_seg: 99.1968 +04/17 22:12:48 - mmengine - INFO - Iter(train) [ 49150/160000] base_lr: 6.9937e-05 lr: 2.5857e-07 eta: 1 day, 6:47:46 time: 1.0009 data_time: 0.0047 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0072 decode.acc_seg: 99.6813 aux.loss_ce: 0.0084 aux.acc_seg: 99.0835 +04/17 22:13:38 - mmengine - INFO - Iter(train) [ 49200/160000] base_lr: 6.9906e-05 lr: 2.5846e-07 eta: 1 day, 6:46:56 time: 1.0013 data_time: 0.0051 memory: 8462 loss: 0.0151 decode.loss_ce: 0.0067 decode.acc_seg: 99.7419 aux.loss_ce: 0.0084 aux.acc_seg: 98.9325 +04/17 22:14:28 - mmengine - INFO - Iter(train) [ 49250/160000] base_lr: 6.9874e-05 lr: 2.5834e-07 eta: 1 day, 6:46:06 time: 1.0021 data_time: 0.0046 memory: 8462 loss: 0.0144 decode.loss_ce: 0.0070 decode.acc_seg: 99.6363 aux.loss_ce: 0.0075 aux.acc_seg: 99.0494 +04/17 22:15:18 - mmengine - INFO - Iter(train) [ 49300/160000] base_lr: 6.9843e-05 lr: 2.5822e-07 eta: 1 day, 6:45:16 time: 1.0011 data_time: 0.0044 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0073 decode.acc_seg: 99.6635 aux.loss_ce: 0.0081 aux.acc_seg: 99.0404 +04/17 22:16:08 - mmengine - INFO - Iter(train) [ 49350/160000] base_lr: 6.9811e-05 lr: 2.5811e-07 eta: 1 day, 6:44:27 time: 1.0020 data_time: 0.0049 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0064 decode.acc_seg: 99.7753 aux.loss_ce: 0.0083 aux.acc_seg: 99.3837 +04/17 22:16:58 - mmengine - INFO - Iter(train) [ 49400/160000] base_lr: 6.9780e-05 lr: 2.5799e-07 eta: 1 day, 6:43:37 time: 0.9983 data_time: 0.0047 memory: 8462 loss: 0.0135 decode.loss_ce: 0.0065 decode.acc_seg: 99.7105 aux.loss_ce: 0.0070 aux.acc_seg: 99.2142 +04/17 22:17:48 - mmengine - INFO - Iter(train) [ 49450/160000] base_lr: 6.9748e-05 lr: 2.5787e-07 eta: 1 day, 6:42:47 time: 1.0017 data_time: 0.0047 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0079 decode.acc_seg: 99.7072 aux.loss_ce: 0.0089 aux.acc_seg: 99.4307 +04/17 22:18:38 - mmengine - INFO - Iter(train) [ 49500/160000] base_lr: 6.9717e-05 lr: 2.5776e-07 eta: 1 day, 6:41:57 time: 1.0010 data_time: 0.0049 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0067 decode.acc_seg: 99.6824 aux.loss_ce: 0.0076 aux.acc_seg: 98.8033 +04/17 22:19:28 - mmengine - INFO - Iter(train) [ 49550/160000] base_lr: 6.9685e-05 lr: 2.5764e-07 eta: 1 day, 6:41:07 time: 1.0011 data_time: 0.0048 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0066 decode.acc_seg: 99.7353 aux.loss_ce: 0.0082 aux.acc_seg: 99.1476 +04/17 22:20:18 - mmengine - INFO - Iter(train) [ 49600/160000] base_lr: 6.9653e-05 lr: 2.5752e-07 eta: 1 day, 6:40:17 time: 1.0010 data_time: 0.0046 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0075 decode.acc_seg: 99.7128 aux.loss_ce: 0.0082 aux.acc_seg: 99.0692 +04/17 22:21:08 - mmengine - INFO - Iter(train) [ 49650/160000] base_lr: 6.9622e-05 lr: 2.5741e-07 eta: 1 day, 6:39:27 time: 1.0017 data_time: 0.0049 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0072 decode.acc_seg: 99.7137 aux.loss_ce: 0.0080 aux.acc_seg: 99.2208 +04/17 22:21:58 - mmengine - INFO - Iter(train) [ 49700/160000] base_lr: 6.9590e-05 lr: 2.5729e-07 eta: 1 day, 6:38:37 time: 1.0017 data_time: 0.0046 memory: 8462 loss: 0.0144 decode.loss_ce: 0.0063 decode.acc_seg: 99.6155 aux.loss_ce: 0.0081 aux.acc_seg: 98.8371 +04/17 22:22:49 - mmengine - INFO - Iter(train) [ 49750/160000] base_lr: 6.9559e-05 lr: 2.5717e-07 eta: 1 day, 6:37:47 time: 1.0002 data_time: 0.0045 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0067 decode.acc_seg: 99.7698 aux.loss_ce: 0.0076 aux.acc_seg: 99.4476 +04/17 22:23:39 - mmengine - INFO - Iter(train) [ 49800/160000] base_lr: 6.9527e-05 lr: 2.5706e-07 eta: 1 day, 6:36:57 time: 1.0008 data_time: 0.0046 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0068 decode.acc_seg: 99.6786 aux.loss_ce: 0.0081 aux.acc_seg: 99.0566 +04/17 22:24:29 - mmengine - INFO - Iter(train) [ 49850/160000] base_lr: 6.9496e-05 lr: 2.5694e-07 eta: 1 day, 6:36:08 time: 1.0015 data_time: 0.0048 memory: 8462 loss: 0.0130 decode.loss_ce: 0.0061 decode.acc_seg: 99.7276 aux.loss_ce: 0.0068 aux.acc_seg: 99.3731 +04/17 22:25:19 - mmengine - INFO - Iter(train) [ 49900/160000] base_lr: 6.9464e-05 lr: 2.5682e-07 eta: 1 day, 6:35:18 time: 1.0006 data_time: 0.0047 memory: 8462 loss: 0.0164 decode.loss_ce: 0.0078 decode.acc_seg: 99.7438 aux.loss_ce: 0.0086 aux.acc_seg: 99.4097 +04/17 22:26:09 - mmengine - INFO - Iter(train) [ 49950/160000] base_lr: 6.9433e-05 lr: 2.5671e-07 eta: 1 day, 6:34:28 time: 1.0024 data_time: 0.0047 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0065 decode.acc_seg: 99.7375 aux.loss_ce: 0.0075 aux.acc_seg: 99.2811 +04/17 22:26:59 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 22:26:59 - mmengine - INFO - Iter(train) [ 50000/160000] base_lr: 6.9401e-05 lr: 2.5659e-07 eta: 1 day, 6:33:38 time: 1.0030 data_time: 0.0045 memory: 8462 loss: 0.0144 decode.loss_ce: 0.0070 decode.acc_seg: 99.6641 aux.loss_ce: 0.0074 aux.acc_seg: 99.1444 +04/17 22:26:59 - mmengine - INFO - Saving checkpoint at 50000 iterations +04/17 22:27:09 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:17 time: 0.1159 data_time: 0.0014 memory: 4004 +04/17 22:27:15 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:11 time: 0.1158 data_time: 0.0014 memory: 4004 +04/17 22:27:20 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:05 time: 0.1160 data_time: 0.0015 memory: 4004 +04/17 22:27:26 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.1161 data_time: 0.0016 memory: 4004 +04/17 22:27:26 - mmengine - INFO - per class results: +04/17 22:27:26 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 99.17 | 99.59 | 99.58 | 99.58 | 99.59 | +| contrast | 81.84 | 89.95 | 90.01 | 90.08 | 89.95 | ++------------+-------+-------+--------+-----------+--------+ +04/17 22:27:26 - mmengine - INFO - Iter(val) [200/200] aAcc: 99.2000 mIoU: 90.5100 mAcc: 94.7700 mFscore: 94.8000 mPrecision: 94.8300 mRecall: 94.7700 data_time: 0.0016 time: 0.1162 +04/17 22:28:17 - mmengine - INFO - Iter(train) [ 50050/160000] base_lr: 6.9370e-05 lr: 2.5647e-07 eta: 1 day, 6:32:49 time: 1.0014 data_time: 0.0044 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0072 decode.acc_seg: 99.6246 aux.loss_ce: 0.0081 aux.acc_seg: 99.0986 +04/17 22:29:07 - mmengine - INFO - Iter(train) [ 50100/160000] base_lr: 6.9338e-05 lr: 2.5636e-07 eta: 1 day, 6:31:59 time: 1.0000 data_time: 0.0045 memory: 8462 loss: 0.0146 decode.loss_ce: 0.0069 decode.acc_seg: 99.6515 aux.loss_ce: 0.0077 aux.acc_seg: 98.9935 +04/17 22:29:57 - mmengine - INFO - Iter(train) [ 50150/160000] base_lr: 6.9306e-05 lr: 2.5624e-07 eta: 1 day, 6:31:09 time: 1.0027 data_time: 0.0048 memory: 8462 loss: 0.0129 decode.loss_ce: 0.0060 decode.acc_seg: 99.7379 aux.loss_ce: 0.0070 aux.acc_seg: 98.9954 +04/17 22:30:47 - mmengine - INFO - Iter(train) [ 50200/160000] base_lr: 6.9275e-05 lr: 2.5612e-07 eta: 1 day, 6:30:19 time: 1.0015 data_time: 0.0045 memory: 8462 loss: 0.0169 decode.loss_ce: 0.0076 decode.acc_seg: 99.6439 aux.loss_ce: 0.0093 aux.acc_seg: 99.0238 +04/17 22:31:37 - mmengine - INFO - Iter(train) [ 50250/160000] base_lr: 6.9243e-05 lr: 2.5601e-07 eta: 1 day, 6:29:29 time: 1.0022 data_time: 0.0044 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0073 decode.acc_seg: 99.6662 aux.loss_ce: 0.0077 aux.acc_seg: 99.1777 +04/17 22:32:27 - mmengine - INFO - Iter(train) [ 50300/160000] base_lr: 6.9212e-05 lr: 2.5589e-07 eta: 1 day, 6:28:39 time: 1.0010 data_time: 0.0045 memory: 8462 loss: 0.0136 decode.loss_ce: 0.0060 decode.acc_seg: 99.7833 aux.loss_ce: 0.0076 aux.acc_seg: 99.2540 +04/17 22:33:17 - mmengine - INFO - Iter(train) [ 50350/160000] base_lr: 6.9180e-05 lr: 2.5577e-07 eta: 1 day, 6:27:50 time: 1.0016 data_time: 0.0046 memory: 8462 loss: 0.0132 decode.loss_ce: 0.0062 decode.acc_seg: 99.7509 aux.loss_ce: 0.0070 aux.acc_seg: 99.2617 +04/17 22:34:07 - mmengine - INFO - Iter(train) [ 50400/160000] base_lr: 6.9149e-05 lr: 2.5566e-07 eta: 1 day, 6:27:00 time: 1.0009 data_time: 0.0046 memory: 8462 loss: 0.0135 decode.loss_ce: 0.0064 decode.acc_seg: 99.6490 aux.loss_ce: 0.0071 aux.acc_seg: 98.7896 +04/17 22:34:57 - mmengine - INFO - Iter(train) [ 50450/160000] base_lr: 6.9117e-05 lr: 2.5554e-07 eta: 1 day, 6:26:10 time: 1.0003 data_time: 0.0046 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0071 decode.acc_seg: 99.5760 aux.loss_ce: 0.0083 aux.acc_seg: 99.0036 +04/17 22:35:47 - mmengine - INFO - Iter(train) [ 50500/160000] base_lr: 6.9086e-05 lr: 2.5542e-07 eta: 1 day, 6:25:20 time: 1.0015 data_time: 0.0043 memory: 8462 loss: 0.0136 decode.loss_ce: 0.0062 decode.acc_seg: 99.7562 aux.loss_ce: 0.0074 aux.acc_seg: 99.2680 +04/17 22:36:37 - mmengine - INFO - Iter(train) [ 50550/160000] base_lr: 6.9054e-05 lr: 2.5531e-07 eta: 1 day, 6:24:30 time: 1.0002 data_time: 0.0044 memory: 8462 loss: 0.0132 decode.loss_ce: 0.0062 decode.acc_seg: 99.6731 aux.loss_ce: 0.0070 aux.acc_seg: 99.0002 +04/17 22:37:27 - mmengine - INFO - Iter(train) [ 50600/160000] base_lr: 6.9023e-05 lr: 2.5519e-07 eta: 1 day, 6:23:40 time: 1.0015 data_time: 0.0046 memory: 8462 loss: 0.0118 decode.loss_ce: 0.0055 decode.acc_seg: 99.8014 aux.loss_ce: 0.0062 aux.acc_seg: 99.4436 +04/17 22:38:17 - mmengine - INFO - Iter(train) [ 50650/160000] base_lr: 6.8991e-05 lr: 2.5507e-07 eta: 1 day, 6:22:50 time: 1.0010 data_time: 0.0048 memory: 8462 loss: 0.0142 decode.loss_ce: 0.0068 decode.acc_seg: 99.7143 aux.loss_ce: 0.0074 aux.acc_seg: 99.1955 +04/17 22:39:07 - mmengine - INFO - Iter(train) [ 50700/160000] base_lr: 6.8959e-05 lr: 2.5496e-07 eta: 1 day, 6:22:00 time: 1.0021 data_time: 0.0047 memory: 8462 loss: 0.0128 decode.loss_ce: 0.0061 decode.acc_seg: 99.7755 aux.loss_ce: 0.0067 aux.acc_seg: 99.3412 +04/17 22:39:57 - mmengine - INFO - Iter(train) [ 50750/160000] base_lr: 6.8928e-05 lr: 2.5484e-07 eta: 1 day, 6:21:10 time: 1.0021 data_time: 0.0046 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0069 decode.acc_seg: 99.7391 aux.loss_ce: 0.0085 aux.acc_seg: 99.3994 +04/17 22:40:47 - mmengine - INFO - Iter(train) [ 50800/160000] base_lr: 6.8896e-05 lr: 2.5472e-07 eta: 1 day, 6:20:20 time: 1.0007 data_time: 0.0046 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0073 decode.acc_seg: 99.6870 aux.loss_ce: 0.0080 aux.acc_seg: 99.1617 +04/17 22:41:37 - mmengine - INFO - Iter(train) [ 50850/160000] base_lr: 6.8865e-05 lr: 2.5461e-07 eta: 1 day, 6:19:30 time: 1.0007 data_time: 0.0047 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0078 decode.acc_seg: 99.5440 aux.loss_ce: 0.0089 aux.acc_seg: 98.8024 +04/17 22:42:27 - mmengine - INFO - Iter(train) [ 50900/160000] base_lr: 6.8833e-05 lr: 2.5449e-07 eta: 1 day, 6:18:40 time: 0.9998 data_time: 0.0046 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0072 decode.acc_seg: 99.7692 aux.loss_ce: 0.0081 aux.acc_seg: 99.2157 +04/17 22:43:18 - mmengine - INFO - Iter(train) [ 50950/160000] base_lr: 6.8802e-05 lr: 2.5437e-07 eta: 1 day, 6:17:50 time: 1.0002 data_time: 0.0047 memory: 8462 loss: 0.0169 decode.loss_ce: 0.0080 decode.acc_seg: 99.6532 aux.loss_ce: 0.0088 aux.acc_seg: 98.7093 +04/17 22:44:08 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 22:44:08 - mmengine - INFO - Iter(train) [ 51000/160000] base_lr: 6.8770e-05 lr: 2.5426e-07 eta: 1 day, 6:17:00 time: 1.0009 data_time: 0.0049 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0073 decode.acc_seg: 99.6750 aux.loss_ce: 0.0079 aux.acc_seg: 98.8974 +04/17 22:44:58 - mmengine - INFO - Iter(train) [ 51050/160000] base_lr: 6.8739e-05 lr: 2.5414e-07 eta: 1 day, 6:16:10 time: 1.0019 data_time: 0.0048 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0065 decode.acc_seg: 99.6536 aux.loss_ce: 0.0078 aux.acc_seg: 99.1037 +04/17 22:45:48 - mmengine - INFO - Iter(train) [ 51100/160000] base_lr: 6.8707e-05 lr: 2.5402e-07 eta: 1 day, 6:15:20 time: 1.0011 data_time: 0.0044 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0068 decode.acc_seg: 99.7547 aux.loss_ce: 0.0089 aux.acc_seg: 99.1705 +04/17 22:46:38 - mmengine - INFO - Iter(train) [ 51150/160000] base_lr: 6.8676e-05 lr: 2.5391e-07 eta: 1 day, 6:14:30 time: 1.0000 data_time: 0.0048 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0072 decode.acc_seg: 99.5274 aux.loss_ce: 0.0080 aux.acc_seg: 99.0046 +04/17 22:47:28 - mmengine - INFO - Iter(train) [ 51200/160000] base_lr: 6.8644e-05 lr: 2.5379e-07 eta: 1 day, 6:13:40 time: 0.9996 data_time: 0.0048 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0071 decode.acc_seg: 99.5813 aux.loss_ce: 0.0081 aux.acc_seg: 99.1209 +04/17 22:48:18 - mmengine - INFO - Iter(train) [ 51250/160000] base_lr: 6.8612e-05 lr: 2.5367e-07 eta: 1 day, 6:12:50 time: 0.9982 data_time: 0.0046 memory: 8462 loss: 0.0126 decode.loss_ce: 0.0056 decode.acc_seg: 99.7341 aux.loss_ce: 0.0070 aux.acc_seg: 99.1972 +04/17 22:49:08 - mmengine - INFO - Iter(train) [ 51300/160000] base_lr: 6.8581e-05 lr: 2.5356e-07 eta: 1 day, 6:12:00 time: 0.9989 data_time: 0.0048 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0069 decode.acc_seg: 99.6454 aux.loss_ce: 0.0079 aux.acc_seg: 99.1423 +04/17 22:49:58 - mmengine - INFO - Iter(train) [ 51350/160000] base_lr: 6.8549e-05 lr: 2.5344e-07 eta: 1 day, 6:11:10 time: 0.9982 data_time: 0.0052 memory: 8462 loss: 0.0146 decode.loss_ce: 0.0063 decode.acc_seg: 99.7147 aux.loss_ce: 0.0083 aux.acc_seg: 98.9279 +04/17 22:50:47 - mmengine - INFO - Iter(train) [ 51400/160000] base_lr: 6.8518e-05 lr: 2.5332e-07 eta: 1 day, 6:10:20 time: 0.9981 data_time: 0.0043 memory: 8462 loss: 0.0168 decode.loss_ce: 0.0076 decode.acc_seg: 99.6479 aux.loss_ce: 0.0092 aux.acc_seg: 98.9697 +04/17 22:51:37 - mmengine - INFO - Iter(train) [ 51450/160000] base_lr: 6.8486e-05 lr: 2.5321e-07 eta: 1 day, 6:09:30 time: 0.9994 data_time: 0.0053 memory: 8462 loss: 0.0194 decode.loss_ce: 0.0108 decode.acc_seg: 99.7135 aux.loss_ce: 0.0086 aux.acc_seg: 99.3416 +04/17 22:52:27 - mmengine - INFO - Iter(train) [ 51500/160000] base_lr: 6.8455e-05 lr: 2.5309e-07 eta: 1 day, 6:08:39 time: 0.9988 data_time: 0.0044 memory: 8462 loss: 0.0124 decode.loss_ce: 0.0060 decode.acc_seg: 99.7625 aux.loss_ce: 0.0064 aux.acc_seg: 99.3465 +04/17 22:53:17 - mmengine - INFO - Iter(train) [ 51550/160000] base_lr: 6.8423e-05 lr: 2.5297e-07 eta: 1 day, 6:07:49 time: 0.9995 data_time: 0.0046 memory: 8462 loss: 0.0138 decode.loss_ce: 0.0062 decode.acc_seg: 99.6857 aux.loss_ce: 0.0076 aux.acc_seg: 98.9225 +04/17 22:54:07 - mmengine - INFO - Iter(train) [ 51600/160000] base_lr: 6.8392e-05 lr: 2.5286e-07 eta: 1 day, 6:06:59 time: 0.9997 data_time: 0.0045 memory: 8462 loss: 0.0159 decode.loss_ce: 0.0074 decode.acc_seg: 99.5653 aux.loss_ce: 0.0085 aux.acc_seg: 98.5556 +04/17 22:54:57 - mmengine - INFO - Iter(train) [ 51650/160000] base_lr: 6.8360e-05 lr: 2.5274e-07 eta: 1 day, 6:06:09 time: 0.9984 data_time: 0.0052 memory: 8462 loss: 0.0139 decode.loss_ce: 0.0063 decode.acc_seg: 99.7093 aux.loss_ce: 0.0076 aux.acc_seg: 99.0751 +04/17 22:55:47 - mmengine - INFO - Iter(train) [ 51700/160000] base_lr: 6.8329e-05 lr: 2.5262e-07 eta: 1 day, 6:05:19 time: 0.9996 data_time: 0.0050 memory: 8462 loss: 0.0129 decode.loss_ce: 0.0059 decode.acc_seg: 99.7099 aux.loss_ce: 0.0070 aux.acc_seg: 99.1999 +04/17 22:56:37 - mmengine - INFO - Iter(train) [ 51750/160000] base_lr: 6.8297e-05 lr: 2.5251e-07 eta: 1 day, 6:04:29 time: 0.9982 data_time: 0.0051 memory: 8462 loss: 0.0144 decode.loss_ce: 0.0068 decode.acc_seg: 99.7078 aux.loss_ce: 0.0076 aux.acc_seg: 99.0191 +04/17 22:57:27 - mmengine - INFO - Iter(train) [ 51800/160000] base_lr: 6.8265e-05 lr: 2.5239e-07 eta: 1 day, 6:03:38 time: 0.9976 data_time: 0.0046 memory: 8462 loss: 0.0138 decode.loss_ce: 0.0062 decode.acc_seg: 99.7412 aux.loss_ce: 0.0076 aux.acc_seg: 99.3313 +04/17 22:58:17 - mmengine - INFO - Iter(train) [ 51850/160000] base_lr: 6.8234e-05 lr: 2.5227e-07 eta: 1 day, 6:02:48 time: 0.9991 data_time: 0.0050 memory: 8462 loss: 0.0133 decode.loss_ce: 0.0064 decode.acc_seg: 99.7326 aux.loss_ce: 0.0068 aux.acc_seg: 99.1741 +04/17 22:59:07 - mmengine - INFO - Iter(train) [ 51900/160000] base_lr: 6.8202e-05 lr: 2.5216e-07 eta: 1 day, 6:01:58 time: 0.9984 data_time: 0.0045 memory: 8462 loss: 0.0146 decode.loss_ce: 0.0067 decode.acc_seg: 99.6784 aux.loss_ce: 0.0079 aux.acc_seg: 99.0480 +04/17 22:59:57 - mmengine - INFO - Iter(train) [ 51950/160000] base_lr: 6.8171e-05 lr: 2.5204e-07 eta: 1 day, 6:01:08 time: 0.9980 data_time: 0.0051 memory: 8462 loss: 0.0172 decode.loss_ce: 0.0078 decode.acc_seg: 99.6328 aux.loss_ce: 0.0094 aux.acc_seg: 99.1133 +04/17 23:00:47 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 23:00:47 - mmengine - INFO - Iter(train) [ 52000/160000] base_lr: 6.8139e-05 lr: 2.5192e-07 eta: 1 day, 6:00:17 time: 0.9974 data_time: 0.0045 memory: 8462 loss: 0.0141 decode.loss_ce: 0.0069 decode.acc_seg: 99.6460 aux.loss_ce: 0.0072 aux.acc_seg: 98.9784 +04/17 23:01:36 - mmengine - INFO - Iter(train) [ 52050/160000] base_lr: 6.8108e-05 lr: 2.5181e-07 eta: 1 day, 5:59:27 time: 0.9974 data_time: 0.0047 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0075 decode.acc_seg: 99.7982 aux.loss_ce: 0.0082 aux.acc_seg: 99.4713 +04/17 23:02:26 - mmengine - INFO - Iter(train) [ 52100/160000] base_lr: 6.8076e-05 lr: 2.5169e-07 eta: 1 day, 5:58:37 time: 0.9978 data_time: 0.0051 memory: 8462 loss: 0.0159 decode.loss_ce: 0.0069 decode.acc_seg: 99.6073 aux.loss_ce: 0.0089 aux.acc_seg: 98.6242 +04/17 23:03:16 - mmengine - INFO - Iter(train) [ 52150/160000] base_lr: 6.8045e-05 lr: 2.5157e-07 eta: 1 day, 5:57:47 time: 0.9961 data_time: 0.0046 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0068 decode.acc_seg: 99.7011 aux.loss_ce: 0.0083 aux.acc_seg: 98.9805 +04/17 23:04:06 - mmengine - INFO - Iter(train) [ 52200/160000] base_lr: 6.8013e-05 lr: 2.5146e-07 eta: 1 day, 5:56:56 time: 0.9984 data_time: 0.0045 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0063 decode.acc_seg: 99.8224 aux.loss_ce: 0.0077 aux.acc_seg: 99.3618 +04/17 23:04:56 - mmengine - INFO - Iter(train) [ 52250/160000] base_lr: 6.7982e-05 lr: 2.5134e-07 eta: 1 day, 5:56:06 time: 0.9972 data_time: 0.0046 memory: 8462 loss: 0.0127 decode.loss_ce: 0.0061 decode.acc_seg: 99.7196 aux.loss_ce: 0.0066 aux.acc_seg: 99.0690 +04/17 23:05:46 - mmengine - INFO - Iter(train) [ 52300/160000] base_lr: 6.7950e-05 lr: 2.5122e-07 eta: 1 day, 5:55:16 time: 0.9983 data_time: 0.0044 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0070 decode.acc_seg: 99.6262 aux.loss_ce: 0.0084 aux.acc_seg: 99.0067 +04/17 23:06:36 - mmengine - INFO - Iter(train) [ 52350/160000] base_lr: 6.7918e-05 lr: 2.5111e-07 eta: 1 day, 5:54:25 time: 0.9967 data_time: 0.0046 memory: 8462 loss: 0.0167 decode.loss_ce: 0.0079 decode.acc_seg: 99.6027 aux.loss_ce: 0.0088 aux.acc_seg: 99.0757 +04/17 23:07:26 - mmengine - INFO - Iter(train) [ 52400/160000] base_lr: 6.7887e-05 lr: 2.5099e-07 eta: 1 day, 5:53:35 time: 0.9973 data_time: 0.0043 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0071 decode.acc_seg: 99.7801 aux.loss_ce: 0.0081 aux.acc_seg: 99.2386 +04/17 23:08:16 - mmengine - INFO - Iter(train) [ 52450/160000] base_lr: 6.7855e-05 lr: 2.5088e-07 eta: 1 day, 5:52:45 time: 0.9972 data_time: 0.0045 memory: 8462 loss: 0.0166 decode.loss_ce: 0.0083 decode.acc_seg: 99.4164 aux.loss_ce: 0.0083 aux.acc_seg: 98.9697 +04/17 23:09:05 - mmengine - INFO - Iter(train) [ 52500/160000] base_lr: 6.7824e-05 lr: 2.5076e-07 eta: 1 day, 5:51:55 time: 0.9969 data_time: 0.0047 memory: 8462 loss: 0.0177 decode.loss_ce: 0.0083 decode.acc_seg: 99.6887 aux.loss_ce: 0.0095 aux.acc_seg: 98.9767 +04/17 23:09:55 - mmengine - INFO - Iter(train) [ 52550/160000] base_lr: 6.7792e-05 lr: 2.5064e-07 eta: 1 day, 5:51:04 time: 0.9958 data_time: 0.0049 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0071 decode.acc_seg: 99.8032 aux.loss_ce: 0.0078 aux.acc_seg: 99.3263 +04/17 23:10:45 - mmengine - INFO - Iter(train) [ 52600/160000] base_lr: 6.7761e-05 lr: 2.5053e-07 eta: 1 day, 5:50:14 time: 0.9976 data_time: 0.0046 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0066 decode.acc_seg: 99.6906 aux.loss_ce: 0.0077 aux.acc_seg: 99.2094 +04/17 23:11:35 - mmengine - INFO - Iter(train) [ 52650/160000] base_lr: 6.7729e-05 lr: 2.5041e-07 eta: 1 day, 5:49:24 time: 0.9961 data_time: 0.0051 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0069 decode.acc_seg: 99.7404 aux.loss_ce: 0.0078 aux.acc_seg: 99.2817 +04/17 23:12:25 - mmengine - INFO - Iter(train) [ 52700/160000] base_lr: 6.7698e-05 lr: 2.5029e-07 eta: 1 day, 5:48:33 time: 0.9960 data_time: 0.0045 memory: 8462 loss: 0.0145 decode.loss_ce: 0.0066 decode.acc_seg: 99.7202 aux.loss_ce: 0.0079 aux.acc_seg: 99.2456 +04/17 23:13:15 - mmengine - INFO - Iter(train) [ 52750/160000] base_lr: 6.7666e-05 lr: 2.5018e-07 eta: 1 day, 5:47:43 time: 0.9966 data_time: 0.0048 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0066 decode.acc_seg: 99.6519 aux.loss_ce: 0.0082 aux.acc_seg: 98.8567 +04/17 23:14:04 - mmengine - INFO - Iter(train) [ 52800/160000] base_lr: 6.7634e-05 lr: 2.5006e-07 eta: 1 day, 5:46:53 time: 0.9974 data_time: 0.0052 memory: 8462 loss: 0.0163 decode.loss_ce: 0.0079 decode.acc_seg: 99.6191 aux.loss_ce: 0.0085 aux.acc_seg: 98.9897 +04/17 23:14:54 - mmengine - INFO - Iter(train) [ 52850/160000] base_lr: 6.7603e-05 lr: 2.4994e-07 eta: 1 day, 5:46:02 time: 0.9955 data_time: 0.0047 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0070 decode.acc_seg: 99.5979 aux.loss_ce: 0.0078 aux.acc_seg: 98.8659 +04/17 23:15:44 - mmengine - INFO - Iter(train) [ 52900/160000] base_lr: 6.7571e-05 lr: 2.4983e-07 eta: 1 day, 5:45:12 time: 0.9964 data_time: 0.0047 memory: 8462 loss: 0.0141 decode.loss_ce: 0.0065 decode.acc_seg: 99.7639 aux.loss_ce: 0.0076 aux.acc_seg: 99.3019 +04/17 23:16:34 - mmengine - INFO - Iter(train) [ 52950/160000] base_lr: 6.7540e-05 lr: 2.4971e-07 eta: 1 day, 5:44:22 time: 0.9974 data_time: 0.0044 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0066 decode.acc_seg: 99.6859 aux.loss_ce: 0.0082 aux.acc_seg: 99.3515 +04/17 23:17:24 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 23:17:24 - mmengine - INFO - Iter(train) [ 53000/160000] base_lr: 6.7508e-05 lr: 2.4959e-07 eta: 1 day, 5:43:31 time: 0.9948 data_time: 0.0046 memory: 8462 loss: 0.0129 decode.loss_ce: 0.0057 decode.acc_seg: 99.8381 aux.loss_ce: 0.0072 aux.acc_seg: 99.4926 +04/17 23:18:14 - mmengine - INFO - Iter(train) [ 53050/160000] base_lr: 6.7477e-05 lr: 2.4948e-07 eta: 1 day, 5:42:41 time: 0.9959 data_time: 0.0048 memory: 8462 loss: 0.0132 decode.loss_ce: 0.0064 decode.acc_seg: 99.7179 aux.loss_ce: 0.0068 aux.acc_seg: 99.3046 +04/17 23:19:04 - mmengine - INFO - Iter(train) [ 53100/160000] base_lr: 6.7445e-05 lr: 2.4936e-07 eta: 1 day, 5:41:50 time: 0.9959 data_time: 0.0046 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0073 decode.acc_seg: 99.5453 aux.loss_ce: 0.0085 aux.acc_seg: 98.7598 +04/17 23:19:53 - mmengine - INFO - Iter(train) [ 53150/160000] base_lr: 6.7414e-05 lr: 2.4924e-07 eta: 1 day, 5:41:00 time: 0.9960 data_time: 0.0047 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0064 decode.acc_seg: 99.7801 aux.loss_ce: 0.0076 aux.acc_seg: 99.2554 +04/17 23:20:43 - mmengine - INFO - Iter(train) [ 53200/160000] base_lr: 6.7382e-05 lr: 2.4913e-07 eta: 1 day, 5:40:10 time: 0.9960 data_time: 0.0049 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0074 decode.acc_seg: 99.6466 aux.loss_ce: 0.0082 aux.acc_seg: 99.1354 +04/17 23:21:33 - mmengine - INFO - Iter(train) [ 53250/160000] base_lr: 6.7351e-05 lr: 2.4901e-07 eta: 1 day, 5:39:19 time: 0.9958 data_time: 0.0050 memory: 8462 loss: 0.0141 decode.loss_ce: 0.0064 decode.acc_seg: 99.7814 aux.loss_ce: 0.0078 aux.acc_seg: 99.4997 +04/17 23:22:23 - mmengine - INFO - Iter(train) [ 53300/160000] base_lr: 6.7319e-05 lr: 2.4889e-07 eta: 1 day, 5:38:29 time: 0.9975 data_time: 0.0049 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0070 decode.acc_seg: 99.6582 aux.loss_ce: 0.0080 aux.acc_seg: 98.8876 +04/17 23:23:13 - mmengine - INFO - Iter(train) [ 53350/160000] base_lr: 6.7287e-05 lr: 2.4878e-07 eta: 1 day, 5:37:39 time: 0.9974 data_time: 0.0055 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0074 decode.acc_seg: 99.5537 aux.loss_ce: 0.0083 aux.acc_seg: 98.9580 +04/17 23:24:02 - mmengine - INFO - Iter(train) [ 53400/160000] base_lr: 6.7256e-05 lr: 2.4866e-07 eta: 1 day, 5:36:48 time: 0.9964 data_time: 0.0046 memory: 8462 loss: 0.0135 decode.loss_ce: 0.0062 decode.acc_seg: 99.8365 aux.loss_ce: 0.0074 aux.acc_seg: 99.4684 +04/17 23:24:52 - mmengine - INFO - Iter(train) [ 53450/160000] base_lr: 6.7224e-05 lr: 2.4854e-07 eta: 1 day, 5:35:58 time: 0.9970 data_time: 0.0045 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0066 decode.acc_seg: 99.6565 aux.loss_ce: 0.0081 aux.acc_seg: 99.1913 +04/17 23:25:42 - mmengine - INFO - Iter(train) [ 53500/160000] base_lr: 6.7193e-05 lr: 2.4843e-07 eta: 1 day, 5:35:07 time: 0.9953 data_time: 0.0050 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0068 decode.acc_seg: 99.6204 aux.loss_ce: 0.0080 aux.acc_seg: 99.0913 +04/17 23:26:32 - mmengine - INFO - Iter(train) [ 53550/160000] base_lr: 6.7161e-05 lr: 2.4831e-07 eta: 1 day, 5:34:17 time: 0.9965 data_time: 0.0046 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0068 decode.acc_seg: 99.7936 aux.loss_ce: 0.0075 aux.acc_seg: 99.4209 +04/17 23:27:22 - mmengine - INFO - Iter(train) [ 53600/160000] base_lr: 6.7130e-05 lr: 2.4819e-07 eta: 1 day, 5:33:27 time: 0.9958 data_time: 0.0044 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0073 decode.acc_seg: 99.5892 aux.loss_ce: 0.0083 aux.acc_seg: 98.8543 +04/17 23:28:11 - mmengine - INFO - Iter(train) [ 53650/160000] base_lr: 6.7098e-05 lr: 2.4808e-07 eta: 1 day, 5:32:36 time: 0.9959 data_time: 0.0046 memory: 8462 loss: 0.0144 decode.loss_ce: 0.0064 decode.acc_seg: 99.7015 aux.loss_ce: 0.0081 aux.acc_seg: 99.0965 +04/17 23:29:01 - mmengine - INFO - Iter(train) [ 53700/160000] base_lr: 6.7067e-05 lr: 2.4796e-07 eta: 1 day, 5:31:46 time: 0.9967 data_time: 0.0048 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0073 decode.acc_seg: 99.7025 aux.loss_ce: 0.0083 aux.acc_seg: 99.1486 +04/17 23:29:51 - mmengine - INFO - Iter(train) [ 53750/160000] base_lr: 6.7035e-05 lr: 2.4784e-07 eta: 1 day, 5:30:55 time: 0.9963 data_time: 0.0048 memory: 8462 loss: 0.0145 decode.loss_ce: 0.0069 decode.acc_seg: 99.5285 aux.loss_ce: 0.0076 aux.acc_seg: 98.7129 +04/17 23:30:41 - mmengine - INFO - Iter(train) [ 53800/160000] base_lr: 6.7004e-05 lr: 2.4773e-07 eta: 1 day, 5:30:05 time: 0.9976 data_time: 0.0048 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0064 decode.acc_seg: 99.8020 aux.loss_ce: 0.0077 aux.acc_seg: 99.3542 +04/17 23:31:31 - mmengine - INFO - Iter(train) [ 53850/160000] base_lr: 6.6972e-05 lr: 2.4761e-07 eta: 1 day, 5:29:15 time: 0.9966 data_time: 0.0043 memory: 8462 loss: 0.0165 decode.loss_ce: 0.0079 decode.acc_seg: 99.6719 aux.loss_ce: 0.0086 aux.acc_seg: 98.8235 +04/17 23:32:21 - mmengine - INFO - Iter(train) [ 53900/160000] base_lr: 6.6940e-05 lr: 2.4749e-07 eta: 1 day, 5:28:24 time: 0.9958 data_time: 0.0046 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0069 decode.acc_seg: 99.6099 aux.loss_ce: 0.0081 aux.acc_seg: 99.0337 +04/17 23:33:10 - mmengine - INFO - Iter(train) [ 53950/160000] base_lr: 6.6909e-05 lr: 2.4738e-07 eta: 1 day, 5:27:34 time: 0.9967 data_time: 0.0045 memory: 8462 loss: 0.0132 decode.loss_ce: 0.0060 decode.acc_seg: 99.6927 aux.loss_ce: 0.0071 aux.acc_seg: 98.9550 +04/17 23:34:00 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 23:34:00 - mmengine - INFO - Iter(train) [ 54000/160000] base_lr: 6.6877e-05 lr: 2.4726e-07 eta: 1 day, 5:26:44 time: 0.9964 data_time: 0.0045 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0070 decode.acc_seg: 99.8268 aux.loss_ce: 0.0079 aux.acc_seg: 99.4104 +04/17 23:34:50 - mmengine - INFO - Iter(train) [ 54050/160000] base_lr: 6.6846e-05 lr: 2.4714e-07 eta: 1 day, 5:25:53 time: 0.9960 data_time: 0.0047 memory: 8462 loss: 0.0142 decode.loss_ce: 0.0067 decode.acc_seg: 99.7950 aux.loss_ce: 0.0075 aux.acc_seg: 99.4024 +04/17 23:35:40 - mmengine - INFO - Iter(train) [ 54100/160000] base_lr: 6.6814e-05 lr: 2.4703e-07 eta: 1 day, 5:25:03 time: 0.9975 data_time: 0.0050 memory: 8462 loss: 0.0144 decode.loss_ce: 0.0066 decode.acc_seg: 99.6904 aux.loss_ce: 0.0078 aux.acc_seg: 99.1680 +04/17 23:36:30 - mmengine - INFO - Iter(train) [ 54150/160000] base_lr: 6.6783e-05 lr: 2.4691e-07 eta: 1 day, 5:24:12 time: 0.9958 data_time: 0.0046 memory: 8462 loss: 0.0132 decode.loss_ce: 0.0060 decode.acc_seg: 99.6975 aux.loss_ce: 0.0073 aux.acc_seg: 98.7625 +04/17 23:37:20 - mmengine - INFO - Iter(train) [ 54200/160000] base_lr: 6.6751e-05 lr: 2.4679e-07 eta: 1 day, 5:23:22 time: 0.9977 data_time: 0.0044 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0068 decode.acc_seg: 99.7915 aux.loss_ce: 0.0076 aux.acc_seg: 99.3511 +04/17 23:38:09 - mmengine - INFO - Iter(train) [ 54250/160000] base_lr: 6.6720e-05 lr: 2.4668e-07 eta: 1 day, 5:22:32 time: 0.9960 data_time: 0.0049 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0067 decode.acc_seg: 99.7839 aux.loss_ce: 0.0079 aux.acc_seg: 99.3517 +04/17 23:38:59 - mmengine - INFO - Iter(train) [ 54300/160000] base_lr: 6.6688e-05 lr: 2.4656e-07 eta: 1 day, 5:21:41 time: 0.9978 data_time: 0.0054 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0066 decode.acc_seg: 99.6439 aux.loss_ce: 0.0077 aux.acc_seg: 99.0408 +04/17 23:39:49 - mmengine - INFO - Iter(train) [ 54350/160000] base_lr: 6.6657e-05 lr: 2.4644e-07 eta: 1 day, 5:20:51 time: 0.9975 data_time: 0.0049 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0072 decode.acc_seg: 99.8219 aux.loss_ce: 0.0085 aux.acc_seg: 99.3736 +04/17 23:40:39 - mmengine - INFO - Iter(train) [ 54400/160000] base_lr: 6.6625e-05 lr: 2.4633e-07 eta: 1 day, 5:20:01 time: 0.9955 data_time: 0.0046 memory: 8462 loss: 0.0118 decode.loss_ce: 0.0055 decode.acc_seg: 99.8287 aux.loss_ce: 0.0063 aux.acc_seg: 99.6265 +04/17 23:41:29 - mmengine - INFO - Iter(train) [ 54450/160000] base_lr: 6.6593e-05 lr: 2.4621e-07 eta: 1 day, 5:19:10 time: 0.9959 data_time: 0.0048 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0067 decode.acc_seg: 99.7332 aux.loss_ce: 0.0076 aux.acc_seg: 99.1829 +04/17 23:42:18 - mmengine - INFO - Iter(train) [ 54500/160000] base_lr: 6.6562e-05 lr: 2.4609e-07 eta: 1 day, 5:18:20 time: 0.9965 data_time: 0.0050 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0068 decode.acc_seg: 99.6954 aux.loss_ce: 0.0080 aux.acc_seg: 98.9834 +04/17 23:43:08 - mmengine - INFO - Iter(train) [ 54550/160000] base_lr: 6.6530e-05 lr: 2.4598e-07 eta: 1 day, 5:17:30 time: 0.9974 data_time: 0.0053 memory: 8462 loss: 0.0124 decode.loss_ce: 0.0056 decode.acc_seg: 99.7581 aux.loss_ce: 0.0068 aux.acc_seg: 98.9933 +04/17 23:43:58 - mmengine - INFO - Iter(train) [ 54600/160000] base_lr: 6.6499e-05 lr: 2.4586e-07 eta: 1 day, 5:16:39 time: 0.9973 data_time: 0.0047 memory: 8462 loss: 0.0170 decode.loss_ce: 0.0077 decode.acc_seg: 99.6363 aux.loss_ce: 0.0093 aux.acc_seg: 98.8234 +04/17 23:44:48 - mmengine - INFO - Iter(train) [ 54650/160000] base_lr: 6.6467e-05 lr: 2.4574e-07 eta: 1 day, 5:15:49 time: 0.9965 data_time: 0.0046 memory: 8462 loss: 0.0151 decode.loss_ce: 0.0070 decode.acc_seg: 99.7231 aux.loss_ce: 0.0081 aux.acc_seg: 99.3288 +04/17 23:45:38 - mmengine - INFO - Iter(train) [ 54700/160000] base_lr: 6.6436e-05 lr: 2.4563e-07 eta: 1 day, 5:14:59 time: 0.9957 data_time: 0.0045 memory: 8462 loss: 0.0185 decode.loss_ce: 0.0079 decode.acc_seg: 99.6387 aux.loss_ce: 0.0106 aux.acc_seg: 99.0807 +04/17 23:46:28 - mmengine - INFO - Iter(train) [ 54750/160000] base_lr: 6.6404e-05 lr: 2.4551e-07 eta: 1 day, 5:14:08 time: 0.9973 data_time: 0.0047 memory: 8462 loss: 0.0146 decode.loss_ce: 0.0064 decode.acc_seg: 99.6508 aux.loss_ce: 0.0082 aux.acc_seg: 98.9162 +04/17 23:47:17 - mmengine - INFO - Iter(train) [ 54800/160000] base_lr: 6.6373e-05 lr: 2.4539e-07 eta: 1 day, 5:13:18 time: 0.9961 data_time: 0.0046 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0072 decode.acc_seg: 99.6748 aux.loss_ce: 0.0078 aux.acc_seg: 98.9868 +04/17 23:48:07 - mmengine - INFO - Iter(train) [ 54850/160000] base_lr: 6.6341e-05 lr: 2.4528e-07 eta: 1 day, 5:12:28 time: 0.9965 data_time: 0.0042 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0074 decode.acc_seg: 99.6634 aux.loss_ce: 0.0082 aux.acc_seg: 99.1253 +04/17 23:48:57 - mmengine - INFO - Iter(train) [ 54900/160000] base_lr: 6.6310e-05 lr: 2.4516e-07 eta: 1 day, 5:11:37 time: 0.9962 data_time: 0.0047 memory: 8462 loss: 0.0151 decode.loss_ce: 0.0071 decode.acc_seg: 99.7988 aux.loss_ce: 0.0080 aux.acc_seg: 99.3633 +04/17 23:49:47 - mmengine - INFO - Iter(train) [ 54950/160000] base_lr: 6.6278e-05 lr: 2.4504e-07 eta: 1 day, 5:10:47 time: 0.9969 data_time: 0.0044 memory: 8462 loss: 0.0144 decode.loss_ce: 0.0067 decode.acc_seg: 99.7210 aux.loss_ce: 0.0077 aux.acc_seg: 99.2727 +04/17 23:50:37 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/17 23:50:37 - mmengine - INFO - Iter(train) [ 55000/160000] base_lr: 6.6246e-05 lr: 2.4493e-07 eta: 1 day, 5:09:57 time: 0.9945 data_time: 0.0045 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0073 decode.acc_seg: 99.7316 aux.loss_ce: 0.0083 aux.acc_seg: 98.9096 +04/17 23:51:27 - mmengine - INFO - Iter(train) [ 55050/160000] base_lr: 6.6215e-05 lr: 2.4481e-07 eta: 1 day, 5:09:06 time: 0.9956 data_time: 0.0046 memory: 8462 loss: 0.0159 decode.loss_ce: 0.0071 decode.acc_seg: 99.8264 aux.loss_ce: 0.0088 aux.acc_seg: 99.1026 +04/17 23:52:16 - mmengine - INFO - Iter(train) [ 55100/160000] base_lr: 6.6183e-05 lr: 2.4469e-07 eta: 1 day, 5:08:16 time: 0.9961 data_time: 0.0047 memory: 8462 loss: 0.0128 decode.loss_ce: 0.0058 decode.acc_seg: 99.7845 aux.loss_ce: 0.0070 aux.acc_seg: 99.3715 +04/17 23:53:06 - mmengine - INFO - Iter(train) [ 55150/160000] base_lr: 6.6152e-05 lr: 2.4458e-07 eta: 1 day, 5:07:26 time: 0.9962 data_time: 0.0046 memory: 8462 loss: 0.0132 decode.loss_ce: 0.0061 decode.acc_seg: 99.7978 aux.loss_ce: 0.0071 aux.acc_seg: 99.4160 +04/17 23:53:56 - mmengine - INFO - Iter(train) [ 55200/160000] base_lr: 6.6120e-05 lr: 2.4446e-07 eta: 1 day, 5:06:35 time: 0.9975 data_time: 0.0052 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0075 decode.acc_seg: 99.6943 aux.loss_ce: 0.0079 aux.acc_seg: 99.1989 +04/17 23:54:46 - mmengine - INFO - Iter(train) [ 55250/160000] base_lr: 6.6089e-05 lr: 2.4434e-07 eta: 1 day, 5:05:45 time: 0.9950 data_time: 0.0046 memory: 8462 loss: 0.0133 decode.loss_ce: 0.0060 decode.acc_seg: 99.7608 aux.loss_ce: 0.0073 aux.acc_seg: 99.4297 +04/17 23:55:36 - mmengine - INFO - Iter(train) [ 55300/160000] base_lr: 6.6057e-05 lr: 2.4423e-07 eta: 1 day, 5:04:55 time: 0.9966 data_time: 0.0049 memory: 8462 loss: 0.0130 decode.loss_ce: 0.0061 decode.acc_seg: 99.7843 aux.loss_ce: 0.0069 aux.acc_seg: 99.4509 +04/17 23:56:26 - mmengine - INFO - Iter(train) [ 55350/160000] base_lr: 6.6026e-05 lr: 2.4411e-07 eta: 1 day, 5:04:04 time: 0.9961 data_time: 0.0047 memory: 8462 loss: 0.0142 decode.loss_ce: 0.0062 decode.acc_seg: 99.7089 aux.loss_ce: 0.0080 aux.acc_seg: 99.0749 +04/17 23:57:15 - mmengine - INFO - Iter(train) [ 55400/160000] base_lr: 6.5994e-05 lr: 2.4399e-07 eta: 1 day, 5:03:14 time: 0.9969 data_time: 0.0048 memory: 8462 loss: 0.0116 decode.loss_ce: 0.0054 decode.acc_seg: 99.7566 aux.loss_ce: 0.0062 aux.acc_seg: 99.4513 +04/17 23:58:05 - mmengine - INFO - Iter(train) [ 55450/160000] base_lr: 6.5963e-05 lr: 2.4388e-07 eta: 1 day, 5:02:24 time: 0.9964 data_time: 0.0046 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0070 decode.acc_seg: 99.7166 aux.loss_ce: 0.0083 aux.acc_seg: 99.0023 +04/17 23:58:55 - mmengine - INFO - Iter(train) [ 55500/160000] base_lr: 6.5931e-05 lr: 2.4376e-07 eta: 1 day, 5:01:33 time: 0.9961 data_time: 0.0045 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0066 decode.acc_seg: 99.7013 aux.loss_ce: 0.0074 aux.acc_seg: 99.1121 +04/17 23:59:45 - mmengine - INFO - Iter(train) [ 55550/160000] base_lr: 6.5899e-05 lr: 2.4364e-07 eta: 1 day, 5:00:43 time: 0.9968 data_time: 0.0048 memory: 8462 loss: 0.0146 decode.loss_ce: 0.0063 decode.acc_seg: 99.6902 aux.loss_ce: 0.0083 aux.acc_seg: 98.8699 +04/18 00:00:35 - mmengine - INFO - Iter(train) [ 55600/160000] base_lr: 6.5868e-05 lr: 2.4353e-07 eta: 1 day, 4:59:53 time: 0.9970 data_time: 0.0051 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0067 decode.acc_seg: 99.6054 aux.loss_ce: 0.0081 aux.acc_seg: 98.8825 +04/18 00:01:25 - mmengine - INFO - Iter(train) [ 55650/160000] base_lr: 6.5836e-05 lr: 2.4341e-07 eta: 1 day, 4:59:02 time: 0.9979 data_time: 0.0051 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0074 decode.acc_seg: 99.7978 aux.loss_ce: 0.0081 aux.acc_seg: 99.5258 +04/18 00:02:14 - mmengine - INFO - Iter(train) [ 55700/160000] base_lr: 6.5805e-05 lr: 2.4329e-07 eta: 1 day, 4:58:12 time: 0.9967 data_time: 0.0047 memory: 8462 loss: 0.0139 decode.loss_ce: 0.0062 decode.acc_seg: 99.7448 aux.loss_ce: 0.0077 aux.acc_seg: 99.2250 +04/18 00:03:04 - mmengine - INFO - Iter(train) [ 55750/160000] base_lr: 6.5773e-05 lr: 2.4318e-07 eta: 1 day, 4:57:22 time: 0.9969 data_time: 0.0045 memory: 8462 loss: 0.0137 decode.loss_ce: 0.0064 decode.acc_seg: 99.8028 aux.loss_ce: 0.0073 aux.acc_seg: 99.2483 +04/18 00:03:54 - mmengine - INFO - Iter(train) [ 55800/160000] base_lr: 6.5742e-05 lr: 2.4306e-07 eta: 1 day, 4:56:31 time: 0.9959 data_time: 0.0045 memory: 8462 loss: 0.0142 decode.loss_ce: 0.0068 decode.acc_seg: 99.6552 aux.loss_ce: 0.0074 aux.acc_seg: 99.2012 +04/18 00:04:44 - mmengine - INFO - Iter(train) [ 55850/160000] base_lr: 6.5710e-05 lr: 2.4294e-07 eta: 1 day, 4:55:41 time: 0.9954 data_time: 0.0048 memory: 8462 loss: 0.0139 decode.loss_ce: 0.0065 decode.acc_seg: 99.7173 aux.loss_ce: 0.0074 aux.acc_seg: 98.9405 +04/18 00:05:34 - mmengine - INFO - Iter(train) [ 55900/160000] base_lr: 6.5679e-05 lr: 2.4283e-07 eta: 1 day, 4:54:51 time: 0.9964 data_time: 0.0045 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0072 decode.acc_seg: 99.7589 aux.loss_ce: 0.0082 aux.acc_seg: 99.3073 +04/18 00:06:23 - mmengine - INFO - Iter(train) [ 55950/160000] base_lr: 6.5647e-05 lr: 2.4271e-07 eta: 1 day, 4:54:00 time: 0.9960 data_time: 0.0048 memory: 8462 loss: 0.0122 decode.loss_ce: 0.0056 decode.acc_seg: 99.7185 aux.loss_ce: 0.0067 aux.acc_seg: 99.1304 +04/18 00:07:13 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/18 00:07:13 - mmengine - INFO - Iter(train) [ 56000/160000] base_lr: 6.5616e-05 lr: 2.4259e-07 eta: 1 day, 4:53:10 time: 0.9960 data_time: 0.0047 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0077 decode.acc_seg: 99.6143 aux.loss_ce: 0.0083 aux.acc_seg: 98.8697 +04/18 00:08:03 - mmengine - INFO - Iter(train) [ 56050/160000] base_lr: 6.5584e-05 lr: 2.4248e-07 eta: 1 day, 4:52:20 time: 0.9964 data_time: 0.0049 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0067 decode.acc_seg: 99.7061 aux.loss_ce: 0.0081 aux.acc_seg: 98.8853 +04/18 00:08:53 - mmengine - INFO - Iter(train) [ 56100/160000] base_lr: 6.5552e-05 lr: 2.4236e-07 eta: 1 day, 4:51:30 time: 0.9964 data_time: 0.0049 memory: 8462 loss: 0.0139 decode.loss_ce: 0.0065 decode.acc_seg: 99.6670 aux.loss_ce: 0.0074 aux.acc_seg: 98.9204 +04/18 00:09:43 - mmengine - INFO - Iter(train) [ 56150/160000] base_lr: 6.5521e-05 lr: 2.4224e-07 eta: 1 day, 4:50:39 time: 0.9961 data_time: 0.0045 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0064 decode.acc_seg: 99.7061 aux.loss_ce: 0.0076 aux.acc_seg: 99.1003 +04/18 00:10:33 - mmengine - INFO - Iter(train) [ 56200/160000] base_lr: 6.5489e-05 lr: 2.4213e-07 eta: 1 day, 4:49:49 time: 0.9963 data_time: 0.0050 memory: 8462 loss: 0.0132 decode.loss_ce: 0.0061 decode.acc_seg: 99.7324 aux.loss_ce: 0.0071 aux.acc_seg: 99.2636 +04/18 00:11:22 - mmengine - INFO - Iter(train) [ 56250/160000] base_lr: 6.5458e-05 lr: 2.4201e-07 eta: 1 day, 4:48:59 time: 0.9962 data_time: 0.0048 memory: 8462 loss: 0.0134 decode.loss_ce: 0.0063 decode.acc_seg: 99.7345 aux.loss_ce: 0.0071 aux.acc_seg: 99.4146 +04/18 00:12:12 - mmengine - INFO - Iter(train) [ 56300/160000] base_lr: 6.5426e-05 lr: 2.4189e-07 eta: 1 day, 4:48:08 time: 0.9969 data_time: 0.0045 memory: 8462 loss: 0.0138 decode.loss_ce: 0.0063 decode.acc_seg: 99.7549 aux.loss_ce: 0.0075 aux.acc_seg: 99.1474 +04/18 00:13:02 - mmengine - INFO - Iter(train) [ 56350/160000] base_lr: 6.5395e-05 lr: 2.4178e-07 eta: 1 day, 4:47:18 time: 0.9976 data_time: 0.0052 memory: 8462 loss: 0.0141 decode.loss_ce: 0.0068 decode.acc_seg: 99.6138 aux.loss_ce: 0.0073 aux.acc_seg: 99.0545 +04/18 00:13:52 - mmengine - INFO - Iter(train) [ 56400/160000] base_lr: 6.5363e-05 lr: 2.4166e-07 eta: 1 day, 4:46:28 time: 0.9971 data_time: 0.0046 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0069 decode.acc_seg: 99.6023 aux.loss_ce: 0.0079 aux.acc_seg: 98.6624 +04/18 00:14:42 - mmengine - INFO - Iter(train) [ 56450/160000] base_lr: 6.5332e-05 lr: 2.4154e-07 eta: 1 day, 4:45:37 time: 0.9973 data_time: 0.0047 memory: 8462 loss: 0.0129 decode.loss_ce: 0.0061 decode.acc_seg: 99.7465 aux.loss_ce: 0.0068 aux.acc_seg: 99.3626 +04/18 00:15:32 - mmengine - INFO - Iter(train) [ 56500/160000] base_lr: 6.5300e-05 lr: 2.4143e-07 eta: 1 day, 4:44:47 time: 0.9969 data_time: 0.0046 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0073 decode.acc_seg: 99.7009 aux.loss_ce: 0.0081 aux.acc_seg: 99.3792 +04/18 00:16:22 - mmengine - INFO - Iter(train) [ 56550/160000] base_lr: 6.5269e-05 lr: 2.4131e-07 eta: 1 day, 4:43:57 time: 0.9969 data_time: 0.0047 memory: 8462 loss: 0.0142 decode.loss_ce: 0.0063 decode.acc_seg: 99.6609 aux.loss_ce: 0.0079 aux.acc_seg: 98.9609 +04/18 00:17:11 - mmengine - INFO - Iter(train) [ 56600/160000] base_lr: 6.5237e-05 lr: 2.4119e-07 eta: 1 day, 4:43:07 time: 0.9974 data_time: 0.0051 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0072 decode.acc_seg: 99.7528 aux.loss_ce: 0.0076 aux.acc_seg: 99.4616 +04/18 00:18:01 - mmengine - INFO - Iter(train) [ 56650/160000] base_lr: 6.5205e-05 lr: 2.4108e-07 eta: 1 day, 4:42:16 time: 0.9964 data_time: 0.0048 memory: 8462 loss: 0.0137 decode.loss_ce: 0.0064 decode.acc_seg: 99.6805 aux.loss_ce: 0.0072 aux.acc_seg: 98.8457 +04/18 00:18:51 - mmengine - INFO - Iter(train) [ 56700/160000] base_lr: 6.5174e-05 lr: 2.4096e-07 eta: 1 day, 4:41:26 time: 0.9971 data_time: 0.0048 memory: 8462 loss: 0.0134 decode.loss_ce: 0.0059 decode.acc_seg: 99.6740 aux.loss_ce: 0.0075 aux.acc_seg: 99.0074 +04/18 00:19:41 - mmengine - INFO - Iter(train) [ 56750/160000] base_lr: 6.5142e-05 lr: 2.4084e-07 eta: 1 day, 4:40:36 time: 0.9969 data_time: 0.0049 memory: 8462 loss: 0.0132 decode.loss_ce: 0.0063 decode.acc_seg: 99.7746 aux.loss_ce: 0.0069 aux.acc_seg: 99.3561 +04/18 00:20:31 - mmengine - INFO - Iter(train) [ 56800/160000] base_lr: 6.5111e-05 lr: 2.4073e-07 eta: 1 day, 4:39:46 time: 0.9974 data_time: 0.0049 memory: 8462 loss: 0.0131 decode.loss_ce: 0.0060 decode.acc_seg: 99.8016 aux.loss_ce: 0.0071 aux.acc_seg: 99.4753 +04/18 00:21:21 - mmengine - INFO - Iter(train) [ 56850/160000] base_lr: 6.5079e-05 lr: 2.4061e-07 eta: 1 day, 4:38:55 time: 0.9979 data_time: 0.0047 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0068 decode.acc_seg: 99.5365 aux.loss_ce: 0.0086 aux.acc_seg: 98.4570 +04/18 00:22:10 - mmengine - INFO - Iter(train) [ 56900/160000] base_lr: 6.5048e-05 lr: 2.4049e-07 eta: 1 day, 4:38:05 time: 0.9958 data_time: 0.0048 memory: 8462 loss: 0.0144 decode.loss_ce: 0.0071 decode.acc_seg: 99.7627 aux.loss_ce: 0.0073 aux.acc_seg: 99.1726 +04/18 00:23:00 - mmengine - INFO - Iter(train) [ 56950/160000] base_lr: 6.5016e-05 lr: 2.4038e-07 eta: 1 day, 4:37:15 time: 0.9971 data_time: 0.0048 memory: 8462 loss: 0.0130 decode.loss_ce: 0.0059 decode.acc_seg: 99.7334 aux.loss_ce: 0.0071 aux.acc_seg: 99.1201 +04/18 00:23:50 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/18 00:23:50 - mmengine - INFO - Iter(train) [ 57000/160000] base_lr: 6.4985e-05 lr: 2.4026e-07 eta: 1 day, 4:36:25 time: 0.9958 data_time: 0.0053 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0069 decode.acc_seg: 99.7732 aux.loss_ce: 0.0080 aux.acc_seg: 99.2191 +04/18 00:24:40 - mmengine - INFO - Iter(train) [ 57050/160000] base_lr: 6.4953e-05 lr: 2.4014e-07 eta: 1 day, 4:35:34 time: 0.9954 data_time: 0.0048 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0063 decode.acc_seg: 99.6063 aux.loss_ce: 0.0080 aux.acc_seg: 98.9212 +04/18 00:25:30 - mmengine - INFO - Iter(train) [ 57100/160000] base_lr: 6.4922e-05 lr: 2.4003e-07 eta: 1 day, 4:34:44 time: 0.9977 data_time: 0.0054 memory: 8462 loss: 0.0133 decode.loss_ce: 0.0061 decode.acc_seg: 99.5913 aux.loss_ce: 0.0071 aux.acc_seg: 98.8617 +04/18 00:26:20 - mmengine - INFO - Iter(train) [ 57150/160000] base_lr: 6.4890e-05 lr: 2.3991e-07 eta: 1 day, 4:33:54 time: 0.9971 data_time: 0.0049 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0066 decode.acc_seg: 99.6857 aux.loss_ce: 0.0086 aux.acc_seg: 98.9967 +04/18 00:27:10 - mmengine - INFO - Iter(train) [ 57200/160000] base_lr: 6.4858e-05 lr: 2.3980e-07 eta: 1 day, 4:33:03 time: 0.9968 data_time: 0.0048 memory: 8462 loss: 0.0134 decode.loss_ce: 0.0062 decode.acc_seg: 99.7454 aux.loss_ce: 0.0072 aux.acc_seg: 99.1982 +04/18 00:27:59 - mmengine - INFO - Iter(train) [ 57250/160000] base_lr: 6.4827e-05 lr: 2.3968e-07 eta: 1 day, 4:32:13 time: 0.9970 data_time: 0.0048 memory: 8462 loss: 0.0136 decode.loss_ce: 0.0065 decode.acc_seg: 99.8083 aux.loss_ce: 0.0071 aux.acc_seg: 99.4089 +04/18 00:28:49 - mmengine - INFO - Iter(train) [ 57300/160000] base_lr: 6.4795e-05 lr: 2.3956e-07 eta: 1 day, 4:31:23 time: 0.9973 data_time: 0.0048 memory: 8462 loss: 0.0131 decode.loss_ce: 0.0059 decode.acc_seg: 99.7503 aux.loss_ce: 0.0073 aux.acc_seg: 99.3639 +04/18 00:29:39 - mmengine - INFO - Iter(train) [ 57350/160000] base_lr: 6.4764e-05 lr: 2.3945e-07 eta: 1 day, 4:30:33 time: 0.9965 data_time: 0.0047 memory: 8462 loss: 0.0142 decode.loss_ce: 0.0068 decode.acc_seg: 99.7700 aux.loss_ce: 0.0074 aux.acc_seg: 99.5041 +04/18 00:30:29 - mmengine - INFO - Iter(train) [ 57400/160000] base_lr: 6.4732e-05 lr: 2.3933e-07 eta: 1 day, 4:29:42 time: 0.9969 data_time: 0.0046 memory: 8462 loss: 0.0138 decode.loss_ce: 0.0064 decode.acc_seg: 99.7704 aux.loss_ce: 0.0074 aux.acc_seg: 99.0435 +04/18 00:31:19 - mmengine - INFO - Iter(train) [ 57450/160000] base_lr: 6.4701e-05 lr: 2.3921e-07 eta: 1 day, 4:28:52 time: 0.9972 data_time: 0.0046 memory: 8462 loss: 0.0136 decode.loss_ce: 0.0063 decode.acc_seg: 99.7816 aux.loss_ce: 0.0073 aux.acc_seg: 99.3179 +04/18 00:32:09 - mmengine - INFO - Iter(train) [ 57500/160000] base_lr: 6.4669e-05 lr: 2.3910e-07 eta: 1 day, 4:28:02 time: 0.9975 data_time: 0.0046 memory: 8462 loss: 0.0122 decode.loss_ce: 0.0058 decode.acc_seg: 99.7908 aux.loss_ce: 0.0064 aux.acc_seg: 99.5922 +04/18 00:32:59 - mmengine - INFO - Iter(train) [ 57550/160000] base_lr: 6.4638e-05 lr: 2.3898e-07 eta: 1 day, 4:27:12 time: 0.9961 data_time: 0.0048 memory: 8462 loss: 0.0146 decode.loss_ce: 0.0068 decode.acc_seg: 99.8287 aux.loss_ce: 0.0078 aux.acc_seg: 99.4864 +04/18 00:33:48 - mmengine - INFO - Iter(train) [ 57600/160000] base_lr: 6.4606e-05 lr: 2.3886e-07 eta: 1 day, 4:26:21 time: 0.9973 data_time: 0.0053 memory: 8462 loss: 0.0128 decode.loss_ce: 0.0060 decode.acc_seg: 99.7719 aux.loss_ce: 0.0067 aux.acc_seg: 99.4776 +04/18 00:34:38 - mmengine - INFO - Iter(train) [ 57650/160000] base_lr: 6.4575e-05 lr: 2.3875e-07 eta: 1 day, 4:25:31 time: 0.9974 data_time: 0.0052 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0066 decode.acc_seg: 99.6058 aux.loss_ce: 0.0084 aux.acc_seg: 98.8892 +04/18 00:35:28 - mmengine - INFO - Iter(train) [ 57700/160000] base_lr: 6.4543e-05 lr: 2.3863e-07 eta: 1 day, 4:24:41 time: 0.9967 data_time: 0.0048 memory: 8462 loss: 0.0142 decode.loss_ce: 0.0067 decode.acc_seg: 99.5640 aux.loss_ce: 0.0074 aux.acc_seg: 98.8661 +04/18 00:36:18 - mmengine - INFO - Iter(train) [ 57750/160000] base_lr: 6.4511e-05 lr: 2.3851e-07 eta: 1 day, 4:23:51 time: 0.9963 data_time: 0.0047 memory: 8462 loss: 0.0125 decode.loss_ce: 0.0060 decode.acc_seg: 99.7713 aux.loss_ce: 0.0066 aux.acc_seg: 99.2237 +04/18 00:37:08 - mmengine - INFO - Iter(train) [ 57800/160000] base_lr: 6.4480e-05 lr: 2.3840e-07 eta: 1 day, 4:23:00 time: 0.9975 data_time: 0.0051 memory: 8462 loss: 0.0121 decode.loss_ce: 0.0054 decode.acc_seg: 99.7616 aux.loss_ce: 0.0067 aux.acc_seg: 99.1951 +04/18 00:37:58 - mmengine - INFO - Iter(train) [ 57850/160000] base_lr: 6.4448e-05 lr: 2.3828e-07 eta: 1 day, 4:22:10 time: 0.9973 data_time: 0.0051 memory: 8462 loss: 0.0126 decode.loss_ce: 0.0057 decode.acc_seg: 99.7337 aux.loss_ce: 0.0069 aux.acc_seg: 99.1795 +04/18 00:38:48 - mmengine - INFO - Iter(train) [ 57900/160000] base_lr: 6.4417e-05 lr: 2.3816e-07 eta: 1 day, 4:21:20 time: 0.9978 data_time: 0.0049 memory: 8462 loss: 0.0120 decode.loss_ce: 0.0058 decode.acc_seg: 99.7416 aux.loss_ce: 0.0062 aux.acc_seg: 99.4034 +04/18 00:39:37 - mmengine - INFO - Iter(train) [ 57950/160000] base_lr: 6.4385e-05 lr: 2.3805e-07 eta: 1 day, 4:20:30 time: 0.9971 data_time: 0.0048 memory: 8462 loss: 0.0160 decode.loss_ce: 0.0074 decode.acc_seg: 99.5874 aux.loss_ce: 0.0086 aux.acc_seg: 98.7101 +04/18 00:40:27 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/18 00:40:27 - mmengine - INFO - Iter(train) [ 58000/160000] base_lr: 6.4354e-05 lr: 2.3793e-07 eta: 1 day, 4:19:40 time: 0.9966 data_time: 0.0047 memory: 8462 loss: 0.0136 decode.loss_ce: 0.0065 decode.acc_seg: 99.6943 aux.loss_ce: 0.0072 aux.acc_seg: 99.1861 +04/18 00:41:17 - mmengine - INFO - Iter(train) [ 58050/160000] base_lr: 6.4322e-05 lr: 2.3781e-07 eta: 1 day, 4:18:49 time: 0.9964 data_time: 0.0048 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0065 decode.acc_seg: 99.6979 aux.loss_ce: 0.0074 aux.acc_seg: 98.9344 +04/18 00:42:07 - mmengine - INFO - Iter(train) [ 58100/160000] base_lr: 6.4291e-05 lr: 2.3770e-07 eta: 1 day, 4:17:59 time: 0.9974 data_time: 0.0054 memory: 8462 loss: 0.0146 decode.loss_ce: 0.0068 decode.acc_seg: 99.7322 aux.loss_ce: 0.0078 aux.acc_seg: 99.4568 +04/18 00:42:57 - mmengine - INFO - Iter(train) [ 58150/160000] base_lr: 6.4259e-05 lr: 2.3758e-07 eta: 1 day, 4:17:09 time: 0.9968 data_time: 0.0051 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0064 decode.acc_seg: 99.7648 aux.loss_ce: 0.0076 aux.acc_seg: 99.4530 +04/18 00:43:47 - mmengine - INFO - Iter(train) [ 58200/160000] base_lr: 6.4228e-05 lr: 2.3746e-07 eta: 1 day, 4:16:18 time: 0.9964 data_time: 0.0047 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0064 decode.acc_seg: 99.7704 aux.loss_ce: 0.0083 aux.acc_seg: 99.3893 +04/18 00:44:36 - mmengine - INFO - Iter(train) [ 58250/160000] base_lr: 6.4196e-05 lr: 2.3735e-07 eta: 1 day, 4:15:28 time: 0.9988 data_time: 0.0049 memory: 8462 loss: 0.0142 decode.loss_ce: 0.0067 decode.acc_seg: 99.6920 aux.loss_ce: 0.0075 aux.acc_seg: 99.1848 +04/18 00:45:26 - mmengine - INFO - Iter(train) [ 58300/160000] base_lr: 6.4164e-05 lr: 2.3723e-07 eta: 1 day, 4:14:38 time: 0.9967 data_time: 0.0046 memory: 8462 loss: 0.0137 decode.loss_ce: 0.0063 decode.acc_seg: 99.8030 aux.loss_ce: 0.0074 aux.acc_seg: 99.2216 +04/18 00:46:16 - mmengine - INFO - Iter(train) [ 58350/160000] base_lr: 6.4133e-05 lr: 2.3711e-07 eta: 1 day, 4:13:48 time: 0.9956 data_time: 0.0046 memory: 8462 loss: 0.0154 decode.loss_ce: 0.0070 decode.acc_seg: 99.6845 aux.loss_ce: 0.0085 aux.acc_seg: 99.1455 +04/18 00:47:06 - mmengine - INFO - Iter(train) [ 58400/160000] base_lr: 6.4101e-05 lr: 2.3700e-07 eta: 1 day, 4:12:58 time: 0.9963 data_time: 0.0046 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0063 decode.acc_seg: 99.7093 aux.loss_ce: 0.0077 aux.acc_seg: 99.0812 +04/18 00:47:56 - mmengine - INFO - Iter(train) [ 58450/160000] base_lr: 6.4070e-05 lr: 2.3688e-07 eta: 1 day, 4:12:07 time: 0.9964 data_time: 0.0047 memory: 8462 loss: 0.0138 decode.loss_ce: 0.0062 decode.acc_seg: 99.6807 aux.loss_ce: 0.0076 aux.acc_seg: 98.7923 +04/18 00:48:46 - mmengine - INFO - Iter(train) [ 58500/160000] base_lr: 6.4038e-05 lr: 2.3676e-07 eta: 1 day, 4:11:17 time: 0.9952 data_time: 0.0047 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0068 decode.acc_seg: 99.6386 aux.loss_ce: 0.0079 aux.acc_seg: 99.0294 +04/18 00:49:36 - mmengine - INFO - Iter(train) [ 58550/160000] base_lr: 6.4007e-05 lr: 2.3665e-07 eta: 1 day, 4:10:27 time: 0.9960 data_time: 0.0053 memory: 8462 loss: 0.0128 decode.loss_ce: 0.0059 decode.acc_seg: 99.6319 aux.loss_ce: 0.0069 aux.acc_seg: 99.0234 +04/18 00:50:25 - mmengine - INFO - Iter(train) [ 58600/160000] base_lr: 6.3975e-05 lr: 2.3653e-07 eta: 1 day, 4:09:37 time: 0.9982 data_time: 0.0047 memory: 8462 loss: 0.0137 decode.loss_ce: 0.0064 decode.acc_seg: 99.7671 aux.loss_ce: 0.0074 aux.acc_seg: 99.1859 +04/18 00:51:15 - mmengine - INFO - Iter(train) [ 58650/160000] base_lr: 6.3944e-05 lr: 2.3641e-07 eta: 1 day, 4:08:46 time: 0.9959 data_time: 0.0049 memory: 8462 loss: 0.0134 decode.loss_ce: 0.0063 decode.acc_seg: 99.7385 aux.loss_ce: 0.0072 aux.acc_seg: 98.9573 +04/18 00:52:05 - mmengine - INFO - Iter(train) [ 58700/160000] base_lr: 6.3912e-05 lr: 2.3630e-07 eta: 1 day, 4:07:56 time: 0.9980 data_time: 0.0051 memory: 8462 loss: 0.0137 decode.loss_ce: 0.0065 decode.acc_seg: 99.7580 aux.loss_ce: 0.0073 aux.acc_seg: 99.1995 +04/18 00:52:55 - mmengine - INFO - Iter(train) [ 58750/160000] base_lr: 6.3881e-05 lr: 2.3618e-07 eta: 1 day, 4:07:06 time: 0.9972 data_time: 0.0050 memory: 8462 loss: 0.0139 decode.loss_ce: 0.0063 decode.acc_seg: 99.6897 aux.loss_ce: 0.0076 aux.acc_seg: 99.0902 +04/18 00:53:45 - mmengine - INFO - Iter(train) [ 58800/160000] base_lr: 6.3849e-05 lr: 2.3606e-07 eta: 1 day, 4:06:16 time: 0.9973 data_time: 0.0050 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0063 decode.acc_seg: 99.7572 aux.loss_ce: 0.0080 aux.acc_seg: 99.1959 +04/18 00:54:35 - mmengine - INFO - Iter(train) [ 58850/160000] base_lr: 6.3817e-05 lr: 2.3595e-07 eta: 1 day, 4:05:25 time: 0.9970 data_time: 0.0052 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0069 decode.acc_seg: 99.7452 aux.loss_ce: 0.0083 aux.acc_seg: 98.9761 +04/18 00:55:25 - mmengine - INFO - Iter(train) [ 58900/160000] base_lr: 6.3786e-05 lr: 2.3583e-07 eta: 1 day, 4:04:35 time: 0.9975 data_time: 0.0047 memory: 8462 loss: 0.0141 decode.loss_ce: 0.0065 decode.acc_seg: 99.7129 aux.loss_ce: 0.0075 aux.acc_seg: 99.3151 +04/18 00:56:14 - mmengine - INFO - Iter(train) [ 58950/160000] base_lr: 6.3754e-05 lr: 2.3571e-07 eta: 1 day, 4:03:45 time: 0.9973 data_time: 0.0049 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0064 decode.acc_seg: 99.7519 aux.loss_ce: 0.0076 aux.acc_seg: 99.1943 +04/18 00:57:04 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/18 00:57:04 - mmengine - INFO - Iter(train) [ 59000/160000] base_lr: 6.3723e-05 lr: 2.3560e-07 eta: 1 day, 4:02:55 time: 0.9972 data_time: 0.0049 memory: 8462 loss: 0.0127 decode.loss_ce: 0.0061 decode.acc_seg: 99.8060 aux.loss_ce: 0.0066 aux.acc_seg: 99.4778 +04/18 00:57:54 - mmengine - INFO - Iter(train) [ 59050/160000] base_lr: 6.3691e-05 lr: 2.3548e-07 eta: 1 day, 4:02:05 time: 0.9980 data_time: 0.0049 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0076 decode.acc_seg: 99.6397 aux.loss_ce: 0.0081 aux.acc_seg: 98.9975 +04/18 00:58:44 - mmengine - INFO - Iter(train) [ 59100/160000] base_lr: 6.3660e-05 lr: 2.3536e-07 eta: 1 day, 4:01:14 time: 0.9956 data_time: 0.0049 memory: 8462 loss: 0.0141 decode.loss_ce: 0.0064 decode.acc_seg: 99.6559 aux.loss_ce: 0.0077 aux.acc_seg: 99.0963 +04/18 00:59:34 - mmengine - INFO - Iter(train) [ 59150/160000] base_lr: 6.3628e-05 lr: 2.3525e-07 eta: 1 day, 4:00:24 time: 0.9964 data_time: 0.0047 memory: 8462 loss: 0.0141 decode.loss_ce: 0.0066 decode.acc_seg: 99.6416 aux.loss_ce: 0.0075 aux.acc_seg: 99.0419 +04/18 01:00:24 - mmengine - INFO - Iter(train) [ 59200/160000] base_lr: 6.3597e-05 lr: 2.3513e-07 eta: 1 day, 3:59:34 time: 0.9968 data_time: 0.0047 memory: 8462 loss: 0.0128 decode.loss_ce: 0.0058 decode.acc_seg: 99.7854 aux.loss_ce: 0.0070 aux.acc_seg: 99.3044 +04/18 01:01:13 - mmengine - INFO - Iter(train) [ 59250/160000] base_lr: 6.3565e-05 lr: 2.3501e-07 eta: 1 day, 3:58:44 time: 0.9961 data_time: 0.0047 memory: 8462 loss: 0.0144 decode.loss_ce: 0.0067 decode.acc_seg: 99.6817 aux.loss_ce: 0.0077 aux.acc_seg: 99.2182 +04/18 01:02:03 - mmengine - INFO - Iter(train) [ 59300/160000] base_lr: 6.3534e-05 lr: 2.3490e-07 eta: 1 day, 3:57:53 time: 0.9978 data_time: 0.0051 memory: 8462 loss: 0.0131 decode.loss_ce: 0.0059 decode.acc_seg: 99.7379 aux.loss_ce: 0.0072 aux.acc_seg: 99.3759 +04/18 01:02:53 - mmengine - INFO - Iter(train) [ 59350/160000] base_lr: 6.3502e-05 lr: 2.3478e-07 eta: 1 day, 3:57:03 time: 0.9962 data_time: 0.0050 memory: 8462 loss: 0.0156 decode.loss_ce: 0.0074 decode.acc_seg: 99.6422 aux.loss_ce: 0.0082 aux.acc_seg: 99.2203 +04/18 01:03:43 - mmengine - INFO - Iter(train) [ 59400/160000] base_lr: 6.3470e-05 lr: 2.3466e-07 eta: 1 day, 3:56:13 time: 0.9984 data_time: 0.0052 memory: 8462 loss: 0.0129 decode.loss_ce: 0.0060 decode.acc_seg: 99.7980 aux.loss_ce: 0.0069 aux.acc_seg: 99.4974 +04/18 01:04:33 - mmengine - INFO - Iter(train) [ 59450/160000] base_lr: 6.3439e-05 lr: 2.3455e-07 eta: 1 day, 3:55:23 time: 0.9968 data_time: 0.0049 memory: 8462 loss: 0.0163 decode.loss_ce: 0.0075 decode.acc_seg: 99.6365 aux.loss_ce: 0.0089 aux.acc_seg: 98.9510 +04/18 01:05:23 - mmengine - INFO - Iter(train) [ 59500/160000] base_lr: 6.3407e-05 lr: 2.3443e-07 eta: 1 day, 3:54:32 time: 0.9970 data_time: 0.0046 memory: 8462 loss: 0.0135 decode.loss_ce: 0.0066 decode.acc_seg: 99.7969 aux.loss_ce: 0.0070 aux.acc_seg: 99.3740 +04/18 01:06:13 - mmengine - INFO - Iter(train) [ 59550/160000] base_lr: 6.3376e-05 lr: 2.3431e-07 eta: 1 day, 3:53:42 time: 0.9966 data_time: 0.0047 memory: 8462 loss: 0.0134 decode.loss_ce: 0.0061 decode.acc_seg: 99.7887 aux.loss_ce: 0.0072 aux.acc_seg: 99.5459 +04/18 01:07:02 - mmengine - INFO - Iter(train) [ 59600/160000] base_lr: 6.3344e-05 lr: 2.3420e-07 eta: 1 day, 3:52:52 time: 0.9981 data_time: 0.0051 memory: 8462 loss: 0.0139 decode.loss_ce: 0.0063 decode.acc_seg: 99.6906 aux.loss_ce: 0.0076 aux.acc_seg: 99.1089 +04/18 01:07:52 - mmengine - INFO - Iter(train) [ 59650/160000] base_lr: 6.3313e-05 lr: 2.3408e-07 eta: 1 day, 3:52:02 time: 0.9966 data_time: 0.0046 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0070 decode.acc_seg: 99.7128 aux.loss_ce: 0.0083 aux.acc_seg: 99.0110 +04/18 01:08:42 - mmengine - INFO - Iter(train) [ 59700/160000] base_lr: 6.3281e-05 lr: 2.3396e-07 eta: 1 day, 3:51:11 time: 0.9968 data_time: 0.0045 memory: 8462 loss: 0.0130 decode.loss_ce: 0.0063 decode.acc_seg: 99.7534 aux.loss_ce: 0.0067 aux.acc_seg: 99.2552 +04/18 01:09:32 - mmengine - INFO - Iter(train) [ 59750/160000] base_lr: 6.3250e-05 lr: 2.3385e-07 eta: 1 day, 3:50:21 time: 0.9968 data_time: 0.0047 memory: 8462 loss: 0.0137 decode.loss_ce: 0.0062 decode.acc_seg: 99.6611 aux.loss_ce: 0.0074 aux.acc_seg: 99.0505 +04/18 01:10:22 - mmengine - INFO - Iter(train) [ 59800/160000] base_lr: 6.3218e-05 lr: 2.3373e-07 eta: 1 day, 3:49:31 time: 0.9975 data_time: 0.0050 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0080 decode.acc_seg: 99.4905 aux.loss_ce: 0.0075 aux.acc_seg: 99.0028 +04/18 01:11:12 - mmengine - INFO - Iter(train) [ 59850/160000] base_lr: 6.3187e-05 lr: 2.3361e-07 eta: 1 day, 3:48:41 time: 0.9974 data_time: 0.0049 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0066 decode.acc_seg: 99.7030 aux.loss_ce: 0.0082 aux.acc_seg: 99.0513 +04/18 01:12:01 - mmengine - INFO - Iter(train) [ 59900/160000] base_lr: 6.3155e-05 lr: 2.3350e-07 eta: 1 day, 3:47:51 time: 0.9963 data_time: 0.0049 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0071 decode.acc_seg: 99.7084 aux.loss_ce: 0.0082 aux.acc_seg: 99.0517 +04/18 01:12:51 - mmengine - INFO - Iter(train) [ 59950/160000] base_lr: 6.3123e-05 lr: 2.3338e-07 eta: 1 day, 3:47:00 time: 0.9974 data_time: 0.0048 memory: 8462 loss: 0.0132 decode.loss_ce: 0.0061 decode.acc_seg: 99.7242 aux.loss_ce: 0.0070 aux.acc_seg: 99.3149 +04/18 01:13:41 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/18 01:13:41 - mmengine - INFO - Iter(train) [ 60000/160000] base_lr: 6.3092e-05 lr: 2.3326e-07 eta: 1 day, 3:46:10 time: 0.9965 data_time: 0.0052 memory: 8462 loss: 0.0152 decode.loss_ce: 0.0069 decode.acc_seg: 99.7009 aux.loss_ce: 0.0083 aux.acc_seg: 99.0499 +04/18 01:13:41 - mmengine - INFO - Saving checkpoint at 60000 iterations +04/18 01:13:51 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:17 time: 0.1160 data_time: 0.0016 memory: 4004 +04/18 01:13:57 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:11 time: 0.1156 data_time: 0.0015 memory: 4004 +04/18 01:14:03 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:05 time: 0.1155 data_time: 0.0015 memory: 4004 +04/18 01:14:08 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.1156 data_time: 0.0015 memory: 4004 +04/18 01:14:09 - mmengine - INFO - per class results: +04/18 01:14:09 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 99.17 | 99.59 | 99.59 | 99.58 | 99.59 | +| contrast | 81.83 | 89.83 | 90.01 | 90.18 | 89.83 | ++------------+-------+-------+--------+-----------+--------+ +04/18 01:14:09 - mmengine - INFO - Iter(val) [200/200] aAcc: 99.2000 mIoU: 90.5000 mAcc: 94.7100 mFscore: 94.8000 mPrecision: 94.8800 mRecall: 94.7100 data_time: 0.0019 time: 0.1161 +04/18 01:14:58 - mmengine - INFO - Iter(train) [ 60050/160000] base_lr: 6.3060e-05 lr: 2.3315e-07 eta: 1 day, 3:45:20 time: 0.9959 data_time: 0.0046 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0066 decode.acc_seg: 99.7782 aux.loss_ce: 0.0074 aux.acc_seg: 99.3649 +04/18 01:15:48 - mmengine - INFO - Iter(train) [ 60100/160000] base_lr: 6.3029e-05 lr: 2.3303e-07 eta: 1 day, 3:44:30 time: 0.9967 data_time: 0.0047 memory: 8462 loss: 0.0173 decode.loss_ce: 0.0083 decode.acc_seg: 99.3910 aux.loss_ce: 0.0090 aux.acc_seg: 98.7228 +04/18 01:16:38 - mmengine - INFO - Iter(train) [ 60150/160000] base_lr: 6.2997e-05 lr: 2.3291e-07 eta: 1 day, 3:43:40 time: 0.9972 data_time: 0.0052 memory: 8462 loss: 0.0124 decode.loss_ce: 0.0055 decode.acc_seg: 99.7696 aux.loss_ce: 0.0069 aux.acc_seg: 99.1947 +04/18 01:17:28 - mmengine - INFO - Iter(train) [ 60200/160000] base_lr: 6.2966e-05 lr: 2.3280e-07 eta: 1 day, 3:42:49 time: 0.9970 data_time: 0.0050 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0062 decode.acc_seg: 99.7618 aux.loss_ce: 0.0085 aux.acc_seg: 99.2031 +04/18 01:18:18 - mmengine - INFO - Iter(train) [ 60250/160000] base_lr: 6.2934e-05 lr: 2.3268e-07 eta: 1 day, 3:41:59 time: 0.9972 data_time: 0.0050 memory: 8462 loss: 0.0144 decode.loss_ce: 0.0065 decode.acc_seg: 99.6517 aux.loss_ce: 0.0079 aux.acc_seg: 99.0055 +04/18 01:19:08 - mmengine - INFO - Iter(train) [ 60300/160000] base_lr: 6.2903e-05 lr: 2.3256e-07 eta: 1 day, 3:41:09 time: 0.9966 data_time: 0.0046 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0066 decode.acc_seg: 99.7400 aux.loss_ce: 0.0082 aux.acc_seg: 99.1520 +04/18 01:19:57 - mmengine - INFO - Iter(train) [ 60350/160000] base_lr: 6.2871e-05 lr: 2.3245e-07 eta: 1 day, 3:40:19 time: 0.9976 data_time: 0.0044 memory: 8462 loss: 0.0122 decode.loss_ce: 0.0057 decode.acc_seg: 99.7942 aux.loss_ce: 0.0066 aux.acc_seg: 99.4230 +04/18 01:20:47 - mmengine - INFO - Iter(train) [ 60400/160000] base_lr: 6.2840e-05 lr: 2.3233e-07 eta: 1 day, 3:39:28 time: 0.9969 data_time: 0.0048 memory: 8462 loss: 0.0151 decode.loss_ce: 0.0068 decode.acc_seg: 99.7206 aux.loss_ce: 0.0083 aux.acc_seg: 98.9014 +04/18 01:21:37 - mmengine - INFO - Iter(train) [ 60450/160000] base_lr: 6.2808e-05 lr: 2.3221e-07 eta: 1 day, 3:38:38 time: 0.9959 data_time: 0.0046 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0065 decode.acc_seg: 99.7553 aux.loss_ce: 0.0082 aux.acc_seg: 99.3738 +04/18 01:22:27 - mmengine - INFO - Iter(train) [ 60500/160000] base_lr: 6.2776e-05 lr: 2.3210e-07 eta: 1 day, 3:37:48 time: 0.9971 data_time: 0.0046 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0064 decode.acc_seg: 99.7364 aux.loss_ce: 0.0079 aux.acc_seg: 99.1091 +04/18 01:23:17 - mmengine - INFO - Iter(train) [ 60550/160000] base_lr: 6.2745e-05 lr: 2.3198e-07 eta: 1 day, 3:36:58 time: 0.9960 data_time: 0.0045 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0068 decode.acc_seg: 99.7866 aux.loss_ce: 0.0082 aux.acc_seg: 99.3393 +04/18 01:24:07 - mmengine - INFO - Iter(train) [ 60600/160000] base_lr: 6.2713e-05 lr: 2.3186e-07 eta: 1 day, 3:36:08 time: 0.9975 data_time: 0.0044 memory: 8462 loss: 0.0135 decode.loss_ce: 0.0062 decode.acc_seg: 99.7169 aux.loss_ce: 0.0073 aux.acc_seg: 99.0591 +04/18 01:24:56 - mmengine - INFO - Iter(train) [ 60650/160000] base_lr: 6.2682e-05 lr: 2.3175e-07 eta: 1 day, 3:35:17 time: 0.9964 data_time: 0.0049 memory: 8462 loss: 0.0137 decode.loss_ce: 0.0064 decode.acc_seg: 99.7566 aux.loss_ce: 0.0073 aux.acc_seg: 99.3406 +04/18 01:25:46 - mmengine - INFO - Iter(train) [ 60700/160000] base_lr: 6.2650e-05 lr: 2.3163e-07 eta: 1 day, 3:34:27 time: 0.9955 data_time: 0.0045 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0065 decode.acc_seg: 99.7856 aux.loss_ce: 0.0075 aux.acc_seg: 99.3752 +04/18 01:26:36 - mmengine - INFO - Iter(train) [ 60750/160000] base_lr: 6.2619e-05 lr: 2.3151e-07 eta: 1 day, 3:33:37 time: 0.9950 data_time: 0.0046 memory: 8462 loss: 0.0146 decode.loss_ce: 0.0067 decode.acc_seg: 99.7543 aux.loss_ce: 0.0079 aux.acc_seg: 99.1589 +04/18 01:27:26 - mmengine - INFO - Iter(train) [ 60800/160000] base_lr: 6.2587e-05 lr: 2.3140e-07 eta: 1 day, 3:32:47 time: 0.9964 data_time: 0.0048 memory: 8462 loss: 0.0130 decode.loss_ce: 0.0062 decode.acc_seg: 99.6580 aux.loss_ce: 0.0068 aux.acc_seg: 98.9519 +04/18 01:28:16 - mmengine - INFO - Iter(train) [ 60850/160000] base_lr: 6.2556e-05 lr: 2.3128e-07 eta: 1 day, 3:31:56 time: 0.9964 data_time: 0.0047 memory: 8462 loss: 0.0133 decode.loss_ce: 0.0060 decode.acc_seg: 99.7095 aux.loss_ce: 0.0073 aux.acc_seg: 99.3185 +04/18 01:29:06 - mmengine - INFO - Iter(train) [ 60900/160000] base_lr: 6.2524e-05 lr: 2.3116e-07 eta: 1 day, 3:31:06 time: 0.9970 data_time: 0.0047 memory: 8462 loss: 0.0153 decode.loss_ce: 0.0069 decode.acc_seg: 99.7679 aux.loss_ce: 0.0084 aux.acc_seg: 99.2031 +04/18 01:29:56 - mmengine - INFO - Iter(train) [ 60950/160000] base_lr: 6.2493e-05 lr: 2.3105e-07 eta: 1 day, 3:30:16 time: 0.9975 data_time: 0.0051 memory: 8462 loss: 0.0150 decode.loss_ce: 0.0065 decode.acc_seg: 99.5949 aux.loss_ce: 0.0085 aux.acc_seg: 98.7904 +04/18 01:30:45 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/18 01:30:45 - mmengine - INFO - Iter(train) [ 61000/160000] base_lr: 6.2461e-05 lr: 2.3093e-07 eta: 1 day, 3:29:26 time: 0.9974 data_time: 0.0051 memory: 8462 loss: 0.0145 decode.loss_ce: 0.0067 decode.acc_seg: 99.8030 aux.loss_ce: 0.0078 aux.acc_seg: 99.4591 +04/18 01:31:35 - mmengine - INFO - Iter(train) [ 61050/160000] base_lr: 6.2429e-05 lr: 2.3081e-07 eta: 1 day, 3:28:36 time: 0.9978 data_time: 0.0047 memory: 8462 loss: 0.0133 decode.loss_ce: 0.0063 decode.acc_seg: 99.8186 aux.loss_ce: 0.0070 aux.acc_seg: 99.3097 +04/18 01:32:25 - mmengine - INFO - Iter(train) [ 61100/160000] base_lr: 6.2398e-05 lr: 2.3070e-07 eta: 1 day, 3:27:46 time: 0.9975 data_time: 0.0047 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0063 decode.acc_seg: 99.6538 aux.loss_ce: 0.0077 aux.acc_seg: 98.9876 +04/18 01:33:15 - mmengine - INFO - Iter(train) [ 61150/160000] base_lr: 6.2366e-05 lr: 2.3058e-07 eta: 1 day, 3:26:55 time: 0.9967 data_time: 0.0051 memory: 8462 loss: 0.0135 decode.loss_ce: 0.0063 decode.acc_seg: 99.7581 aux.loss_ce: 0.0072 aux.acc_seg: 99.3671 +04/18 01:34:05 - mmengine - INFO - Iter(train) [ 61200/160000] base_lr: 6.2335e-05 lr: 2.3046e-07 eta: 1 day, 3:26:05 time: 0.9977 data_time: 0.0049 memory: 8462 loss: 0.0158 decode.loss_ce: 0.0068 decode.acc_seg: 99.7314 aux.loss_ce: 0.0090 aux.acc_seg: 99.1180 +04/18 01:34:55 - mmengine - INFO - Iter(train) [ 61250/160000] base_lr: 6.2303e-05 lr: 2.3035e-07 eta: 1 day, 3:25:15 time: 0.9984 data_time: 0.0047 memory: 8462 loss: 0.0143 decode.loss_ce: 0.0065 decode.acc_seg: 99.7868 aux.loss_ce: 0.0078 aux.acc_seg: 99.3105 +04/18 01:35:45 - mmengine - INFO - Iter(train) [ 61300/160000] base_lr: 6.2272e-05 lr: 2.3023e-07 eta: 1 day, 3:24:25 time: 0.9972 data_time: 0.0048 memory: 8462 loss: 0.0161 decode.loss_ce: 0.0076 decode.acc_seg: 99.7469 aux.loss_ce: 0.0085 aux.acc_seg: 99.2950 +04/18 01:36:34 - mmengine - INFO - Iter(train) [ 61350/160000] base_lr: 6.2240e-05 lr: 2.3011e-07 eta: 1 day, 3:23:35 time: 0.9977 data_time: 0.0048 memory: 8462 loss: 0.0133 decode.loss_ce: 0.0062 decode.acc_seg: 99.6969 aux.loss_ce: 0.0070 aux.acc_seg: 99.0492 +04/18 01:37:24 - mmengine - INFO - Iter(train) [ 61400/160000] base_lr: 6.2209e-05 lr: 2.3000e-07 eta: 1 day, 3:22:44 time: 0.9963 data_time: 0.0048 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0067 decode.acc_seg: 99.6357 aux.loss_ce: 0.0073 aux.acc_seg: 98.8478 +04/18 01:38:14 - mmengine - INFO - Iter(train) [ 61450/160000] base_lr: 6.2177e-05 lr: 2.2988e-07 eta: 1 day, 3:21:54 time: 0.9967 data_time: 0.0048 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0067 decode.acc_seg: 99.7492 aux.loss_ce: 0.0080 aux.acc_seg: 99.3401 +04/18 01:39:04 - mmengine - INFO - Iter(train) [ 61500/160000] base_lr: 6.2146e-05 lr: 2.2976e-07 eta: 1 day, 3:21:04 time: 0.9971 data_time: 0.0048 memory: 8462 loss: 0.0116 decode.loss_ce: 0.0053 decode.acc_seg: 99.7766 aux.loss_ce: 0.0063 aux.acc_seg: 99.2941 +04/18 01:39:54 - mmengine - INFO - Iter(train) [ 61550/160000] base_lr: 6.2114e-05 lr: 2.2965e-07 eta: 1 day, 3:20:14 time: 0.9985 data_time: 0.0051 memory: 8462 loss: 0.0145 decode.loss_ce: 0.0066 decode.acc_seg: 99.6946 aux.loss_ce: 0.0079 aux.acc_seg: 99.2332 +04/18 01:40:44 - mmengine - INFO - Iter(train) [ 61600/160000] base_lr: 6.2082e-05 lr: 2.2953e-07 eta: 1 day, 3:19:24 time: 0.9981 data_time: 0.0046 memory: 8462 loss: 0.0155 decode.loss_ce: 0.0071 decode.acc_seg: 99.8177 aux.loss_ce: 0.0085 aux.acc_seg: 99.5119 +04/18 01:41:34 - mmengine - INFO - Iter(train) [ 61650/160000] base_lr: 6.2051e-05 lr: 2.2941e-07 eta: 1 day, 3:18:34 time: 0.9976 data_time: 0.0047 memory: 8462 loss: 0.0139 decode.loss_ce: 0.0062 decode.acc_seg: 99.7576 aux.loss_ce: 0.0076 aux.acc_seg: 99.3904 +04/18 01:42:24 - mmengine - INFO - Iter(train) [ 61700/160000] base_lr: 6.2019e-05 lr: 2.2930e-07 eta: 1 day, 3:17:43 time: 0.9971 data_time: 0.0048 memory: 8462 loss: 0.0136 decode.loss_ce: 0.0060 decode.acc_seg: 99.7002 aux.loss_ce: 0.0075 aux.acc_seg: 99.2342 +04/18 01:43:13 - mmengine - INFO - Iter(train) [ 61750/160000] base_lr: 6.1988e-05 lr: 2.2918e-07 eta: 1 day, 3:16:53 time: 0.9980 data_time: 0.0053 memory: 8462 loss: 0.0144 decode.loss_ce: 0.0069 decode.acc_seg: 99.7612 aux.loss_ce: 0.0075 aux.acc_seg: 99.1423 +04/18 01:44:03 - mmengine - INFO - Iter(train) [ 61800/160000] base_lr: 6.1956e-05 lr: 2.2906e-07 eta: 1 day, 3:16:03 time: 0.9973 data_time: 0.0052 memory: 8462 loss: 0.0142 decode.loss_ce: 0.0067 decode.acc_seg: 99.6469 aux.loss_ce: 0.0075 aux.acc_seg: 98.9435 +04/18 01:44:53 - mmengine - INFO - Iter(train) [ 61850/160000] base_lr: 6.1925e-05 lr: 2.2895e-07 eta: 1 day, 3:15:13 time: 0.9978 data_time: 0.0045 memory: 8462 loss: 0.0123 decode.loss_ce: 0.0058 decode.acc_seg: 99.7133 aux.loss_ce: 0.0065 aux.acc_seg: 99.1039 +04/18 01:45:43 - mmengine - INFO - Iter(train) [ 61900/160000] base_lr: 6.1893e-05 lr: 2.2883e-07 eta: 1 day, 3:14:23 time: 0.9968 data_time: 0.0050 memory: 8462 loss: 0.0149 decode.loss_ce: 0.0069 decode.acc_seg: 99.7814 aux.loss_ce: 0.0079 aux.acc_seg: 99.4324 +04/18 01:46:33 - mmengine - INFO - Iter(train) [ 61950/160000] base_lr: 6.1862e-05 lr: 2.2872e-07 eta: 1 day, 3:13:33 time: 0.9974 data_time: 0.0047 memory: 8462 loss: 0.0142 decode.loss_ce: 0.0065 decode.acc_seg: 99.6693 aux.loss_ce: 0.0077 aux.acc_seg: 99.1661 +04/18 01:47:23 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/18 01:47:23 - mmengine - INFO - Iter(train) [ 62000/160000] base_lr: 6.1830e-05 lr: 2.2860e-07 eta: 1 day, 3:12:42 time: 0.9955 data_time: 0.0049 memory: 8462 loss: 0.0130 decode.loss_ce: 0.0055 decode.acc_seg: 99.8205 aux.loss_ce: 0.0075 aux.acc_seg: 99.5090 +04/18 01:48:13 - mmengine - INFO - Iter(train) [ 62050/160000] base_lr: 6.1798e-05 lr: 2.2848e-07 eta: 1 day, 3:11:52 time: 0.9960 data_time: 0.0050 memory: 8462 loss: 0.0140 decode.loss_ce: 0.0064 decode.acc_seg: 99.7391 aux.loss_ce: 0.0076 aux.acc_seg: 99.2418 +04/18 01:49:02 - mmengine - INFO - Iter(train) [ 62100/160000] base_lr: 6.1767e-05 lr: 2.2837e-07 eta: 1 day, 3:11:02 time: 0.9969 data_time: 0.0045 memory: 8462 loss: 0.0145 decode.loss_ce: 0.0068 decode.acc_seg: 99.7988 aux.loss_ce: 0.0077 aux.acc_seg: 99.5373 +04/18 01:49:52 - mmengine - INFO - Iter(train) [ 62150/160000] base_lr: 6.1735e-05 lr: 2.2825e-07 eta: 1 day, 3:10:12 time: 0.9966 data_time: 0.0047 memory: 8462 loss: 0.0122 decode.loss_ce: 0.0056 decode.acc_seg: 99.6365 aux.loss_ce: 0.0066 aux.acc_seg: 99.0641 +04/18 01:50:42 - mmengine - INFO - Iter(train) [ 62200/160000] base_lr: 6.1704e-05 lr: 2.2813e-07 eta: 1 day, 3:09:22 time: 0.9986 data_time: 0.0054 memory: 8462 loss: 0.0139 decode.loss_ce: 0.0062 decode.acc_seg: 99.7803 aux.loss_ce: 0.0077 aux.acc_seg: 99.3336 +04/18 01:51:32 - mmengine - INFO - Iter(train) [ 62250/160000] base_lr: 6.1672e-05 lr: 2.2802e-07 eta: 1 day, 3:08:31 time: 0.9963 data_time: 0.0048 memory: 8462 loss: 0.0119 decode.loss_ce: 0.0053 decode.acc_seg: 99.7755 aux.loss_ce: 0.0066 aux.acc_seg: 99.2903 +04/18 01:52:22 - mmengine - INFO - Iter(train) [ 62300/160000] base_lr: 6.1641e-05 lr: 2.2790e-07 eta: 1 day, 3:07:41 time: 0.9970 data_time: 0.0051 memory: 8462 loss: 0.0139 decode.loss_ce: 0.0064 decode.acc_seg: 99.7145 aux.loss_ce: 0.0075 aux.acc_seg: 99.3723 +04/18 01:53:12 - mmengine - INFO - Iter(train) [ 62350/160000] base_lr: 6.1609e-05 lr: 2.2778e-07 eta: 1 day, 3:06:51 time: 0.9967 data_time: 0.0046 memory: 8462 loss: 0.0138 decode.loss_ce: 0.0065 decode.acc_seg: 99.7660 aux.loss_ce: 0.0073 aux.acc_seg: 99.2767 +04/18 01:54:02 - mmengine - INFO - Iter(train) [ 62400/160000] base_lr: 6.1578e-05 lr: 2.2767e-07 eta: 1 day, 3:06:01 time: 0.9980 data_time: 0.0047 memory: 8462 loss: 0.0133 decode.loss_ce: 0.0061 decode.acc_seg: 99.7074 aux.loss_ce: 0.0072 aux.acc_seg: 99.1819 +04/18 01:54:52 - mmengine - INFO - Iter(train) [ 62450/160000] base_lr: 6.1546e-05 lr: 2.2755e-07 eta: 1 day, 3:05:11 time: 0.9972 data_time: 0.0046 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0065 decode.acc_seg: 99.6447 aux.loss_ce: 0.0082 aux.acc_seg: 98.9105 +04/18 01:55:41 - mmengine - INFO - Iter(train) [ 62500/160000] base_lr: 6.1515e-05 lr: 2.2743e-07 eta: 1 day, 3:04:21 time: 0.9970 data_time: 0.0047 memory: 8462 loss: 0.0123 decode.loss_ce: 0.0054 decode.acc_seg: 99.7746 aux.loss_ce: 0.0069 aux.acc_seg: 99.0730 +04/18 01:56:31 - mmengine - INFO - Iter(train) [ 62550/160000] base_lr: 6.1483e-05 lr: 2.2732e-07 eta: 1 day, 3:03:30 time: 0.9986 data_time: 0.0045 memory: 8462 loss: 0.0136 decode.loss_ce: 0.0061 decode.acc_seg: 99.7644 aux.loss_ce: 0.0075 aux.acc_seg: 99.2481 +04/18 01:57:21 - mmengine - INFO - Iter(train) [ 62600/160000] base_lr: 6.1451e-05 lr: 2.2720e-07 eta: 1 day, 3:02:40 time: 0.9972 data_time: 0.0051 memory: 8462 loss: 0.0144 decode.loss_ce: 0.0066 decode.acc_seg: 99.8302 aux.loss_ce: 0.0078 aux.acc_seg: 99.4967 +04/18 01:58:11 - mmengine - INFO - Iter(train) [ 62650/160000] base_lr: 6.1420e-05 lr: 2.2708e-07 eta: 1 day, 3:01:50 time: 0.9978 data_time: 0.0046 memory: 8462 loss: 0.0128 decode.loss_ce: 0.0060 decode.acc_seg: 99.7139 aux.loss_ce: 0.0068 aux.acc_seg: 99.2033 +04/18 01:59:01 - mmengine - INFO - Iter(train) [ 62700/160000] base_lr: 6.1388e-05 lr: 2.2697e-07 eta: 1 day, 3:01:00 time: 0.9977 data_time: 0.0045 memory: 8462 loss: 0.0117 decode.loss_ce: 0.0053 decode.acc_seg: 99.8123 aux.loss_ce: 0.0064 aux.acc_seg: 99.2870 +04/18 01:59:51 - mmengine - INFO - Iter(train) [ 62750/160000] base_lr: 6.1357e-05 lr: 2.2685e-07 eta: 1 day, 3:00:10 time: 0.9971 data_time: 0.0048 memory: 8462 loss: 0.0139 decode.loss_ce: 0.0062 decode.acc_seg: 99.8270 aux.loss_ce: 0.0077 aux.acc_seg: 99.4904 +04/18 02:00:41 - mmengine - INFO - Iter(train) [ 62800/160000] base_lr: 6.1325e-05 lr: 2.2673e-07 eta: 1 day, 2:59:20 time: 0.9975 data_time: 0.0046 memory: 8462 loss: 0.0134 decode.loss_ce: 0.0062 decode.acc_seg: 99.7332 aux.loss_ce: 0.0072 aux.acc_seg: 99.1310 +04/18 02:01:30 - mmengine - INFO - Iter(train) [ 62850/160000] base_lr: 6.1294e-05 lr: 2.2662e-07 eta: 1 day, 2:58:29 time: 0.9970 data_time: 0.0047 memory: 8462 loss: 0.0147 decode.loss_ce: 0.0068 decode.acc_seg: 99.6639 aux.loss_ce: 0.0079 aux.acc_seg: 99.0673 +04/18 02:02:20 - mmengine - INFO - Iter(train) [ 62900/160000] base_lr: 6.1262e-05 lr: 2.2650e-07 eta: 1 day, 2:57:39 time: 0.9967 data_time: 0.0046 memory: 8462 loss: 0.0126 decode.loss_ce: 0.0059 decode.acc_seg: 99.7194 aux.loss_ce: 0.0068 aux.acc_seg: 99.1749 +04/18 02:03:10 - mmengine - INFO - Iter(train) [ 62950/160000] base_lr: 6.1231e-05 lr: 2.2638e-07 eta: 1 day, 2:56:49 time: 0.9971 data_time: 0.0049 memory: 8462 loss: 0.0132 decode.loss_ce: 0.0061 decode.acc_seg: 99.6910 aux.loss_ce: 0.0071 aux.acc_seg: 99.0683 +04/18 02:04:00 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240417_083130 +04/18 02:04:00 - mmengine - INFO - Iter(train) [ 63000/160000] base_lr: 6.1199e-05 lr: 2.2627e-07 eta: 1 day, 2:55:59 time: 0.9970 data_time: 0.0051 memory: 8462 loss: 0.0132 decode.loss_ce: 0.0059 decode.acc_seg: 99.7391 aux.loss_ce: 0.0073 aux.acc_seg: 99.4577 +04/18 02:04:50 - mmengine - INFO - Iter(train) [ 63050/160000] base_lr: 6.1168e-05 lr: 2.2615e-07 eta: 1 day, 2:55:09 time: 0.9969 data_time: 0.0047 memory: 8462 loss: 0.0121 decode.loss_ce: 0.0053 decode.acc_seg: 99.7906 aux.loss_ce: 0.0067 aux.acc_seg: 99.2176 +04/18 02:05:40 - mmengine - INFO - Iter(train) [ 63100/160000] base_lr: 6.1136e-05 lr: 2.2603e-07 eta: 1 day, 2:54:19 time: 0.9977 data_time: 0.0049 memory: 8462 loss: 0.0127 decode.loss_ce: 0.0059 decode.acc_seg: 99.7208 aux.loss_ce: 0.0068 aux.acc_seg: 99.0906 +04/18 02:06:30 - mmengine - INFO - Iter(train) [ 63150/160000] base_lr: 6.1104e-05 lr: 2.2592e-07 eta: 1 day, 2:53:29 time: 0.9964 data_time: 0.0047 memory: 8462 loss: 0.0129 decode.loss_ce: 0.0059 decode.acc_seg: 99.7169 aux.loss_ce: 0.0070 aux.acc_seg: 99.2624 +04/18 02:07:20 - mmengine - INFO - Iter(train) [ 63200/160000] base_lr: 6.1073e-05 lr: 2.2580e-07 eta: 1 day, 2:52:38 time: 0.9967 data_time: 0.0047 memory: 8462 loss: 0.0148 decode.loss_ce: 0.0068 decode.acc_seg: 99.7906 aux.loss_ce: 0.0080 aux.acc_seg: 99.3170 +04/18 02:08:09 - mmengine - INFO - Iter(train) [ 63250/160000] base_lr: 6.1041e-05 lr: 2.2568e-07 eta: 1 day, 2:51:48 time: 0.9979 data_time: 0.0050 memory: 8462 loss: 0.0128 decode.loss_ce: 0.0057 decode.acc_seg: 99.7709 aux.loss_ce: 0.0072 aux.acc_seg: 99.2205 +04/18 02:08:59 - mmengine - INFO - Iter(train) [ 63300/160000] base_lr: 6.1010e-05 lr: 2.2557e-07 eta: 1 day, 2:50:58 time: 0.9969 data_time: 0.0047 memory: 8462 loss: 0.0132 decode.loss_ce: 0.0057 decode.acc_seg: 99.6817 aux.loss_ce: 0.0075 aux.acc_seg: 99.1745 +04/18 02:09:49 - mmengine - INFO - Iter(train) [ 63350/160000] base_lr: 6.0978e-05 lr: 2.2545e-07 eta: 1 day, 2:50:08 time: 0.9983 data_time: 0.0056 memory: 8462 loss: 0.0157 decode.loss_ce: 0.0069 decode.acc_seg: 99.7620 aux.loss_ce: 0.0089 aux.acc_seg: 99.2807 +04/18 02:10:39 - mmengine - INFO - Iter(train) [ 63400/160000] base_lr: 6.0947e-05 lr: 2.2533e-07 eta: 1 day, 2:49:18 time: 0.9960 data_time: 0.0049 memory: 8462 loss: 0.0151 decode.loss_ce: 0.0068 decode.acc_seg: 99.7185 aux.loss_ce: 0.0083 aux.acc_seg: 99.2039 +04/18 02:11:29 - mmengine - INFO - Iter(train) [ 63450/160000] base_lr: 6.0915e-05 lr: 2.2522e-07 eta: 1 day, 2:48:28 time: 0.9980 data_time: 0.0053 memory: 8462 loss: 0.0141 decode.loss_ce: 0.0064 decode.acc_seg: 99.8201 aux.loss_ce: 0.0077 aux.acc_seg: 99.3877 +WARNING:torch.distributed.elastic.agent.server.api:Received 2 death signal, shutting down workers +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 174801 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 174802 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 174803 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 174804 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 174801 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 174802 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 174803 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 174804 closing signal SIGTERM +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 709, in run + result = self._invoke_run(role) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 850, in _invoke_run + time.sleep(monitor_interval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 174770 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 716, in run + self._shutdown(e.sigval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 174770 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 236, in launch_agent + result = agent.run() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/metrics/api.py", line 125, in wrapper + result = f(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 721, in run + self._shutdown() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 174770 got signal: 2 +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/18 02:20:10 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 1925317996 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 1925317996 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/18 02:20:11 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = None +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=1024, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=2, + num_convs=1, + type='FCNHead'), + backbone=dict( + contract_dilation=True, + depth=101, + dilations=( + 1, + 1, + 1, + 1, + ), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + norm_eval=False, + num_stages=4, + out_indices=( + 0, + 1, + 2, + 3, + ), + strides=( + 1, + 2, + 2, + 2, + ), + style='pytorch', + type='ResNetV1c'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=512, + dropout_ratio=0.1, + in_channels=[ + 256, + 512, + 1024, + 2048, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=2, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + pretrained='open-mmlab://resnet101_v1c', + test_cfg=dict(mode='whole'), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + clip_grad=None, + optimizer=dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, + by_epoch=False, + end=160000, + eta_min=0.0001, + power=0.9, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=4, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = '/workspaces/mmsegmentation-1/work_dirs/R101_4000' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/18 02:20:13 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +04/18 02:20:14 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +04/18 02:20:14 - mmengine - INFO - load model from: open-mmlab://resnet101_v1c +04/18 02:20:14 - mmengine - INFO - Loads checkpoint by openmmlab backend from path: open-mmlab://resnet101_v1c +04/18 02:20:14 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: fc.weight, fc.bias + +04/18 02:20:14 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +04/18 02:20:14 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +04/18 02:20:14 - mmengine - INFO - Checkpoints will be saved to /workspaces/mmsegmentation-1/work_dirs/R101_4000. +04/18 02:20:46 - mmengine - INFO - Iter(train) [ 50/160000] lr: 9.9973e-03 eta: 1 day, 4:42:15 time: 0.5461 data_time: 0.0064 memory: 7635 loss: 0.1394 decode.loss_ce: 0.0903 decode.acc_seg: 96.5318 aux.loss_ce: 0.0491 aux.acc_seg: 96.5318 +04/18 02:21:14 - mmengine - INFO - Iter(train) [ 100/160000] lr: 9.9945e-03 eta: 1 day, 2:30:49 time: 0.5490 data_time: 0.0057 memory: 7635 loss: 0.0858 decode.loss_ce: 0.0528 decode.acc_seg: 97.9885 aux.loss_ce: 0.0330 aux.acc_seg: 97.1264 +04/18 02:21:41 - mmengine - INFO - Iter(train) [ 150/160000] lr: 9.9917e-03 eta: 1 day, 1:47:10 time: 0.5487 data_time: 0.0061 memory: 7635 loss: 0.0797 decode.loss_ce: 0.0480 decode.acc_seg: 98.1838 aux.loss_ce: 0.0317 aux.acc_seg: 96.6573 +Traceback (most recent call last): +Traceback (most recent call last): +Traceback (most recent call last): +Traceback (most recent call last): + File "tools/train.py", line 104, in + File "tools/train.py", line 104, in + File "tools/train.py", line 104, in + File "tools/train.py", line 104, in +WARNING:torch.distributed.elastic.agent.server.api:Received 2 death signal, shutting down workers + main()main()main()main() +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 592936 closing signal SIGINT + + + + File "tools/train.py", line 100, in main + File "tools/train.py", line 100, in main + File "tools/train.py", line 100, in main + File "tools/train.py", line 100, in main +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 592937 closing signal SIGINT + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 592938 closing signal SIGINT +runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 592939 closing signal SIGINT + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + self.run_iter(data_batch) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + model = self.train_loop.run() # type: ignoreoutputs = self.runner.model.train_step( + +model = self.train_loop.run() # type: ignoremodel = self.train_loop.run() # type: ignore File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + + + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + optim_wrapper.update_params(parsed_loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 201, in update_params + self.run_iter(data_batch)self.run_iter(data_batch) + + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter +self.run_iter(data_batch) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + self.step(**step_kwargs) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/scheduler/param_scheduler.py", line 115, in wrapper + outputs = self.runner.model.train_step(outputs = self.runner.model.train_step( + + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + return wrapped(*args, **kwargs)outputs = self.runner.model.train_step( + + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 253, in step + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + optim_wrapper.update_params(parsed_loss)optim_wrapper.update_params(parsed_loss) + + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 201, in update_params + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 201, in update_params + optim_wrapper.update_params(parsed_loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 201, in update_params +self.optimizer.step(**kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/optim/optimizer.py", line 88, in wrapper + self.step(**step_kwargs) +self.step(**step_kwargs) File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/scheduler/param_scheduler.py", line 115, in wrapper + + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/scheduler/param_scheduler.py", line 115, in wrapper + return func(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context +self.step(**step_kwargs) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/scheduler/param_scheduler.py", line 115, in wrapper +return wrapped(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 253, in step +return wrapped(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 253, in step + return func(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/optim/sgd.py", line 144, in step + return wrapped(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 253, in step + self.optimizer.step(**kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/optim/optimizer.py", line 88, in wrapper + self.optimizer.step(**kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/optim/optimizer.py", line 88, in wrapper + return func(*args, **kwargs) +self.optimizer.step(**kwargs) File "/opt/conda/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context + + File "/opt/conda/lib/python3.8/site-packages/torch/optim/optimizer.py", line 88, in wrapper + return func(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context + return func(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/optim/sgd.py", line 144, in step + return func(*args, **kwargs)return func(*args, **kwargs) + + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context + File "/opt/conda/lib/python3.8/site-packages/torch/optim/sgd.py", line 144, in step + return func(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/optim/sgd.py", line 144, in step + F.sgd(params_with_grad,F.sgd(params_with_grad,F.sgd(params_with_grad,F.sgd(params_with_grad, + + + + File "/opt/conda/lib/python3.8/site-packages/torch/optim/_functional.py", line 186, in sgd + File "/opt/conda/lib/python3.8/site-packages/torch/optim/_functional.py", line 194, in sgd + File "/opt/conda/lib/python3.8/site-packages/torch/optim/_functional.py", line 177, in sgd + File "/opt/conda/lib/python3.8/site-packages/torch/optim/_functional.py", line 194, in sgd + param.add_(d_p, alpha=alpha) buf.mul_(momentum).add_(d_p, alpha=1 - dampening)d_p = d_p.add(param, alpha=weight_decay) +param.add_(d_p, alpha=alpha) +KeyboardInterrupt + + +KeyboardInterruptKeyboardInterruptKeyboardInterrupt + + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 236, in launch_agent + result = agent.run() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/metrics/api.py", line 125, in wrapper + result = f(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 709, in run + result = self._invoke_run(role) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 850, in _invoke_run + time.sleep(monitor_interval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 592901 got signal: 2 +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/18 02:22:14 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 1770541458 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 1770541458 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/18 02:22:14 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = None +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=1024, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=2, + num_convs=1, + type='FCNHead'), + backbone=dict( + contract_dilation=True, + depth=101, + dilations=( + 1, + 1, + 1, + 1, + ), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + norm_eval=False, + num_stages=4, + out_indices=( + 0, + 1, + 2, + 3, + ), + strides=( + 1, + 2, + 2, + 2, + ), + style='pytorch', + type='ResNetV1c'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=512, + dropout_ratio=0.1, + in_channels=[ + 256, + 512, + 1024, + 2048, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=2, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + pretrained='open-mmlab://resnet101_v1c', + test_cfg=dict(mode='whole'), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + clip_grad=None, + optimizer=dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, + by_epoch=False, + end=160000, + eta_min=0.0001, + power=0.9, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=4, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = '/workspaces/mmsegmentation-1/work_dirs/R101_4000' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/resnet.py:431: UserWarning: DeprecationWarning: pretrained is a deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is a deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py:120: UserWarning: For binary segmentation, we suggest using`out_channels = 1` to define the outputchannels of segmentor, and use `threshold`to convert `seg_logits` into a predictionapplying a threshold + warnings.warn('For binary segmentation, we suggest using' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/18 02:22:16 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +04/18 02:22:17 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +04/18 02:22:18 - mmengine - INFO - load model from: open-mmlab://resnet101_v1c +04/18 02:22:18 - mmengine - INFO - Loads checkpoint by openmmlab backend from path: open-mmlab://resnet101_v1c +04/18 02:22:18 - mmengine - WARNING - The model and loaded state dict do not match exactly + +unexpected key in source state_dict: fc.weight, fc.bias + +04/18 02:22:18 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +04/18 02:22:18 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +04/18 02:22:18 - mmengine - INFO - Checkpoints will be saved to /workspaces/mmsegmentation-1/work_dirs/R101_4000. +04/18 02:22:50 - mmengine - INFO - Iter(train) [ 50/160000] lr: 9.9973e-03 eta: 1 day, 4:43:49 time: 0.5456 data_time: 0.0068 memory: 7635 loss: 0.1351 decode.loss_ce: 0.0875 decode.acc_seg: 97.6007 aux.loss_ce: 0.0476 aux.acc_seg: 97.6006 +04/18 02:23:17 - mmengine - INFO - Iter(train) [ 100/160000] lr: 9.9945e-03 eta: 1 day, 2:29:41 time: 0.5467 data_time: 0.0057 memory: 7635 loss: 0.0936 decode.loss_ce: 0.0562 decode.acc_seg: 97.6175 aux.loss_ce: 0.0374 aux.acc_seg: 96.2901 +04/18 02:23:45 - mmengine - INFO - Iter(train) [ 150/160000] lr: 9.9917e-03 eta: 1 day, 1:44:59 time: 0.5467 data_time: 0.0065 memory: 7635 loss: 0.0822 decode.loss_ce: 0.0503 decode.acc_seg: 97.9509 aux.loss_ce: 0.0319 aux.acc_seg: 97.1464 +04/18 02:24:12 - mmengine - INFO - Iter(train) [ 200/160000] lr: 9.9889e-03 eta: 1 day, 1:22:56 time: 0.5485 data_time: 0.0064 memory: 7635 loss: 0.0688 decode.loss_ce: 0.0416 decode.acc_seg: 98.1911 aux.loss_ce: 0.0272 aux.acc_seg: 96.7990 +04/18 02:24:40 - mmengine - INFO - Iter(train) [ 250/160000] lr: 9.9861e-03 eta: 1 day, 1:09:55 time: 0.5487 data_time: 0.0058 memory: 7635 loss: 0.0830 decode.loss_ce: 0.0518 decode.acc_seg: 98.1723 aux.loss_ce: 0.0313 aux.acc_seg: 96.9854 +04/18 02:25:07 - mmengine - INFO - Iter(train) [ 300/160000] lr: 9.9833e-03 eta: 1 day, 1:00:47 time: 0.5491 data_time: 0.0063 memory: 7635 loss: 0.0735 decode.loss_ce: 0.0464 decode.acc_seg: 98.2391 aux.loss_ce: 0.0270 aux.acc_seg: 97.6198 +04/18 02:25:34 - mmengine - INFO - Iter(train) [ 350/160000] lr: 9.9806e-03 eta: 1 day, 0:54:07 time: 0.5471 data_time: 0.0063 memory: 7635 loss: 0.0653 decode.loss_ce: 0.0413 decode.acc_seg: 98.3918 aux.loss_ce: 0.0241 aux.acc_seg: 97.8115 +04/18 02:26:02 - mmengine - INFO - Iter(train) [ 400/160000] lr: 9.9778e-03 eta: 1 day, 0:49:12 time: 0.5473 data_time: 0.0061 memory: 7635 loss: 0.0583 decode.loss_ce: 0.0371 decode.acc_seg: 98.5509 aux.loss_ce: 0.0212 aux.acc_seg: 98.1630 +04/18 02:26:29 - mmengine - INFO - Iter(train) [ 450/160000] lr: 9.9750e-03 eta: 1 day, 0:45:04 time: 0.5478 data_time: 0.0058 memory: 7635 loss: 0.0662 decode.loss_ce: 0.0430 decode.acc_seg: 98.4915 aux.loss_ce: 0.0232 aux.acc_seg: 98.0277 +04/18 02:26:57 - mmengine - INFO - Iter(train) [ 500/160000] lr: 9.9722e-03 eta: 1 day, 0:43:03 time: 0.5475 data_time: 0.0059 memory: 7635 loss: 0.0593 decode.loss_ce: 0.0378 decode.acc_seg: 98.2600 aux.loss_ce: 0.0215 aux.acc_seg: 97.6308 +04/18 02:27:24 - mmengine - INFO - Iter(train) [ 550/160000] lr: 9.9694e-03 eta: 1 day, 0:40:14 time: 0.5483 data_time: 0.0059 memory: 7635 loss: 0.0571 decode.loss_ce: 0.0364 decode.acc_seg: 97.7746 aux.loss_ce: 0.0207 aux.acc_seg: 96.9963 +04/18 02:27:52 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 02:27:52 - mmengine - INFO - Iter(train) [ 600/160000] lr: 9.9666e-03 eta: 1 day, 0:37:48 time: 0.5475 data_time: 0.0061 memory: 7635 loss: 0.0485 decode.loss_ce: 0.0307 decode.acc_seg: 99.0183 aux.loss_ce: 0.0178 aux.acc_seg: 98.3242 +04/18 02:28:19 - mmengine - INFO - Iter(train) [ 650/160000] lr: 9.9639e-03 eta: 1 day, 0:35:39 time: 0.5482 data_time: 0.0057 memory: 7635 loss: 0.0573 decode.loss_ce: 0.0367 decode.acc_seg: 99.0075 aux.loss_ce: 0.0206 aux.acc_seg: 98.4173 +04/18 02:28:46 - mmengine - INFO - Iter(train) [ 700/160000] lr: 9.9611e-03 eta: 1 day, 0:33:52 time: 0.5494 data_time: 0.0071 memory: 7635 loss: 0.0468 decode.loss_ce: 0.0293 decode.acc_seg: 98.9671 aux.loss_ce: 0.0175 aux.acc_seg: 98.4134 +04/18 02:29:14 - mmengine - INFO - Iter(train) [ 750/160000] lr: 9.9583e-03 eta: 1 day, 0:32:04 time: 0.5473 data_time: 0.0054 memory: 7635 loss: 0.0509 decode.loss_ce: 0.0322 decode.acc_seg: 98.5582 aux.loss_ce: 0.0187 aux.acc_seg: 97.6395 +04/18 02:29:41 - mmengine - INFO - Iter(train) [ 800/160000] lr: 9.9555e-03 eta: 1 day, 0:30:34 time: 0.5486 data_time: 0.0062 memory: 7635 loss: 0.0491 decode.loss_ce: 0.0310 decode.acc_seg: 99.2507 aux.loss_ce: 0.0181 aux.acc_seg: 98.8086 +04/18 02:30:09 - mmengine - INFO - Iter(train) [ 850/160000] lr: 9.9527e-03 eta: 1 day, 0:29:11 time: 0.5487 data_time: 0.0062 memory: 7635 loss: 0.0514 decode.loss_ce: 0.0329 decode.acc_seg: 98.4890 aux.loss_ce: 0.0185 aux.acc_seg: 97.9950 +04/18 02:30:36 - mmengine - INFO - Iter(train) [ 900/160000] lr: 9.9499e-03 eta: 1 day, 0:27:56 time: 0.5475 data_time: 0.0056 memory: 7635 loss: 0.0542 decode.loss_ce: 0.0350 decode.acc_seg: 98.8522 aux.loss_ce: 0.0192 aux.acc_seg: 98.3143 +04/18 02:31:03 - mmengine - INFO - Iter(train) [ 950/160000] lr: 9.9471e-03 eta: 1 day, 0:26:46 time: 0.5476 data_time: 0.0065 memory: 7635 loss: 0.0471 decode.loss_ce: 0.0298 decode.acc_seg: 98.2690 aux.loss_ce: 0.0173 aux.acc_seg: 97.9364 +04/18 02:31:31 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 02:31:31 - mmengine - INFO - Iter(train) [ 1000/160000] lr: 9.9444e-03 eta: 1 day, 0:25:43 time: 0.5505 data_time: 0.0067 memory: 7635 loss: 0.0458 decode.loss_ce: 0.0291 decode.acc_seg: 98.5917 aux.loss_ce: 0.0167 aux.acc_seg: 98.0520 +04/18 02:31:58 - mmengine - INFO - Iter(train) [ 1050/160000] lr: 9.9416e-03 eta: 1 day, 0:24:38 time: 0.5486 data_time: 0.0060 memory: 7635 loss: 0.0461 decode.loss_ce: 0.0289 decode.acc_seg: 99.0119 aux.loss_ce: 0.0172 aux.acc_seg: 98.5758 +04/18 02:32:26 - mmengine - INFO - Iter(train) [ 1100/160000] lr: 9.9388e-03 eta: 1 day, 0:23:44 time: 0.5486 data_time: 0.0055 memory: 7635 loss: 0.0462 decode.loss_ce: 0.0294 decode.acc_seg: 98.2671 aux.loss_ce: 0.0168 aux.acc_seg: 97.6591 +04/18 02:32:53 - mmengine - INFO - Iter(train) [ 1150/160000] lr: 9.9360e-03 eta: 1 day, 0:22:49 time: 0.5476 data_time: 0.0059 memory: 7635 loss: 0.0401 decode.loss_ce: 0.0249 decode.acc_seg: 98.8012 aux.loss_ce: 0.0152 aux.acc_seg: 98.5517 +04/18 02:33:21 - mmengine - INFO - Iter(train) [ 1200/160000] lr: 9.9332e-03 eta: 1 day, 0:21:56 time: 0.5482 data_time: 0.0059 memory: 7635 loss: 0.0461 decode.loss_ce: 0.0291 decode.acc_seg: 98.8809 aux.loss_ce: 0.0170 aux.acc_seg: 98.1678 +04/18 02:33:48 - mmengine - INFO - Iter(train) [ 1250/160000] lr: 9.9304e-03 eta: 1 day, 0:21:04 time: 0.5486 data_time: 0.0064 memory: 7635 loss: 0.0391 decode.loss_ce: 0.0240 decode.acc_seg: 98.9322 aux.loss_ce: 0.0151 aux.acc_seg: 98.2064 +04/18 02:34:16 - mmengine - INFO - Iter(train) [ 1300/160000] lr: 9.9276e-03 eta: 1 day, 0:20:12 time: 0.5475 data_time: 0.0066 memory: 7635 loss: 0.0476 decode.loss_ce: 0.0303 decode.acc_seg: 98.8741 aux.loss_ce: 0.0173 aux.acc_seg: 98.2177 +04/18 02:34:43 - mmengine - INFO - Iter(train) [ 1350/160000] lr: 9.9248e-03 eta: 1 day, 0:19:26 time: 0.5486 data_time: 0.0058 memory: 7635 loss: 0.0481 decode.loss_ce: 0.0312 decode.acc_seg: 98.3006 aux.loss_ce: 0.0170 aux.acc_seg: 97.7006 +04/18 02:35:10 - mmengine - INFO - Iter(train) [ 1400/160000] lr: 9.9221e-03 eta: 1 day, 0:18:39 time: 0.5489 data_time: 0.0063 memory: 7635 loss: 0.0418 decode.loss_ce: 0.0259 decode.acc_seg: 99.1931 aux.loss_ce: 0.0158 aux.acc_seg: 98.7233 +04/18 02:35:38 - mmengine - INFO - Iter(train) [ 1450/160000] lr: 9.9193e-03 eta: 1 day, 0:17:54 time: 0.5478 data_time: 0.0061 memory: 7635 loss: 0.0464 decode.loss_ce: 0.0294 decode.acc_seg: 98.9993 aux.loss_ce: 0.0170 aux.acc_seg: 98.6215 +04/18 02:36:05 - mmengine - INFO - Iter(train) [ 1500/160000] lr: 9.9165e-03 eta: 1 day, 0:17:09 time: 0.5479 data_time: 0.0060 memory: 7635 loss: 0.0441 decode.loss_ce: 0.0276 decode.acc_seg: 98.9414 aux.loss_ce: 0.0164 aux.acc_seg: 98.5452 +04/18 02:36:33 - mmengine - INFO - Iter(train) [ 1550/160000] lr: 9.9137e-03 eta: 1 day, 0:16:37 time: 0.5576 data_time: 0.0061 memory: 7635 loss: 0.0364 decode.loss_ce: 0.0221 decode.acc_seg: 99.5236 aux.loss_ce: 0.0143 aux.acc_seg: 99.0460 +04/18 02:37:00 - mmengine - INFO - Iter(train) [ 1600/160000] lr: 9.9109e-03 eta: 1 day, 0:16:04 time: 0.5485 data_time: 0.0061 memory: 7635 loss: 0.0416 decode.loss_ce: 0.0258 decode.acc_seg: 99.0429 aux.loss_ce: 0.0157 aux.acc_seg: 98.5963 +04/18 02:37:28 - mmengine - INFO - Iter(train) [ 1650/160000] lr: 9.9081e-03 eta: 1 day, 0:15:23 time: 0.5487 data_time: 0.0064 memory: 7635 loss: 0.0422 decode.loss_ce: 0.0264 decode.acc_seg: 99.1405 aux.loss_ce: 0.0158 aux.acc_seg: 98.7473 +04/18 02:37:55 - mmengine - INFO - Iter(train) [ 1700/160000] lr: 9.9053e-03 eta: 1 day, 0:14:43 time: 0.5485 data_time: 0.0064 memory: 7635 loss: 0.0418 decode.loss_ce: 0.0261 decode.acc_seg: 99.0495 aux.loss_ce: 0.0156 aux.acc_seg: 98.5112 +04/18 02:38:23 - mmengine - INFO - Iter(train) [ 1750/160000] lr: 9.9025e-03 eta: 1 day, 0:14:03 time: 0.5489 data_time: 0.0058 memory: 7635 loss: 0.0423 decode.loss_ce: 0.0264 decode.acc_seg: 98.7845 aux.loss_ce: 0.0159 aux.acc_seg: 98.3470 +04/18 02:38:50 - mmengine - INFO - Iter(train) [ 1800/160000] lr: 9.8998e-03 eta: 1 day, 0:13:24 time: 0.5480 data_time: 0.0059 memory: 7635 loss: 0.0401 decode.loss_ce: 0.0251 decode.acc_seg: 98.5924 aux.loss_ce: 0.0149 aux.acc_seg: 98.2556 +04/18 02:39:17 - mmengine - INFO - Iter(train) [ 1850/160000] lr: 9.8970e-03 eta: 1 day, 0:12:45 time: 0.5498 data_time: 0.0057 memory: 7635 loss: 0.0381 decode.loss_ce: 0.0236 decode.acc_seg: 99.0651 aux.loss_ce: 0.0144 aux.acc_seg: 98.5510 +04/18 02:39:45 - mmengine - INFO - Iter(train) [ 1900/160000] lr: 9.8942e-03 eta: 1 day, 0:12:09 time: 0.5500 data_time: 0.0064 memory: 7635 loss: 0.0368 decode.loss_ce: 0.0227 decode.acc_seg: 98.8979 aux.loss_ce: 0.0140 aux.acc_seg: 98.2182 +04/18 02:40:12 - mmengine - INFO - Iter(train) [ 1950/160000] lr: 9.8914e-03 eta: 1 day, 0:11:31 time: 0.5484 data_time: 0.0064 memory: 7635 loss: 0.0408 decode.loss_ce: 0.0254 decode.acc_seg: 98.6368 aux.loss_ce: 0.0154 aux.acc_seg: 98.1456 +04/18 02:40:40 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 02:40:40 - mmengine - INFO - Iter(train) [ 2000/160000] lr: 9.8886e-03 eta: 1 day, 0:10:54 time: 0.5498 data_time: 0.0057 memory: 7635 loss: 0.0379 decode.loss_ce: 0.0240 decode.acc_seg: 99.2053 aux.loss_ce: 0.0139 aux.acc_seg: 98.7584 +04/18 02:41:07 - mmengine - INFO - Iter(train) [ 2050/160000] lr: 9.8858e-03 eta: 1 day, 0:10:20 time: 0.5499 data_time: 0.0070 memory: 7635 loss: 0.0347 decode.loss_ce: 0.0211 decode.acc_seg: 99.1835 aux.loss_ce: 0.0135 aux.acc_seg: 98.8031 +04/18 02:41:35 - mmengine - INFO - Iter(train) [ 2100/160000] lr: 9.8830e-03 eta: 1 day, 0:09:45 time: 0.5491 data_time: 0.0069 memory: 7635 loss: 0.0420 decode.loss_ce: 0.0262 decode.acc_seg: 98.9922 aux.loss_ce: 0.0159 aux.acc_seg: 98.4855 +04/18 02:42:02 - mmengine - INFO - Iter(train) [ 2150/160000] lr: 9.8802e-03 eta: 1 day, 0:09:10 time: 0.5495 data_time: 0.0058 memory: 7635 loss: 0.0358 decode.loss_ce: 0.0222 decode.acc_seg: 99.1411 aux.loss_ce: 0.0136 aux.acc_seg: 98.6343 +04/18 02:42:30 - mmengine - INFO - Iter(train) [ 2200/160000] lr: 9.8775e-03 eta: 1 day, 0:08:32 time: 0.5481 data_time: 0.0061 memory: 7635 loss: 0.0408 decode.loss_ce: 0.0260 decode.acc_seg: 99.0074 aux.loss_ce: 0.0148 aux.acc_seg: 98.6520 +04/18 02:42:57 - mmengine - INFO - Iter(train) [ 2250/160000] lr: 9.8747e-03 eta: 1 day, 0:07:57 time: 0.5498 data_time: 0.0075 memory: 7635 loss: 0.0407 decode.loss_ce: 0.0253 decode.acc_seg: 98.9144 aux.loss_ce: 0.0154 aux.acc_seg: 98.1933 +04/18 02:43:24 - mmengine - INFO - Iter(train) [ 2300/160000] lr: 9.8719e-03 eta: 1 day, 0:07:21 time: 0.5472 data_time: 0.0059 memory: 7635 loss: 0.0358 decode.loss_ce: 0.0220 decode.acc_seg: 99.2257 aux.loss_ce: 0.0138 aux.acc_seg: 98.7956 +04/18 02:43:52 - mmengine - INFO - Iter(train) [ 2350/160000] lr: 9.8691e-03 eta: 1 day, 0:06:49 time: 0.5498 data_time: 0.0065 memory: 7635 loss: 0.0411 decode.loss_ce: 0.0260 decode.acc_seg: 98.4962 aux.loss_ce: 0.0151 aux.acc_seg: 98.0080 +04/18 02:44:19 - mmengine - INFO - Iter(train) [ 2400/160000] lr: 9.8663e-03 eta: 1 day, 0:06:16 time: 0.5484 data_time: 0.0061 memory: 7635 loss: 0.0372 decode.loss_ce: 0.0232 decode.acc_seg: 99.0525 aux.loss_ce: 0.0140 aux.acc_seg: 98.3671 +04/18 02:44:47 - mmengine - INFO - Iter(train) [ 2450/160000] lr: 9.8635e-03 eta: 1 day, 0:05:42 time: 0.5489 data_time: 0.0065 memory: 7635 loss: 0.0434 decode.loss_ce: 0.0275 decode.acc_seg: 98.9281 aux.loss_ce: 0.0159 aux.acc_seg: 98.5132 +04/18 02:45:14 - mmengine - INFO - Iter(train) [ 2500/160000] lr: 9.8607e-03 eta: 1 day, 0:05:09 time: 0.5488 data_time: 0.0062 memory: 7635 loss: 0.0360 decode.loss_ce: 0.0225 decode.acc_seg: 98.5716 aux.loss_ce: 0.0135 aux.acc_seg: 98.1297 +04/18 02:45:42 - mmengine - INFO - Iter(train) [ 2550/160000] lr: 9.8579e-03 eta: 1 day, 0:04:37 time: 0.5486 data_time: 0.0056 memory: 7635 loss: 0.0358 decode.loss_ce: 0.0223 decode.acc_seg: 99.2933 aux.loss_ce: 0.0136 aux.acc_seg: 98.6950 +04/18 02:46:09 - mmengine - INFO - Iter(train) [ 2600/160000] lr: 9.8551e-03 eta: 1 day, 0:04:06 time: 0.5507 data_time: 0.0060 memory: 7635 loss: 0.0354 decode.loss_ce: 0.0215 decode.acc_seg: 99.2549 aux.loss_ce: 0.0139 aux.acc_seg: 98.7345 +04/18 02:46:37 - mmengine - INFO - Iter(train) [ 2650/160000] lr: 9.8524e-03 eta: 1 day, 0:03:45 time: 0.5579 data_time: 0.0060 memory: 7635 loss: 0.0389 decode.loss_ce: 0.0242 decode.acc_seg: 99.0269 aux.loss_ce: 0.0147 aux.acc_seg: 98.4343 +04/18 02:47:04 - mmengine - INFO - Iter(train) [ 2700/160000] lr: 9.8496e-03 eta: 1 day, 0:03:14 time: 0.5500 data_time: 0.0064 memory: 7635 loss: 0.0323 decode.loss_ce: 0.0193 decode.acc_seg: 99.0790 aux.loss_ce: 0.0130 aux.acc_seg: 98.3951 +04/18 02:47:32 - mmengine - INFO - Iter(train) [ 2750/160000] lr: 9.8468e-03 eta: 1 day, 0:02:42 time: 0.5483 data_time: 0.0061 memory: 7635 loss: 0.0368 decode.loss_ce: 0.0225 decode.acc_seg: 99.2489 aux.loss_ce: 0.0142 aux.acc_seg: 98.7843 +04/18 02:47:59 - mmengine - INFO - Iter(train) [ 2800/160000] lr: 9.8440e-03 eta: 1 day, 0:02:10 time: 0.5489 data_time: 0.0057 memory: 7635 loss: 0.0415 decode.loss_ce: 0.0263 decode.acc_seg: 99.0236 aux.loss_ce: 0.0152 aux.acc_seg: 98.5974 +04/18 02:48:27 - mmengine - INFO - Iter(train) [ 2850/160000] lr: 9.8412e-03 eta: 1 day, 0:01:39 time: 0.5501 data_time: 0.0062 memory: 7635 loss: 0.0386 decode.loss_ce: 0.0239 decode.acc_seg: 99.1429 aux.loss_ce: 0.0147 aux.acc_seg: 98.5730 +04/18 02:48:54 - mmengine - INFO - Iter(train) [ 2900/160000] lr: 9.8384e-03 eta: 1 day, 0:01:06 time: 0.5494 data_time: 0.0059 memory: 7635 loss: 0.0350 decode.loss_ce: 0.0218 decode.acc_seg: 98.8841 aux.loss_ce: 0.0132 aux.acc_seg: 98.5057 +04/18 02:49:21 - mmengine - INFO - Iter(train) [ 2950/160000] lr: 9.8356e-03 eta: 1 day, 0:00:34 time: 0.5494 data_time: 0.0061 memory: 7635 loss: 0.0334 decode.loss_ce: 0.0203 decode.acc_seg: 99.3465 aux.loss_ce: 0.0130 aux.acc_seg: 98.6574 +04/18 02:49:49 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 02:49:49 - mmengine - INFO - Iter(train) [ 3000/160000] lr: 9.8328e-03 eta: 1 day, 0:00:02 time: 0.5490 data_time: 0.0057 memory: 7635 loss: 0.0348 decode.loss_ce: 0.0213 decode.acc_seg: 98.5440 aux.loss_ce: 0.0136 aux.acc_seg: 98.1440 +04/18 02:50:16 - mmengine - INFO - Iter(train) [ 3050/160000] lr: 9.8300e-03 eta: 23:59:30 time: 0.5473 data_time: 0.0061 memory: 7635 loss: 0.0345 decode.loss_ce: 0.0212 decode.acc_seg: 99.1501 aux.loss_ce: 0.0133 aux.acc_seg: 98.5846 +04/18 02:50:44 - mmengine - INFO - Iter(train) [ 3100/160000] lr: 9.8273e-03 eta: 23:58:59 time: 0.5493 data_time: 0.0062 memory: 7635 loss: 0.0375 decode.loss_ce: 0.0233 decode.acc_seg: 98.6369 aux.loss_ce: 0.0142 aux.acc_seg: 98.0972 +04/18 02:51:11 - mmengine - INFO - Iter(train) [ 3150/160000] lr: 9.8245e-03 eta: 23:58:28 time: 0.5484 data_time: 0.0059 memory: 7635 loss: 0.0327 decode.loss_ce: 0.0198 decode.acc_seg: 99.2376 aux.loss_ce: 0.0129 aux.acc_seg: 98.8457 +04/18 02:51:39 - mmengine - INFO - Iter(train) [ 3200/160000] lr: 9.8217e-03 eta: 23:57:55 time: 0.5483 data_time: 0.0059 memory: 7635 loss: 0.0337 decode.loss_ce: 0.0206 decode.acc_seg: 99.3318 aux.loss_ce: 0.0131 aux.acc_seg: 98.8579 +04/18 02:52:06 - mmengine - INFO - Iter(train) [ 3250/160000] lr: 9.8189e-03 eta: 23:57:24 time: 0.5481 data_time: 0.0056 memory: 7635 loss: 0.0344 decode.loss_ce: 0.0214 decode.acc_seg: 99.1192 aux.loss_ce: 0.0130 aux.acc_seg: 98.8572 +04/18 02:52:33 - mmengine - INFO - Iter(train) [ 3300/160000] lr: 9.8161e-03 eta: 23:56:51 time: 0.5476 data_time: 0.0056 memory: 7635 loss: 0.0364 decode.loss_ce: 0.0222 decode.acc_seg: 99.0077 aux.loss_ce: 0.0142 aux.acc_seg: 98.1403 +04/18 02:53:01 - mmengine - INFO - Iter(train) [ 3350/160000] lr: 9.8133e-03 eta: 23:56:17 time: 0.5477 data_time: 0.0063 memory: 7635 loss: 0.0344 decode.loss_ce: 0.0212 decode.acc_seg: 98.9128 aux.loss_ce: 0.0132 aux.acc_seg: 98.4770 +04/18 02:53:28 - mmengine - INFO - Iter(train) [ 3400/160000] lr: 9.8105e-03 eta: 23:55:46 time: 0.5483 data_time: 0.0056 memory: 7635 loss: 0.0381 decode.loss_ce: 0.0239 decode.acc_seg: 98.9027 aux.loss_ce: 0.0143 aux.acc_seg: 98.3586 +04/18 02:53:56 - mmengine - INFO - Iter(train) [ 3450/160000] lr: 9.8077e-03 eta: 23:55:15 time: 0.5497 data_time: 0.0059 memory: 7635 loss: 0.0315 decode.loss_ce: 0.0190 decode.acc_seg: 99.1760 aux.loss_ce: 0.0126 aux.acc_seg: 98.7520 +04/18 02:54:23 - mmengine - INFO - Iter(train) [ 3500/160000] lr: 9.8049e-03 eta: 23:54:45 time: 0.5499 data_time: 0.0064 memory: 7635 loss: 0.0307 decode.loss_ce: 0.0185 decode.acc_seg: 99.1366 aux.loss_ce: 0.0122 aux.acc_seg: 98.8712 +04/18 02:54:51 - mmengine - INFO - Iter(train) [ 3550/160000] lr: 9.8021e-03 eta: 23:54:15 time: 0.5485 data_time: 0.0060 memory: 7635 loss: 0.0334 decode.loss_ce: 0.0206 decode.acc_seg: 98.8063 aux.loss_ce: 0.0128 aux.acc_seg: 98.5059 +04/18 02:55:18 - mmengine - INFO - Iter(train) [ 3600/160000] lr: 9.7994e-03 eta: 23:53:46 time: 0.5482 data_time: 0.0075 memory: 7635 loss: 0.0346 decode.loss_ce: 0.0208 decode.acc_seg: 99.3115 aux.loss_ce: 0.0138 aux.acc_seg: 98.7263 +04/18 02:55:45 - mmengine - INFO - Iter(train) [ 3650/160000] lr: 9.7966e-03 eta: 23:53:18 time: 0.5492 data_time: 0.0069 memory: 7635 loss: 0.0293 decode.loss_ce: 0.0177 decode.acc_seg: 99.4010 aux.loss_ce: 0.0115 aux.acc_seg: 98.9786 +04/18 02:56:13 - mmengine - INFO - Iter(train) [ 3700/160000] lr: 9.7938e-03 eta: 23:52:47 time: 0.5486 data_time: 0.0057 memory: 7635 loss: 0.0324 decode.loss_ce: 0.0198 decode.acc_seg: 99.3436 aux.loss_ce: 0.0126 aux.acc_seg: 98.9145 +04/18 02:56:41 - mmengine - INFO - Iter(train) [ 3750/160000] lr: 9.7910e-03 eta: 23:52:26 time: 0.5494 data_time: 0.0065 memory: 7635 loss: 0.0305 decode.loss_ce: 0.0186 decode.acc_seg: 98.8435 aux.loss_ce: 0.0119 aux.acc_seg: 98.3727 +04/18 02:57:08 - mmengine - INFO - Iter(train) [ 3800/160000] lr: 9.7882e-03 eta: 23:51:57 time: 0.5487 data_time: 0.0059 memory: 7635 loss: 0.0302 decode.loss_ce: 0.0181 decode.acc_seg: 99.4102 aux.loss_ce: 0.0121 aux.acc_seg: 98.9926 +04/18 02:57:35 - mmengine - INFO - Iter(train) [ 3850/160000] lr: 9.7854e-03 eta: 23:51:27 time: 0.5496 data_time: 0.0064 memory: 7635 loss: 0.0317 decode.loss_ce: 0.0197 decode.acc_seg: 99.2857 aux.loss_ce: 0.0120 aux.acc_seg: 98.7676 +04/18 02:58:03 - mmengine - INFO - Iter(train) [ 3900/160000] lr: 9.7826e-03 eta: 23:50:56 time: 0.5485 data_time: 0.0058 memory: 7635 loss: 0.0323 decode.loss_ce: 0.0196 decode.acc_seg: 98.9332 aux.loss_ce: 0.0127 aux.acc_seg: 98.2819 +04/18 02:58:30 - mmengine - INFO - Iter(train) [ 3950/160000] lr: 9.7798e-03 eta: 23:50:26 time: 0.5494 data_time: 0.0066 memory: 7635 loss: 0.0311 decode.loss_ce: 0.0189 decode.acc_seg: 99.3972 aux.loss_ce: 0.0122 aux.acc_seg: 98.9279 +04/18 02:58:58 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 02:58:58 - mmengine - INFO - Iter(train) [ 4000/160000] lr: 9.7770e-03 eta: 23:49:57 time: 0.5496 data_time: 0.0055 memory: 7635 loss: 0.0345 decode.loss_ce: 0.0210 decode.acc_seg: 99.1607 aux.loss_ce: 0.0135 aux.acc_seg: 98.4842 +04/18 02:59:25 - mmengine - INFO - Iter(train) [ 4050/160000] lr: 9.7742e-03 eta: 23:49:28 time: 0.5498 data_time: 0.0063 memory: 7635 loss: 0.0309 decode.loss_ce: 0.0184 decode.acc_seg: 99.1215 aux.loss_ce: 0.0125 aux.acc_seg: 98.7163 +04/18 02:59:53 - mmengine - INFO - Iter(train) [ 4100/160000] lr: 9.7714e-03 eta: 23:48:59 time: 0.5489 data_time: 0.0068 memory: 7635 loss: 0.0299 decode.loss_ce: 0.0181 decode.acc_seg: 99.1440 aux.loss_ce: 0.0118 aux.acc_seg: 98.6404 +04/18 03:00:20 - mmengine - INFO - Iter(train) [ 4150/160000] lr: 9.7686e-03 eta: 23:48:29 time: 0.5491 data_time: 0.0066 memory: 7635 loss: 0.0283 decode.loss_ce: 0.0166 decode.acc_seg: 99.3775 aux.loss_ce: 0.0116 aux.acc_seg: 98.9055 +04/18 03:00:48 - mmengine - INFO - Iter(train) [ 4200/160000] lr: 9.7659e-03 eta: 23:48:00 time: 0.5488 data_time: 0.0059 memory: 7635 loss: 0.0281 decode.loss_ce: 0.0168 decode.acc_seg: 99.5010 aux.loss_ce: 0.0114 aux.acc_seg: 99.1861 +04/18 03:01:15 - mmengine - INFO - Iter(train) [ 4250/160000] lr: 9.7631e-03 eta: 23:47:31 time: 0.5507 data_time: 0.0066 memory: 7635 loss: 0.0352 decode.loss_ce: 0.0211 decode.acc_seg: 99.2168 aux.loss_ce: 0.0141 aux.acc_seg: 98.4035 +04/18 03:01:42 - mmengine - INFO - Iter(train) [ 4300/160000] lr: 9.7603e-03 eta: 23:47:01 time: 0.5481 data_time: 0.0060 memory: 7635 loss: 0.0297 decode.loss_ce: 0.0178 decode.acc_seg: 99.3208 aux.loss_ce: 0.0119 aux.acc_seg: 98.8874 +04/18 03:02:10 - mmengine - INFO - Iter(train) [ 4350/160000] lr: 9.7575e-03 eta: 23:46:31 time: 0.5472 data_time: 0.0059 memory: 7635 loss: 0.0298 decode.loss_ce: 0.0177 decode.acc_seg: 99.3001 aux.loss_ce: 0.0120 aux.acc_seg: 98.7158 +04/18 03:02:37 - mmengine - INFO - Iter(train) [ 4400/160000] lr: 9.7547e-03 eta: 23:46:02 time: 0.5486 data_time: 0.0063 memory: 7635 loss: 0.0314 decode.loss_ce: 0.0191 decode.acc_seg: 99.0993 aux.loss_ce: 0.0123 aux.acc_seg: 98.6669 +04/18 03:03:05 - mmengine - INFO - Iter(train) [ 4450/160000] lr: 9.7519e-03 eta: 23:45:33 time: 0.5493 data_time: 0.0058 memory: 7635 loss: 0.0312 decode.loss_ce: 0.0191 decode.acc_seg: 99.4136 aux.loss_ce: 0.0121 aux.acc_seg: 98.9943 +04/18 03:03:32 - mmengine - INFO - Iter(train) [ 4500/160000] lr: 9.7491e-03 eta: 23:45:04 time: 0.5473 data_time: 0.0061 memory: 7635 loss: 0.0307 decode.loss_ce: 0.0186 decode.acc_seg: 98.9962 aux.loss_ce: 0.0121 aux.acc_seg: 98.5571 +04/18 03:04:00 - mmengine - INFO - Iter(train) [ 4550/160000] lr: 9.7463e-03 eta: 23:44:33 time: 0.5480 data_time: 0.0068 memory: 7635 loss: 0.0311 decode.loss_ce: 0.0189 decode.acc_seg: 99.1083 aux.loss_ce: 0.0122 aux.acc_seg: 98.6803 +04/18 03:04:27 - mmengine - INFO - Iter(train) [ 4600/160000] lr: 9.7435e-03 eta: 23:44:04 time: 0.5486 data_time: 0.0063 memory: 7635 loss: 0.0330 decode.loss_ce: 0.0208 decode.acc_seg: 99.1116 aux.loss_ce: 0.0123 aux.acc_seg: 98.6354 +04/18 03:04:55 - mmengine - INFO - Iter(train) [ 4650/160000] lr: 9.7407e-03 eta: 23:43:34 time: 0.5480 data_time: 0.0057 memory: 7635 loss: 0.0337 decode.loss_ce: 0.0208 decode.acc_seg: 99.4368 aux.loss_ce: 0.0128 aux.acc_seg: 99.0678 +04/18 03:05:22 - mmengine - INFO - Iter(train) [ 4700/160000] lr: 9.7379e-03 eta: 23:43:04 time: 0.5483 data_time: 0.0067 memory: 7635 loss: 0.0294 decode.loss_ce: 0.0175 decode.acc_seg: 99.5893 aux.loss_ce: 0.0119 aux.acc_seg: 99.0988 +04/18 03:05:49 - mmengine - INFO - Iter(train) [ 4750/160000] lr: 9.7351e-03 eta: 23:42:35 time: 0.5488 data_time: 0.0060 memory: 7635 loss: 0.0282 decode.loss_ce: 0.0171 decode.acc_seg: 99.1003 aux.loss_ce: 0.0111 aux.acc_seg: 98.5092 +04/18 03:06:17 - mmengine - INFO - Iter(train) [ 4800/160000] lr: 9.7323e-03 eta: 23:42:12 time: 0.5575 data_time: 0.0061 memory: 7635 loss: 0.0288 decode.loss_ce: 0.0173 decode.acc_seg: 99.3198 aux.loss_ce: 0.0115 aux.acc_seg: 98.9315 +04/18 03:06:44 - mmengine - INFO - Iter(train) [ 4850/160000] lr: 9.7296e-03 eta: 23:41:43 time: 0.5492 data_time: 0.0058 memory: 7635 loss: 0.0287 decode.loss_ce: 0.0169 decode.acc_seg: 99.5131 aux.loss_ce: 0.0118 aux.acc_seg: 98.9789 +04/18 03:07:12 - mmengine - INFO - Iter(train) [ 4900/160000] lr: 9.7268e-03 eta: 23:41:13 time: 0.5480 data_time: 0.0066 memory: 7635 loss: 0.0279 decode.loss_ce: 0.0165 decode.acc_seg: 99.4431 aux.loss_ce: 0.0114 aux.acc_seg: 98.9892 +04/18 03:07:39 - mmengine - INFO - Iter(train) [ 4950/160000] lr: 9.7240e-03 eta: 23:40:44 time: 0.5482 data_time: 0.0063 memory: 7635 loss: 0.0294 decode.loss_ce: 0.0177 decode.acc_seg: 99.3356 aux.loss_ce: 0.0117 aux.acc_seg: 98.8717 +04/18 03:08:07 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 03:08:07 - mmengine - INFO - Iter(train) [ 5000/160000] lr: 9.7212e-03 eta: 23:40:14 time: 0.5493 data_time: 0.0061 memory: 7635 loss: 0.0278 decode.loss_ce: 0.0166 decode.acc_seg: 99.2800 aux.loss_ce: 0.0113 aux.acc_seg: 98.8735 +04/18 03:08:34 - mmengine - INFO - Iter(train) [ 5050/160000] lr: 9.7184e-03 eta: 23:39:44 time: 0.5496 data_time: 0.0063 memory: 7635 loss: 0.0299 decode.loss_ce: 0.0178 decode.acc_seg: 99.3108 aux.loss_ce: 0.0121 aux.acc_seg: 98.7395 +04/18 03:09:02 - mmengine - INFO - Iter(train) [ 5100/160000] lr: 9.7156e-03 eta: 23:39:15 time: 0.5485 data_time: 0.0059 memory: 7635 loss: 0.0287 decode.loss_ce: 0.0170 decode.acc_seg: 99.3157 aux.loss_ce: 0.0117 aux.acc_seg: 98.9011 +04/18 03:09:29 - mmengine - INFO - Iter(train) [ 5150/160000] lr: 9.7128e-03 eta: 23:38:46 time: 0.5492 data_time: 0.0058 memory: 7635 loss: 0.0331 decode.loss_ce: 0.0203 decode.acc_seg: 98.6803 aux.loss_ce: 0.0128 aux.acc_seg: 98.2692 +04/18 03:09:56 - mmengine - INFO - Iter(train) [ 5200/160000] lr: 9.7100e-03 eta: 23:38:18 time: 0.5486 data_time: 0.0059 memory: 7635 loss: 0.0281 decode.loss_ce: 0.0166 decode.acc_seg: 99.4009 aux.loss_ce: 0.0115 aux.acc_seg: 98.8865 +04/18 03:10:24 - mmengine - INFO - Iter(train) [ 5250/160000] lr: 9.7072e-03 eta: 23:37:48 time: 0.5481 data_time: 0.0061 memory: 7635 loss: 0.0294 decode.loss_ce: 0.0179 decode.acc_seg: 99.3403 aux.loss_ce: 0.0116 aux.acc_seg: 98.8310 +04/18 03:10:51 - mmengine - INFO - Iter(train) [ 5300/160000] lr: 9.7044e-03 eta: 23:37:19 time: 0.5487 data_time: 0.0071 memory: 7635 loss: 0.0315 decode.loss_ce: 0.0190 decode.acc_seg: 99.2280 aux.loss_ce: 0.0125 aux.acc_seg: 98.6730 +04/18 03:11:19 - mmengine - INFO - Iter(train) [ 5350/160000] lr: 9.7016e-03 eta: 23:36:50 time: 0.5479 data_time: 0.0060 memory: 7635 loss: 0.0302 decode.loss_ce: 0.0182 decode.acc_seg: 99.4350 aux.loss_ce: 0.0120 aux.acc_seg: 98.9041 +04/18 03:11:46 - mmengine - INFO - Iter(train) [ 5400/160000] lr: 9.6988e-03 eta: 23:36:20 time: 0.5480 data_time: 0.0066 memory: 7635 loss: 0.0296 decode.loss_ce: 0.0179 decode.acc_seg: 99.2143 aux.loss_ce: 0.0117 aux.acc_seg: 98.7977 +04/18 03:12:14 - mmengine - INFO - Iter(train) [ 5450/160000] lr: 9.6960e-03 eta: 23:35:50 time: 0.5475 data_time: 0.0060 memory: 7635 loss: 0.0304 decode.loss_ce: 0.0183 decode.acc_seg: 99.0879 aux.loss_ce: 0.0121 aux.acc_seg: 98.5758 +04/18 03:12:41 - mmengine - INFO - Iter(train) [ 5500/160000] lr: 9.6932e-03 eta: 23:35:20 time: 0.5481 data_time: 0.0062 memory: 7635 loss: 0.0290 decode.loss_ce: 0.0173 decode.acc_seg: 99.0100 aux.loss_ce: 0.0118 aux.acc_seg: 98.5935 +04/18 03:13:08 - mmengine - INFO - Iter(train) [ 5550/160000] lr: 9.6904e-03 eta: 23:34:51 time: 0.5483 data_time: 0.0055 memory: 7635 loss: 0.0260 decode.loss_ce: 0.0151 decode.acc_seg: 99.4328 aux.loss_ce: 0.0109 aux.acc_seg: 98.9295 +04/18 03:13:36 - mmengine - INFO - Iter(train) [ 5600/160000] lr: 9.6877e-03 eta: 23:34:22 time: 0.5495 data_time: 0.0065 memory: 7635 loss: 0.0307 decode.loss_ce: 0.0186 decode.acc_seg: 98.7765 aux.loss_ce: 0.0121 aux.acc_seg: 98.3252 +04/18 03:14:03 - mmengine - INFO - Iter(train) [ 5650/160000] lr: 9.6849e-03 eta: 23:33:53 time: 0.5488 data_time: 0.0062 memory: 7635 loss: 0.0275 decode.loss_ce: 0.0164 decode.acc_seg: 99.4640 aux.loss_ce: 0.0112 aux.acc_seg: 99.0702 +04/18 03:14:31 - mmengine - INFO - Iter(train) [ 5700/160000] lr: 9.6821e-03 eta: 23:33:24 time: 0.5480 data_time: 0.0063 memory: 7635 loss: 0.0277 decode.loss_ce: 0.0167 decode.acc_seg: 99.5169 aux.loss_ce: 0.0110 aux.acc_seg: 99.0870 +04/18 03:14:58 - mmengine - INFO - Iter(train) [ 5750/160000] lr: 9.6793e-03 eta: 23:32:55 time: 0.5487 data_time: 0.0065 memory: 7635 loss: 0.0268 decode.loss_ce: 0.0159 decode.acc_seg: 99.3650 aux.loss_ce: 0.0109 aux.acc_seg: 98.9162 +04/18 03:15:26 - mmengine - INFO - Iter(train) [ 5800/160000] lr: 9.6765e-03 eta: 23:32:26 time: 0.5484 data_time: 0.0058 memory: 7635 loss: 0.0273 decode.loss_ce: 0.0165 decode.acc_seg: 99.4540 aux.loss_ce: 0.0109 aux.acc_seg: 99.1590 +04/18 03:15:53 - mmengine - INFO - Iter(train) [ 5850/160000] lr: 9.6737e-03 eta: 23:31:57 time: 0.5490 data_time: 0.0069 memory: 7635 loss: 0.0270 decode.loss_ce: 0.0156 decode.acc_seg: 99.2741 aux.loss_ce: 0.0114 aux.acc_seg: 98.6178 +04/18 03:16:21 - mmengine - INFO - Iter(train) [ 5900/160000] lr: 9.6709e-03 eta: 23:31:34 time: 0.5472 data_time: 0.0066 memory: 7635 loss: 0.0246 decode.loss_ce: 0.0144 decode.acc_seg: 99.6537 aux.loss_ce: 0.0102 aux.acc_seg: 99.2810 +04/18 03:16:48 - mmengine - INFO - Iter(train) [ 5950/160000] lr: 9.6681e-03 eta: 23:31:03 time: 0.5462 data_time: 0.0064 memory: 7635 loss: 0.0311 decode.loss_ce: 0.0186 decode.acc_seg: 99.5471 aux.loss_ce: 0.0125 aux.acc_seg: 99.0972 +04/18 03:17:15 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 03:17:15 - mmengine - INFO - Iter(train) [ 6000/160000] lr: 9.6653e-03 eta: 23:30:34 time: 0.5482 data_time: 0.0069 memory: 7635 loss: 0.0296 decode.loss_ce: 0.0179 decode.acc_seg: 99.0438 aux.loss_ce: 0.0117 aux.acc_seg: 98.4300 +04/18 03:17:43 - mmengine - INFO - Iter(train) [ 6050/160000] lr: 9.6625e-03 eta: 23:30:05 time: 0.5487 data_time: 0.0067 memory: 7635 loss: 0.0271 decode.loss_ce: 0.0160 decode.acc_seg: 99.4235 aux.loss_ce: 0.0111 aux.acc_seg: 98.7734 +04/18 03:18:10 - mmengine - INFO - Iter(train) [ 6100/160000] lr: 9.6597e-03 eta: 23:29:36 time: 0.5486 data_time: 0.0073 memory: 7635 loss: 0.0248 decode.loss_ce: 0.0142 decode.acc_seg: 99.3794 aux.loss_ce: 0.0106 aux.acc_seg: 98.8412 +04/18 03:18:38 - mmengine - INFO - Iter(train) [ 6150/160000] lr: 9.6569e-03 eta: 23:29:07 time: 0.5479 data_time: 0.0065 memory: 7635 loss: 0.0274 decode.loss_ce: 0.0160 decode.acc_seg: 99.5241 aux.loss_ce: 0.0115 aux.acc_seg: 98.9821 +04/18 03:19:05 - mmengine - INFO - Iter(train) [ 6200/160000] lr: 9.6541e-03 eta: 23:28:39 time: 0.5495 data_time: 0.0062 memory: 7635 loss: 0.0243 decode.loss_ce: 0.0142 decode.acc_seg: 99.4569 aux.loss_ce: 0.0101 aux.acc_seg: 99.0397 +04/18 03:19:32 - mmengine - INFO - Iter(train) [ 6250/160000] lr: 9.6513e-03 eta: 23:28:10 time: 0.5487 data_time: 0.0070 memory: 7635 loss: 0.0260 decode.loss_ce: 0.0156 decode.acc_seg: 99.1268 aux.loss_ce: 0.0105 aux.acc_seg: 98.6498 +04/18 03:20:00 - mmengine - INFO - Iter(train) [ 6300/160000] lr: 9.6485e-03 eta: 23:27:41 time: 0.5485 data_time: 0.0056 memory: 7635 loss: 0.0288 decode.loss_ce: 0.0171 decode.acc_seg: 99.1766 aux.loss_ce: 0.0117 aux.acc_seg: 98.7195 +04/18 03:20:27 - mmengine - INFO - Iter(train) [ 6350/160000] lr: 9.6457e-03 eta: 23:27:11 time: 0.5473 data_time: 0.0059 memory: 7635 loss: 0.0277 decode.loss_ce: 0.0166 decode.acc_seg: 99.3714 aux.loss_ce: 0.0111 aux.acc_seg: 99.0213 +04/18 03:20:55 - mmengine - INFO - Iter(train) [ 6400/160000] lr: 9.6429e-03 eta: 23:26:42 time: 0.5484 data_time: 0.0065 memory: 7635 loss: 0.0238 decode.loss_ce: 0.0140 decode.acc_seg: 99.4395 aux.loss_ce: 0.0098 aux.acc_seg: 98.9776 +04/18 03:21:22 - mmengine - INFO - Iter(train) [ 6450/160000] lr: 9.6401e-03 eta: 23:26:13 time: 0.5465 data_time: 0.0062 memory: 7635 loss: 0.0280 decode.loss_ce: 0.0163 decode.acc_seg: 99.2767 aux.loss_ce: 0.0118 aux.acc_seg: 98.8447 +04/18 03:21:49 - mmengine - INFO - Iter(train) [ 6500/160000] lr: 9.6373e-03 eta: 23:25:43 time: 0.5475 data_time: 0.0065 memory: 7635 loss: 0.0260 decode.loss_ce: 0.0153 decode.acc_seg: 98.8082 aux.loss_ce: 0.0106 aux.acc_seg: 98.3924 +04/18 03:22:17 - mmengine - INFO - Iter(train) [ 6550/160000] lr: 9.6345e-03 eta: 23:25:14 time: 0.5489 data_time: 0.0065 memory: 7635 loss: 0.0227 decode.loss_ce: 0.0132 decode.acc_seg: 99.4503 aux.loss_ce: 0.0095 aux.acc_seg: 99.0750 +04/18 03:22:44 - mmengine - INFO - Iter(train) [ 6600/160000] lr: 9.6317e-03 eta: 23:24:44 time: 0.5475 data_time: 0.0064 memory: 7635 loss: 0.0270 decode.loss_ce: 0.0161 decode.acc_seg: 98.9626 aux.loss_ce: 0.0109 aux.acc_seg: 98.5077 +04/18 03:23:12 - mmengine - INFO - Iter(train) [ 6650/160000] lr: 9.6290e-03 eta: 23:24:15 time: 0.5468 data_time: 0.0059 memory: 7635 loss: 0.0272 decode.loss_ce: 0.0159 decode.acc_seg: 99.0431 aux.loss_ce: 0.0113 aux.acc_seg: 98.3189 +04/18 03:23:39 - mmengine - INFO - Iter(train) [ 6700/160000] lr: 9.6262e-03 eta: 23:23:46 time: 0.5487 data_time: 0.0060 memory: 7635 loss: 0.0306 decode.loss_ce: 0.0186 decode.acc_seg: 99.2869 aux.loss_ce: 0.0120 aux.acc_seg: 98.6516 +04/18 03:24:06 - mmengine - INFO - Iter(train) [ 6750/160000] lr: 9.6234e-03 eta: 23:23:17 time: 0.5487 data_time: 0.0064 memory: 7635 loss: 0.0287 decode.loss_ce: 0.0174 decode.acc_seg: 99.5493 aux.loss_ce: 0.0113 aux.acc_seg: 99.0694 +04/18 03:24:34 - mmengine - INFO - Iter(train) [ 6800/160000] lr: 9.6206e-03 eta: 23:22:49 time: 0.5477 data_time: 0.0062 memory: 7635 loss: 0.0279 decode.loss_ce: 0.0163 decode.acc_seg: 99.2764 aux.loss_ce: 0.0116 aux.acc_seg: 98.6606 +04/18 03:25:01 - mmengine - INFO - Iter(train) [ 6850/160000] lr: 9.6178e-03 eta: 23:22:20 time: 0.5486 data_time: 0.0058 memory: 7635 loss: 0.0270 decode.loss_ce: 0.0162 decode.acc_seg: 99.4166 aux.loss_ce: 0.0108 aux.acc_seg: 98.8656 +04/18 03:25:29 - mmengine - INFO - Iter(train) [ 6900/160000] lr: 9.6150e-03 eta: 23:21:51 time: 0.5471 data_time: 0.0054 memory: 7635 loss: 0.0233 decode.loss_ce: 0.0134 decode.acc_seg: 99.4481 aux.loss_ce: 0.0099 aux.acc_seg: 98.8849 +04/18 03:25:56 - mmengine - INFO - Iter(train) [ 6950/160000] lr: 9.6122e-03 eta: 23:21:26 time: 0.5485 data_time: 0.0058 memory: 7635 loss: 0.0272 decode.loss_ce: 0.0164 decode.acc_seg: 99.3545 aux.loss_ce: 0.0108 aux.acc_seg: 99.0042 +04/18 03:26:24 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 03:26:24 - mmengine - INFO - Iter(train) [ 7000/160000] lr: 9.6094e-03 eta: 23:20:57 time: 0.5484 data_time: 0.0065 memory: 7635 loss: 0.0262 decode.loss_ce: 0.0153 decode.acc_seg: 99.1550 aux.loss_ce: 0.0109 aux.acc_seg: 98.4638 +04/18 03:26:51 - mmengine - INFO - Iter(train) [ 7050/160000] lr: 9.6066e-03 eta: 23:20:28 time: 0.5472 data_time: 0.0060 memory: 7635 loss: 0.0252 decode.loss_ce: 0.0146 decode.acc_seg: 99.2325 aux.loss_ce: 0.0106 aux.acc_seg: 98.6634 +04/18 03:27:19 - mmengine - INFO - Iter(train) [ 7100/160000] lr: 9.6038e-03 eta: 23:20:00 time: 0.5488 data_time: 0.0067 memory: 7635 loss: 0.0264 decode.loss_ce: 0.0153 decode.acc_seg: 99.6102 aux.loss_ce: 0.0111 aux.acc_seg: 99.2232 +04/18 03:27:46 - mmengine - INFO - Iter(train) [ 7150/160000] lr: 9.6010e-03 eta: 23:19:31 time: 0.5486 data_time: 0.0063 memory: 7635 loss: 0.0264 decode.loss_ce: 0.0157 decode.acc_seg: 99.2992 aux.loss_ce: 0.0107 aux.acc_seg: 98.8308 +04/18 03:28:13 - mmengine - INFO - Iter(train) [ 7200/160000] lr: 9.5982e-03 eta: 23:19:02 time: 0.5487 data_time: 0.0066 memory: 7635 loss: 0.0250 decode.loss_ce: 0.0145 decode.acc_seg: 99.5803 aux.loss_ce: 0.0104 aux.acc_seg: 99.1144 +04/18 03:28:41 - mmengine - INFO - Iter(train) [ 7250/160000] lr: 9.5954e-03 eta: 23:18:33 time: 0.5475 data_time: 0.0060 memory: 7635 loss: 0.0232 decode.loss_ce: 0.0135 decode.acc_seg: 99.2552 aux.loss_ce: 0.0097 aux.acc_seg: 98.7329 +04/18 03:29:08 - mmengine - INFO - Iter(train) [ 7300/160000] lr: 9.5926e-03 eta: 23:18:05 time: 0.5490 data_time: 0.0067 memory: 7635 loss: 0.0291 decode.loss_ce: 0.0174 decode.acc_seg: 99.5982 aux.loss_ce: 0.0116 aux.acc_seg: 99.2081 +04/18 03:29:36 - mmengine - INFO - Iter(train) [ 7350/160000] lr: 9.5898e-03 eta: 23:17:36 time: 0.5472 data_time: 0.0069 memory: 7635 loss: 0.0245 decode.loss_ce: 0.0144 decode.acc_seg: 99.4645 aux.loss_ce: 0.0101 aux.acc_seg: 98.9813 +04/18 03:30:03 - mmengine - INFO - Iter(train) [ 7400/160000] lr: 9.5870e-03 eta: 23:17:07 time: 0.5478 data_time: 0.0062 memory: 7635 loss: 0.0235 decode.loss_ce: 0.0137 decode.acc_seg: 99.5255 aux.loss_ce: 0.0098 aux.acc_seg: 98.9691 +04/18 03:30:30 - mmengine - INFO - Iter(train) [ 7450/160000] lr: 9.5842e-03 eta: 23:16:38 time: 0.5476 data_time: 0.0056 memory: 7635 loss: 0.0258 decode.loss_ce: 0.0151 decode.acc_seg: 99.4922 aux.loss_ce: 0.0107 aux.acc_seg: 98.9164 +04/18 03:30:58 - mmengine - INFO - Iter(train) [ 7500/160000] lr: 9.5814e-03 eta: 23:16:09 time: 0.5492 data_time: 0.0056 memory: 7635 loss: 0.0243 decode.loss_ce: 0.0143 decode.acc_seg: 99.3121 aux.loss_ce: 0.0100 aux.acc_seg: 98.9280 +04/18 03:31:25 - mmengine - INFO - Iter(train) [ 7550/160000] lr: 9.5786e-03 eta: 23:15:40 time: 0.5472 data_time: 0.0057 memory: 7635 loss: 0.0219 decode.loss_ce: 0.0126 decode.acc_seg: 99.5523 aux.loss_ce: 0.0093 aux.acc_seg: 99.1000 +04/18 03:31:53 - mmengine - INFO - Iter(train) [ 7600/160000] lr: 9.5758e-03 eta: 23:15:12 time: 0.5488 data_time: 0.0066 memory: 7635 loss: 0.0246 decode.loss_ce: 0.0144 decode.acc_seg: 99.3397 aux.loss_ce: 0.0102 aux.acc_seg: 98.9182 +04/18 03:32:20 - mmengine - INFO - Iter(train) [ 7650/160000] lr: 9.5730e-03 eta: 23:14:43 time: 0.5485 data_time: 0.0061 memory: 7635 loss: 0.0265 decode.loss_ce: 0.0158 decode.acc_seg: 99.3401 aux.loss_ce: 0.0107 aux.acc_seg: 98.8952 +04/18 03:32:47 - mmengine - INFO - Iter(train) [ 7700/160000] lr: 9.5702e-03 eta: 23:14:16 time: 0.5497 data_time: 0.0062 memory: 7635 loss: 0.0237 decode.loss_ce: 0.0136 decode.acc_seg: 99.4659 aux.loss_ce: 0.0101 aux.acc_seg: 98.9071 +04/18 03:33:15 - mmengine - INFO - Iter(train) [ 7750/160000] lr: 9.5674e-03 eta: 23:13:47 time: 0.5481 data_time: 0.0063 memory: 7635 loss: 0.0251 decode.loss_ce: 0.0149 decode.acc_seg: 99.3489 aux.loss_ce: 0.0101 aux.acc_seg: 98.9504 +04/18 03:33:42 - mmengine - INFO - Iter(train) [ 7800/160000] lr: 9.5646e-03 eta: 23:13:18 time: 0.5488 data_time: 0.0061 memory: 7635 loss: 0.0258 decode.loss_ce: 0.0152 decode.acc_seg: 99.0079 aux.loss_ce: 0.0105 aux.acc_seg: 98.6009 +04/18 03:34:10 - mmengine - INFO - Iter(train) [ 7850/160000] lr: 9.5618e-03 eta: 23:12:49 time: 0.5474 data_time: 0.0061 memory: 7635 loss: 0.0212 decode.loss_ce: 0.0120 decode.acc_seg: 99.4060 aux.loss_ce: 0.0092 aux.acc_seg: 98.8746 +04/18 03:34:37 - mmengine - INFO - Iter(train) [ 7900/160000] lr: 9.5590e-03 eta: 23:12:20 time: 0.5460 data_time: 0.0062 memory: 7635 loss: 0.0242 decode.loss_ce: 0.0140 decode.acc_seg: 99.5603 aux.loss_ce: 0.0102 aux.acc_seg: 99.1700 +04/18 03:35:04 - mmengine - INFO - Iter(train) [ 7950/160000] lr: 9.5562e-03 eta: 23:11:52 time: 0.5482 data_time: 0.0066 memory: 7635 loss: 0.0298 decode.loss_ce: 0.0182 decode.acc_seg: 99.1676 aux.loss_ce: 0.0115 aux.acc_seg: 98.7913 +04/18 03:35:32 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 03:35:32 - mmengine - INFO - Iter(train) [ 8000/160000] lr: 9.5534e-03 eta: 23:11:25 time: 0.5589 data_time: 0.0057 memory: 7635 loss: 0.0228 decode.loss_ce: 0.0131 decode.acc_seg: 99.5768 aux.loss_ce: 0.0096 aux.acc_seg: 99.0025 +04/18 03:35:59 - mmengine - INFO - Iter(train) [ 8050/160000] lr: 9.5506e-03 eta: 23:10:59 time: 0.5485 data_time: 0.0062 memory: 7635 loss: 0.0239 decode.loss_ce: 0.0138 decode.acc_seg: 99.4815 aux.loss_ce: 0.0101 aux.acc_seg: 99.1381 +04/18 03:36:27 - mmengine - INFO - Iter(train) [ 8100/160000] lr: 9.5478e-03 eta: 23:10:30 time: 0.5485 data_time: 0.0061 memory: 7635 loss: 0.0244 decode.loss_ce: 0.0143 decode.acc_seg: 99.3567 aux.loss_ce: 0.0101 aux.acc_seg: 98.9232 +04/18 03:36:54 - mmengine - INFO - Iter(train) [ 8150/160000] lr: 9.5450e-03 eta: 23:10:03 time: 0.5492 data_time: 0.0059 memory: 7635 loss: 0.0249 decode.loss_ce: 0.0142 decode.acc_seg: 99.4530 aux.loss_ce: 0.0107 aux.acc_seg: 98.9367 +04/18 03:37:22 - mmengine - INFO - Iter(train) [ 8200/160000] lr: 9.5422e-03 eta: 23:09:35 time: 0.5485 data_time: 0.0064 memory: 7635 loss: 0.0246 decode.loss_ce: 0.0140 decode.acc_seg: 99.4133 aux.loss_ce: 0.0106 aux.acc_seg: 98.7640 +04/18 03:37:49 - mmengine - INFO - Iter(train) [ 8250/160000] lr: 9.5394e-03 eta: 23:09:06 time: 0.5481 data_time: 0.0059 memory: 7635 loss: 0.0251 decode.loss_ce: 0.0143 decode.acc_seg: 99.2445 aux.loss_ce: 0.0109 aux.acc_seg: 98.7236 +04/18 03:38:17 - mmengine - INFO - Iter(train) [ 8300/160000] lr: 9.5366e-03 eta: 23:08:37 time: 0.5480 data_time: 0.0059 memory: 7635 loss: 0.0239 decode.loss_ce: 0.0136 decode.acc_seg: 99.4275 aux.loss_ce: 0.0103 aux.acc_seg: 98.7861 +04/18 03:38:44 - mmengine - INFO - Iter(train) [ 8350/160000] lr: 9.5338e-03 eta: 23:08:08 time: 0.5482 data_time: 0.0070 memory: 7635 loss: 0.0264 decode.loss_ce: 0.0158 decode.acc_seg: 99.3280 aux.loss_ce: 0.0106 aux.acc_seg: 99.0131 +04/18 03:39:11 - mmengine - INFO - Iter(train) [ 8400/160000] lr: 9.5310e-03 eta: 23:07:40 time: 0.5491 data_time: 0.0057 memory: 7635 loss: 0.0277 decode.loss_ce: 0.0164 decode.acc_seg: 99.3407 aux.loss_ce: 0.0113 aux.acc_seg: 98.7556 +04/18 03:39:39 - mmengine - INFO - Iter(train) [ 8450/160000] lr: 9.5282e-03 eta: 23:07:11 time: 0.5479 data_time: 0.0067 memory: 7635 loss: 0.0247 decode.loss_ce: 0.0143 decode.acc_seg: 99.2147 aux.loss_ce: 0.0104 aux.acc_seg: 98.7143 +04/18 03:40:06 - mmengine - INFO - Iter(train) [ 8500/160000] lr: 9.5254e-03 eta: 23:06:43 time: 0.5479 data_time: 0.0066 memory: 7635 loss: 0.0243 decode.loss_ce: 0.0140 decode.acc_seg: 99.5707 aux.loss_ce: 0.0103 aux.acc_seg: 99.1184 +04/18 03:40:34 - mmengine - INFO - Iter(train) [ 8550/160000] lr: 9.5226e-03 eta: 23:06:14 time: 0.5482 data_time: 0.0063 memory: 7635 loss: 0.0246 decode.loss_ce: 0.0145 decode.acc_seg: 99.4205 aux.loss_ce: 0.0101 aux.acc_seg: 99.0105 +04/18 03:41:01 - mmengine - INFO - Iter(train) [ 8600/160000] lr: 9.5198e-03 eta: 23:05:46 time: 0.5482 data_time: 0.0064 memory: 7635 loss: 0.0248 decode.loss_ce: 0.0146 decode.acc_seg: 99.4130 aux.loss_ce: 0.0102 aux.acc_seg: 98.9951 +04/18 03:41:28 - mmengine - INFO - Iter(train) [ 8650/160000] lr: 9.5170e-03 eta: 23:05:17 time: 0.5483 data_time: 0.0064 memory: 7635 loss: 0.0276 decode.loss_ce: 0.0165 decode.acc_seg: 99.3017 aux.loss_ce: 0.0111 aux.acc_seg: 98.7202 +04/18 03:41:56 - mmengine - INFO - Iter(train) [ 8700/160000] lr: 9.5142e-03 eta: 23:04:49 time: 0.5479 data_time: 0.0062 memory: 7635 loss: 0.0253 decode.loss_ce: 0.0152 decode.acc_seg: 99.3296 aux.loss_ce: 0.0101 aux.acc_seg: 98.8963 +04/18 03:42:23 - mmengine - INFO - Iter(train) [ 8750/160000] lr: 9.5114e-03 eta: 23:04:21 time: 0.5482 data_time: 0.0061 memory: 7635 loss: 0.0240 decode.loss_ce: 0.0138 decode.acc_seg: 98.9489 aux.loss_ce: 0.0102 aux.acc_seg: 98.0512 +04/18 03:42:51 - mmengine - INFO - Iter(train) [ 8800/160000] lr: 9.5086e-03 eta: 23:03:52 time: 0.5478 data_time: 0.0057 memory: 7635 loss: 0.0225 decode.loss_ce: 0.0129 decode.acc_seg: 99.4946 aux.loss_ce: 0.0096 aux.acc_seg: 99.1075 +04/18 03:43:18 - mmengine - INFO - Iter(train) [ 8850/160000] lr: 9.5058e-03 eta: 23:03:23 time: 0.5477 data_time: 0.0064 memory: 7635 loss: 0.0240 decode.loss_ce: 0.0139 decode.acc_seg: 99.2327 aux.loss_ce: 0.0100 aux.acc_seg: 98.6191 +04/18 03:43:45 - mmengine - INFO - Iter(train) [ 8900/160000] lr: 9.5030e-03 eta: 23:02:55 time: 0.5486 data_time: 0.0062 memory: 7635 loss: 0.0227 decode.loss_ce: 0.0128 decode.acc_seg: 99.4642 aux.loss_ce: 0.0099 aux.acc_seg: 99.0090 +04/18 03:44:13 - mmengine - INFO - Iter(train) [ 8950/160000] lr: 9.5002e-03 eta: 23:02:27 time: 0.5496 data_time: 0.0064 memory: 7635 loss: 0.0220 decode.loss_ce: 0.0126 decode.acc_seg: 99.5750 aux.loss_ce: 0.0095 aux.acc_seg: 99.3172 +04/18 03:44:40 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 03:44:40 - mmengine - INFO - Iter(train) [ 9000/160000] lr: 9.4974e-03 eta: 23:02:00 time: 0.5492 data_time: 0.0061 memory: 7635 loss: 0.0223 decode.loss_ce: 0.0130 decode.acc_seg: 99.4808 aux.loss_ce: 0.0094 aux.acc_seg: 98.8997 +04/18 03:45:08 - mmengine - INFO - Iter(train) [ 9050/160000] lr: 9.4946e-03 eta: 23:01:32 time: 0.5477 data_time: 0.0061 memory: 7635 loss: 0.0222 decode.loss_ce: 0.0129 decode.acc_seg: 99.4164 aux.loss_ce: 0.0093 aux.acc_seg: 98.9769 +04/18 03:45:35 - mmengine - INFO - Iter(train) [ 9100/160000] lr: 9.4918e-03 eta: 23:01:07 time: 0.5567 data_time: 0.0059 memory: 7635 loss: 0.0238 decode.loss_ce: 0.0135 decode.acc_seg: 99.5447 aux.loss_ce: 0.0102 aux.acc_seg: 99.0881 +04/18 03:46:03 - mmengine - INFO - Iter(train) [ 9150/160000] lr: 9.4890e-03 eta: 23:00:38 time: 0.5485 data_time: 0.0065 memory: 7635 loss: 0.0246 decode.loss_ce: 0.0141 decode.acc_seg: 99.6255 aux.loss_ce: 0.0104 aux.acc_seg: 99.1031 +04/18 03:46:30 - mmengine - INFO - Iter(train) [ 9200/160000] lr: 9.4862e-03 eta: 23:00:10 time: 0.5483 data_time: 0.0057 memory: 7635 loss: 0.0230 decode.loss_ce: 0.0131 decode.acc_seg: 99.3305 aux.loss_ce: 0.0099 aux.acc_seg: 98.7550 +04/18 03:46:57 - mmengine - INFO - Iter(train) [ 9250/160000] lr: 9.4834e-03 eta: 22:59:42 time: 0.5492 data_time: 0.0058 memory: 7635 loss: 0.0222 decode.loss_ce: 0.0125 decode.acc_seg: 99.6259 aux.loss_ce: 0.0097 aux.acc_seg: 99.1253 +04/18 03:47:25 - mmengine - INFO - Iter(train) [ 9300/160000] lr: 9.4806e-03 eta: 22:59:14 time: 0.5475 data_time: 0.0067 memory: 7635 loss: 0.0248 decode.loss_ce: 0.0144 decode.acc_seg: 99.5245 aux.loss_ce: 0.0104 aux.acc_seg: 98.9099 +04/18 03:47:52 - mmengine - INFO - Iter(train) [ 9350/160000] lr: 9.4778e-03 eta: 22:58:46 time: 0.5485 data_time: 0.0058 memory: 7635 loss: 0.0215 decode.loss_ce: 0.0121 decode.acc_seg: 99.5957 aux.loss_ce: 0.0094 aux.acc_seg: 99.1341 +04/18 03:48:20 - mmengine - INFO - Iter(train) [ 9400/160000] lr: 9.4750e-03 eta: 22:58:18 time: 0.5485 data_time: 0.0064 memory: 7635 loss: 0.0239 decode.loss_ce: 0.0140 decode.acc_seg: 99.3546 aux.loss_ce: 0.0099 aux.acc_seg: 98.8052 +04/18 03:48:47 - mmengine - INFO - Iter(train) [ 9450/160000] lr: 9.4722e-03 eta: 22:57:50 time: 0.5483 data_time: 0.0065 memory: 7635 loss: 0.0260 decode.loss_ce: 0.0153 decode.acc_seg: 99.1859 aux.loss_ce: 0.0108 aux.acc_seg: 98.5625 +04/18 03:49:15 - mmengine - INFO - Iter(train) [ 9500/160000] lr: 9.4694e-03 eta: 22:57:22 time: 0.5482 data_time: 0.0062 memory: 7635 loss: 0.0219 decode.loss_ce: 0.0127 decode.acc_seg: 99.4252 aux.loss_ce: 0.0092 aux.acc_seg: 98.7907 +04/18 03:49:42 - mmengine - INFO - Iter(train) [ 9550/160000] lr: 9.4666e-03 eta: 22:56:54 time: 0.5474 data_time: 0.0068 memory: 7635 loss: 0.0230 decode.loss_ce: 0.0130 decode.acc_seg: 99.5365 aux.loss_ce: 0.0100 aux.acc_seg: 98.8636 +04/18 03:50:09 - mmengine - INFO - Iter(train) [ 9600/160000] lr: 9.4638e-03 eta: 22:56:26 time: 0.5483 data_time: 0.0063 memory: 7635 loss: 0.0244 decode.loss_ce: 0.0140 decode.acc_seg: 99.4267 aux.loss_ce: 0.0104 aux.acc_seg: 98.9110 +04/18 03:50:37 - mmengine - INFO - Iter(train) [ 9650/160000] lr: 9.4610e-03 eta: 22:55:58 time: 0.5474 data_time: 0.0058 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0118 decode.acc_seg: 99.5340 aux.loss_ce: 0.0089 aux.acc_seg: 99.1682 +04/18 03:51:04 - mmengine - INFO - Iter(train) [ 9700/160000] lr: 9.4582e-03 eta: 22:55:31 time: 0.5489 data_time: 0.0068 memory: 7635 loss: 0.0216 decode.loss_ce: 0.0123 decode.acc_seg: 99.4224 aux.loss_ce: 0.0093 aux.acc_seg: 98.6237 +04/18 03:51:32 - mmengine - INFO - Iter(train) [ 9750/160000] lr: 9.4554e-03 eta: 22:55:03 time: 0.5488 data_time: 0.0058 memory: 7635 loss: 0.0211 decode.loss_ce: 0.0120 decode.acc_seg: 99.6173 aux.loss_ce: 0.0091 aux.acc_seg: 99.2518 +04/18 03:51:59 - mmengine - INFO - Iter(train) [ 9800/160000] lr: 9.4526e-03 eta: 22:54:34 time: 0.5483 data_time: 0.0066 memory: 7635 loss: 0.0229 decode.loss_ce: 0.0133 decode.acc_seg: 99.6406 aux.loss_ce: 0.0096 aux.acc_seg: 99.3035 +04/18 03:52:27 - mmengine - INFO - Iter(train) [ 9850/160000] lr: 9.4498e-03 eta: 22:54:06 time: 0.5480 data_time: 0.0064 memory: 7635 loss: 0.0220 decode.loss_ce: 0.0125 decode.acc_seg: 99.4288 aux.loss_ce: 0.0094 aux.acc_seg: 98.8976 +04/18 03:52:54 - mmengine - INFO - Iter(train) [ 9900/160000] lr: 9.4470e-03 eta: 22:53:38 time: 0.5480 data_time: 0.0068 memory: 7635 loss: 0.0232 decode.loss_ce: 0.0132 decode.acc_seg: 99.6505 aux.loss_ce: 0.0099 aux.acc_seg: 99.2436 +04/18 03:53:21 - mmengine - INFO - Iter(train) [ 9950/160000] lr: 9.4442e-03 eta: 22:53:10 time: 0.5498 data_time: 0.0057 memory: 7635 loss: 0.0249 decode.loss_ce: 0.0142 decode.acc_seg: 99.4144 aux.loss_ce: 0.0106 aux.acc_seg: 98.5002 +04/18 03:53:49 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 03:53:49 - mmengine - INFO - Iter(train) [ 10000/160000] lr: 9.4414e-03 eta: 22:52:42 time: 0.5481 data_time: 0.0058 memory: 7635 loss: 0.0230 decode.loss_ce: 0.0132 decode.acc_seg: 99.3968 aux.loss_ce: 0.0098 aux.acc_seg: 99.0562 +04/18 03:53:49 - mmengine - INFO - Saving checkpoint at 10000 iterations +04/18 03:53:54 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:10 time: 0.0475 data_time: 0.0015 memory: 5542 +04/18 03:53:56 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:05 time: 0.0473 data_time: 0.0014 memory: 1657 +04/18 03:53:59 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:02 time: 0.0462 data_time: 0.0015 memory: 1657 +04/18 03:54:01 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.0450 data_time: 0.0012 memory: 1657 +04/18 03:54:01 - mmengine - INFO - per class results: +04/18 03:54:01 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 98.95 | 99.35 | 99.47 | 99.6 | 99.35 | +| contrast | 78.15 | 90.4 | 87.74 | 85.23 | 90.4 | ++------------+-------+-------+--------+-----------+--------+ +04/18 03:54:01 - mmengine - INFO - Iter(val) [200/200] aAcc: 98.9900 mIoU: 88.5500 mAcc: 94.8700 mFscore: 93.6100 mPrecision: 92.4200 mRecall: 94.8700 data_time: 0.0023 time: 0.0518 +04/18 03:54:29 - mmengine - INFO - Iter(train) [ 10050/160000] lr: 9.4386e-03 eta: 22:52:17 time: 0.5487 data_time: 0.0069 memory: 7635 loss: 0.0211 decode.loss_ce: 0.0122 decode.acc_seg: 99.3901 aux.loss_ce: 0.0089 aux.acc_seg: 98.9715 +04/18 03:54:56 - mmengine - INFO - Iter(train) [ 10100/160000] lr: 9.4358e-03 eta: 22:51:49 time: 0.5489 data_time: 0.0062 memory: 7635 loss: 0.0208 decode.loss_ce: 0.0118 decode.acc_seg: 99.5948 aux.loss_ce: 0.0090 aux.acc_seg: 99.1281 +04/18 03:55:24 - mmengine - INFO - Iter(train) [ 10150/160000] lr: 9.4330e-03 eta: 22:51:24 time: 0.5693 data_time: 0.0059 memory: 7635 loss: 0.0214 decode.loss_ce: 0.0120 decode.acc_seg: 99.5793 aux.loss_ce: 0.0094 aux.acc_seg: 99.0387 +04/18 03:55:51 - mmengine - INFO - Iter(train) [ 10200/160000] lr: 9.4302e-03 eta: 22:50:55 time: 0.5478 data_time: 0.0067 memory: 7635 loss: 0.0213 decode.loss_ce: 0.0122 decode.acc_seg: 99.4923 aux.loss_ce: 0.0092 aux.acc_seg: 98.9777 +04/18 03:56:18 - mmengine - INFO - Iter(train) [ 10250/160000] lr: 9.4274e-03 eta: 22:50:27 time: 0.5491 data_time: 0.0055 memory: 7635 loss: 0.0257 decode.loss_ce: 0.0153 decode.acc_seg: 98.8754 aux.loss_ce: 0.0104 aux.acc_seg: 98.3413 +04/18 03:56:46 - mmengine - INFO - Iter(train) [ 10300/160000] lr: 9.4246e-03 eta: 22:50:00 time: 0.5485 data_time: 0.0058 memory: 7635 loss: 0.0215 decode.loss_ce: 0.0124 decode.acc_seg: 99.2995 aux.loss_ce: 0.0091 aux.acc_seg: 98.8486 +04/18 03:57:13 - mmengine - INFO - Iter(train) [ 10350/160000] lr: 9.4218e-03 eta: 22:49:31 time: 0.5486 data_time: 0.0059 memory: 7635 loss: 0.0210 decode.loss_ce: 0.0121 decode.acc_seg: 99.4930 aux.loss_ce: 0.0089 aux.acc_seg: 98.9901 +04/18 03:57:41 - mmengine - INFO - Iter(train) [ 10400/160000] lr: 9.4190e-03 eta: 22:49:03 time: 0.5489 data_time: 0.0062 memory: 7635 loss: 0.0225 decode.loss_ce: 0.0127 decode.acc_seg: 99.5553 aux.loss_ce: 0.0098 aux.acc_seg: 99.2027 +04/18 03:58:08 - mmengine - INFO - Iter(train) [ 10450/160000] lr: 9.4162e-03 eta: 22:48:35 time: 0.5489 data_time: 0.0060 memory: 7635 loss: 0.0230 decode.loss_ce: 0.0133 decode.acc_seg: 99.4981 aux.loss_ce: 0.0098 aux.acc_seg: 99.0229 +04/18 03:58:36 - mmengine - INFO - Iter(train) [ 10500/160000] lr: 9.4134e-03 eta: 22:48:09 time: 0.5479 data_time: 0.0055 memory: 7635 loss: 0.0215 decode.loss_ce: 0.0122 decode.acc_seg: 99.5740 aux.loss_ce: 0.0094 aux.acc_seg: 99.2808 +04/18 03:59:03 - mmengine - INFO - Iter(train) [ 10550/160000] lr: 9.4106e-03 eta: 22:47:41 time: 0.5484 data_time: 0.0060 memory: 7635 loss: 0.0215 decode.loss_ce: 0.0124 decode.acc_seg: 99.5886 aux.loss_ce: 0.0091 aux.acc_seg: 99.0820 +04/18 03:59:30 - mmengine - INFO - Iter(train) [ 10600/160000] lr: 9.4078e-03 eta: 22:47:12 time: 0.5491 data_time: 0.0067 memory: 7635 loss: 0.0223 decode.loss_ce: 0.0129 decode.acc_seg: 99.5251 aux.loss_ce: 0.0094 aux.acc_seg: 99.0472 +04/18 03:59:58 - mmengine - INFO - Iter(train) [ 10650/160000] lr: 9.4050e-03 eta: 22:46:45 time: 0.5491 data_time: 0.0063 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0110 decode.acc_seg: 99.5851 aux.loss_ce: 0.0085 aux.acc_seg: 99.1895 +04/18 04:00:25 - mmengine - INFO - Iter(train) [ 10700/160000] lr: 9.4022e-03 eta: 22:46:17 time: 0.5487 data_time: 0.0068 memory: 7635 loss: 0.0242 decode.loss_ce: 0.0137 decode.acc_seg: 99.5319 aux.loss_ce: 0.0105 aux.acc_seg: 98.8897 +04/18 04:00:53 - mmengine - INFO - Iter(train) [ 10750/160000] lr: 9.3993e-03 eta: 22:45:49 time: 0.5480 data_time: 0.0059 memory: 7635 loss: 0.0221 decode.loss_ce: 0.0127 decode.acc_seg: 99.5176 aux.loss_ce: 0.0094 aux.acc_seg: 99.0525 +04/18 04:01:20 - mmengine - INFO - Iter(train) [ 10800/160000] lr: 9.3965e-03 eta: 22:45:21 time: 0.5488 data_time: 0.0066 memory: 7635 loss: 0.0220 decode.loss_ce: 0.0124 decode.acc_seg: 99.5129 aux.loss_ce: 0.0096 aux.acc_seg: 98.9785 +04/18 04:01:48 - mmengine - INFO - Iter(train) [ 10850/160000] lr: 9.3937e-03 eta: 22:44:54 time: 0.5498 data_time: 0.0065 memory: 7635 loss: 0.0226 decode.loss_ce: 0.0131 decode.acc_seg: 99.6201 aux.loss_ce: 0.0095 aux.acc_seg: 99.2350 +04/18 04:02:15 - mmengine - INFO - Iter(train) [ 10900/160000] lr: 9.3909e-03 eta: 22:44:26 time: 0.5482 data_time: 0.0068 memory: 7635 loss: 0.0211 decode.loss_ce: 0.0119 decode.acc_seg: 99.4088 aux.loss_ce: 0.0092 aux.acc_seg: 99.0273 +04/18 04:02:42 - mmengine - INFO - Iter(train) [ 10950/160000] lr: 9.3881e-03 eta: 22:43:58 time: 0.5486 data_time: 0.0064 memory: 7635 loss: 0.0201 decode.loss_ce: 0.0112 decode.acc_seg: 99.6139 aux.loss_ce: 0.0089 aux.acc_seg: 99.1864 +04/18 04:03:10 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 04:03:10 - mmengine - INFO - Iter(train) [ 11000/160000] lr: 9.3853e-03 eta: 22:43:30 time: 0.5479 data_time: 0.0064 memory: 7635 loss: 0.0235 decode.loss_ce: 0.0134 decode.acc_seg: 99.5029 aux.loss_ce: 0.0101 aux.acc_seg: 98.9841 +04/18 04:03:37 - mmengine - INFO - Iter(train) [ 11050/160000] lr: 9.3825e-03 eta: 22:43:03 time: 0.5497 data_time: 0.0067 memory: 7635 loss: 0.0235 decode.loss_ce: 0.0134 decode.acc_seg: 99.4128 aux.loss_ce: 0.0101 aux.acc_seg: 99.0168 +04/18 04:04:05 - mmengine - INFO - Iter(train) [ 11100/160000] lr: 9.3797e-03 eta: 22:42:35 time: 0.5491 data_time: 0.0060 memory: 7635 loss: 0.0225 decode.loss_ce: 0.0129 decode.acc_seg: 99.6453 aux.loss_ce: 0.0096 aux.acc_seg: 99.2537 +04/18 04:04:32 - mmengine - INFO - Iter(train) [ 11150/160000] lr: 9.3769e-03 eta: 22:42:07 time: 0.5487 data_time: 0.0060 memory: 7635 loss: 0.0259 decode.loss_ce: 0.0149 decode.acc_seg: 99.3362 aux.loss_ce: 0.0110 aux.acc_seg: 98.9176 +04/18 04:05:00 - mmengine - INFO - Iter(train) [ 11200/160000] lr: 9.3741e-03 eta: 22:41:40 time: 0.5480 data_time: 0.0065 memory: 7635 loss: 0.0226 decode.loss_ce: 0.0128 decode.acc_seg: 99.4145 aux.loss_ce: 0.0098 aux.acc_seg: 98.8956 +04/18 04:05:27 - mmengine - INFO - Iter(train) [ 11250/160000] lr: 9.3713e-03 eta: 22:41:14 time: 0.5502 data_time: 0.0061 memory: 7635 loss: 0.0227 decode.loss_ce: 0.0132 decode.acc_seg: 99.4885 aux.loss_ce: 0.0095 aux.acc_seg: 99.0182 +04/18 04:05:55 - mmengine - INFO - Iter(train) [ 11300/160000] lr: 9.3685e-03 eta: 22:40:47 time: 0.5487 data_time: 0.0064 memory: 7635 loss: 0.0200 decode.loss_ce: 0.0111 decode.acc_seg: 99.3999 aux.loss_ce: 0.0089 aux.acc_seg: 98.7281 +04/18 04:06:22 - mmengine - INFO - Iter(train) [ 11350/160000] lr: 9.3657e-03 eta: 22:40:19 time: 0.5476 data_time: 0.0072 memory: 7635 loss: 0.0202 decode.loss_ce: 0.0113 decode.acc_seg: 99.4430 aux.loss_ce: 0.0089 aux.acc_seg: 98.7743 +04/18 04:06:49 - mmengine - INFO - Iter(train) [ 11400/160000] lr: 9.3629e-03 eta: 22:39:51 time: 0.5485 data_time: 0.0065 memory: 7635 loss: 0.0211 decode.loss_ce: 0.0120 decode.acc_seg: 99.4951 aux.loss_ce: 0.0091 aux.acc_seg: 99.0124 +04/18 04:07:17 - mmengine - INFO - Iter(train) [ 11450/160000] lr: 9.3601e-03 eta: 22:39:23 time: 0.5487 data_time: 0.0058 memory: 7635 loss: 0.0198 decode.loss_ce: 0.0110 decode.acc_seg: 99.5588 aux.loss_ce: 0.0088 aux.acc_seg: 99.0472 +04/18 04:07:44 - mmengine - INFO - Iter(train) [ 11500/160000] lr: 9.3573e-03 eta: 22:38:55 time: 0.5485 data_time: 0.0064 memory: 7635 loss: 0.0213 decode.loss_ce: 0.0119 decode.acc_seg: 99.4657 aux.loss_ce: 0.0093 aux.acc_seg: 98.9974 +04/18 04:08:12 - mmengine - INFO - Iter(train) [ 11550/160000] lr: 9.3545e-03 eta: 22:38:27 time: 0.5477 data_time: 0.0062 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0121 decode.acc_seg: 99.6387 aux.loss_ce: 0.0088 aux.acc_seg: 99.1696 +04/18 04:08:39 - mmengine - INFO - Iter(train) [ 11600/160000] lr: 9.3517e-03 eta: 22:38:00 time: 0.5493 data_time: 0.0060 memory: 7635 loss: 0.0200 decode.loss_ce: 0.0113 decode.acc_seg: 99.5864 aux.loss_ce: 0.0087 aux.acc_seg: 99.1900 +04/18 04:09:07 - mmengine - INFO - Iter(train) [ 11650/160000] lr: 9.3489e-03 eta: 22:37:32 time: 0.5483 data_time: 0.0061 memory: 7635 loss: 0.0250 decode.loss_ce: 0.0142 decode.acc_seg: 99.3481 aux.loss_ce: 0.0108 aux.acc_seg: 98.6688 +04/18 04:09:34 - mmengine - INFO - Iter(train) [ 11700/160000] lr: 9.3461e-03 eta: 22:37:04 time: 0.5491 data_time: 0.0063 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0107 decode.acc_seg: 99.5366 aux.loss_ce: 0.0085 aux.acc_seg: 99.1977 +04/18 04:10:01 - mmengine - INFO - Iter(train) [ 11750/160000] lr: 9.3433e-03 eta: 22:36:36 time: 0.5486 data_time: 0.0058 memory: 7635 loss: 0.0200 decode.loss_ce: 0.0115 decode.acc_seg: 99.5639 aux.loss_ce: 0.0085 aux.acc_seg: 99.1105 +04/18 04:10:29 - mmengine - INFO - Iter(train) [ 11800/160000] lr: 9.3404e-03 eta: 22:36:08 time: 0.5485 data_time: 0.0070 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0115 decode.acc_seg: 99.4719 aux.loss_ce: 0.0091 aux.acc_seg: 99.0666 +04/18 04:10:56 - mmengine - INFO - Iter(train) [ 11850/160000] lr: 9.3376e-03 eta: 22:35:40 time: 0.5481 data_time: 0.0062 memory: 7635 loss: 0.0214 decode.loss_ce: 0.0122 decode.acc_seg: 99.5543 aux.loss_ce: 0.0092 aux.acc_seg: 99.1239 +04/18 04:11:24 - mmengine - INFO - Iter(train) [ 11900/160000] lr: 9.3348e-03 eta: 22:35:12 time: 0.5481 data_time: 0.0058 memory: 7635 loss: 0.0222 decode.loss_ce: 0.0126 decode.acc_seg: 99.5976 aux.loss_ce: 0.0096 aux.acc_seg: 98.9672 +04/18 04:11:51 - mmengine - INFO - Iter(train) [ 11950/160000] lr: 9.3320e-03 eta: 22:34:44 time: 0.5485 data_time: 0.0063 memory: 7635 loss: 0.0226 decode.loss_ce: 0.0131 decode.acc_seg: 99.4691 aux.loss_ce: 0.0095 aux.acc_seg: 98.9560 +04/18 04:12:18 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 04:12:18 - mmengine - INFO - Iter(train) [ 12000/160000] lr: 9.3292e-03 eta: 22:34:16 time: 0.5470 data_time: 0.0058 memory: 7635 loss: 0.0211 decode.loss_ce: 0.0122 decode.acc_seg: 99.4463 aux.loss_ce: 0.0089 aux.acc_seg: 98.8301 +04/18 04:12:46 - mmengine - INFO - Iter(train) [ 12050/160000] lr: 9.3264e-03 eta: 22:33:48 time: 0.5477 data_time: 0.0060 memory: 7635 loss: 0.0216 decode.loss_ce: 0.0122 decode.acc_seg: 99.6193 aux.loss_ce: 0.0094 aux.acc_seg: 99.1158 +04/18 04:13:13 - mmengine - INFO - Iter(train) [ 12100/160000] lr: 9.3236e-03 eta: 22:33:20 time: 0.5478 data_time: 0.0063 memory: 7635 loss: 0.0225 decode.loss_ce: 0.0131 decode.acc_seg: 99.3755 aux.loss_ce: 0.0094 aux.acc_seg: 98.8880 +04/18 04:13:41 - mmengine - INFO - Iter(train) [ 12150/160000] lr: 9.3208e-03 eta: 22:32:52 time: 0.5485 data_time: 0.0058 memory: 7635 loss: 0.0214 decode.loss_ce: 0.0120 decode.acc_seg: 99.4139 aux.loss_ce: 0.0094 aux.acc_seg: 98.9518 +04/18 04:14:08 - mmengine - INFO - Iter(train) [ 12200/160000] lr: 9.3180e-03 eta: 22:32:25 time: 0.5485 data_time: 0.0063 memory: 7635 loss: 0.0235 decode.loss_ce: 0.0138 decode.acc_seg: 99.3939 aux.loss_ce: 0.0097 aux.acc_seg: 98.9902 +04/18 04:14:36 - mmengine - INFO - Iter(train) [ 12250/160000] lr: 9.3152e-03 eta: 22:31:57 time: 0.5483 data_time: 0.0069 memory: 7635 loss: 0.0229 decode.loss_ce: 0.0133 decode.acc_seg: 99.5385 aux.loss_ce: 0.0096 aux.acc_seg: 99.1913 +04/18 04:15:03 - mmengine - INFO - Iter(train) [ 12300/160000] lr: 9.3124e-03 eta: 22:31:32 time: 0.5593 data_time: 0.0068 memory: 7635 loss: 0.0205 decode.loss_ce: 0.0116 decode.acc_seg: 99.5448 aux.loss_ce: 0.0089 aux.acc_seg: 98.9612 +04/18 04:15:31 - mmengine - INFO - Iter(train) [ 12350/160000] lr: 9.3096e-03 eta: 22:31:04 time: 0.5497 data_time: 0.0062 memory: 7635 loss: 0.0208 decode.loss_ce: 0.0116 decode.acc_seg: 99.5140 aux.loss_ce: 0.0092 aux.acc_seg: 99.0891 +04/18 04:15:58 - mmengine - INFO - Iter(train) [ 12400/160000] lr: 9.3068e-03 eta: 22:30:37 time: 0.5491 data_time: 0.0061 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0105 decode.acc_seg: 99.5073 aux.loss_ce: 0.0086 aux.acc_seg: 98.8900 +04/18 04:16:26 - mmengine - INFO - Iter(train) [ 12450/160000] lr: 9.3040e-03 eta: 22:30:09 time: 0.5496 data_time: 0.0067 memory: 7635 loss: 0.0222 decode.loss_ce: 0.0127 decode.acc_seg: 99.5250 aux.loss_ce: 0.0096 aux.acc_seg: 98.9526 +04/18 04:16:53 - mmengine - INFO - Iter(train) [ 12500/160000] lr: 9.3012e-03 eta: 22:29:42 time: 0.5496 data_time: 0.0061 memory: 7635 loss: 0.0215 decode.loss_ce: 0.0120 decode.acc_seg: 99.2399 aux.loss_ce: 0.0095 aux.acc_seg: 98.6911 +04/18 04:17:20 - mmengine - INFO - Iter(train) [ 12550/160000] lr: 9.2984e-03 eta: 22:29:14 time: 0.5482 data_time: 0.0061 memory: 7635 loss: 0.0219 decode.loss_ce: 0.0126 decode.acc_seg: 99.5748 aux.loss_ce: 0.0093 aux.acc_seg: 99.1762 +04/18 04:17:48 - mmengine - INFO - Iter(train) [ 12600/160000] lr: 9.2955e-03 eta: 22:28:46 time: 0.5485 data_time: 0.0060 memory: 7635 loss: 0.0204 decode.loss_ce: 0.0115 decode.acc_seg: 99.2990 aux.loss_ce: 0.0089 aux.acc_seg: 98.8092 +04/18 04:18:15 - mmengine - INFO - Iter(train) [ 12650/160000] lr: 9.2927e-03 eta: 22:28:19 time: 0.5488 data_time: 0.0057 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0107 decode.acc_seg: 99.5843 aux.loss_ce: 0.0084 aux.acc_seg: 99.1313 +04/18 04:18:43 - mmengine - INFO - Iter(train) [ 12700/160000] lr: 9.2899e-03 eta: 22:27:53 time: 0.5496 data_time: 0.0061 memory: 7635 loss: 0.0194 decode.loss_ce: 0.0107 decode.acc_seg: 99.6871 aux.loss_ce: 0.0087 aux.acc_seg: 99.2489 +04/18 04:19:10 - mmengine - INFO - Iter(train) [ 12750/160000] lr: 9.2871e-03 eta: 22:27:25 time: 0.5492 data_time: 0.0060 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0118 decode.acc_seg: 99.5813 aux.loss_ce: 0.0091 aux.acc_seg: 99.1307 +04/18 04:19:38 - mmengine - INFO - Iter(train) [ 12800/160000] lr: 9.2843e-03 eta: 22:26:57 time: 0.5488 data_time: 0.0054 memory: 7635 loss: 0.0201 decode.loss_ce: 0.0111 decode.acc_seg: 99.3687 aux.loss_ce: 0.0090 aux.acc_seg: 98.7031 +04/18 04:20:05 - mmengine - INFO - Iter(train) [ 12850/160000] lr: 9.2815e-03 eta: 22:26:30 time: 0.5492 data_time: 0.0058 memory: 7635 loss: 0.0205 decode.loss_ce: 0.0116 decode.acc_seg: 99.6304 aux.loss_ce: 0.0089 aux.acc_seg: 99.1039 +04/18 04:20:33 - mmengine - INFO - Iter(train) [ 12900/160000] lr: 9.2787e-03 eta: 22:26:02 time: 0.5487 data_time: 0.0063 memory: 7635 loss: 0.0214 decode.loss_ce: 0.0121 decode.acc_seg: 99.5192 aux.loss_ce: 0.0094 aux.acc_seg: 99.0376 +04/18 04:21:00 - mmengine - INFO - Iter(train) [ 12950/160000] lr: 9.2759e-03 eta: 22:25:34 time: 0.5485 data_time: 0.0060 memory: 7635 loss: 0.0203 decode.loss_ce: 0.0113 decode.acc_seg: 99.5872 aux.loss_ce: 0.0090 aux.acc_seg: 99.1456 +04/18 04:21:27 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 04:21:27 - mmengine - INFO - Iter(train) [ 13000/160000] lr: 9.2731e-03 eta: 22:25:07 time: 0.5489 data_time: 0.0058 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0104 decode.acc_seg: 99.6518 aux.loss_ce: 0.0085 aux.acc_seg: 99.3849 +04/18 04:21:55 - mmengine - INFO - Iter(train) [ 13050/160000] lr: 9.2703e-03 eta: 22:24:39 time: 0.5494 data_time: 0.0070 memory: 7635 loss: 0.0203 decode.loss_ce: 0.0116 decode.acc_seg: 99.5750 aux.loss_ce: 0.0086 aux.acc_seg: 99.0207 +04/18 04:22:22 - mmengine - INFO - Iter(train) [ 13100/160000] lr: 9.2675e-03 eta: 22:24:11 time: 0.5494 data_time: 0.0060 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0106 decode.acc_seg: 99.4333 aux.loss_ce: 0.0081 aux.acc_seg: 99.0598 +04/18 04:22:50 - mmengine - INFO - Iter(train) [ 13150/160000] lr: 9.2647e-03 eta: 22:23:44 time: 0.5481 data_time: 0.0058 memory: 7635 loss: 0.0201 decode.loss_ce: 0.0111 decode.acc_seg: 99.7065 aux.loss_ce: 0.0089 aux.acc_seg: 99.2218 +04/18 04:23:17 - mmengine - INFO - Iter(train) [ 13200/160000] lr: 9.2618e-03 eta: 22:23:16 time: 0.5474 data_time: 0.0059 memory: 7635 loss: 0.0204 decode.loss_ce: 0.0115 decode.acc_seg: 99.3992 aux.loss_ce: 0.0089 aux.acc_seg: 98.8688 +04/18 04:23:45 - mmengine - INFO - Iter(train) [ 13250/160000] lr: 9.2590e-03 eta: 22:22:49 time: 0.5493 data_time: 0.0065 memory: 7635 loss: 0.0231 decode.loss_ce: 0.0133 decode.acc_seg: 99.5006 aux.loss_ce: 0.0099 aux.acc_seg: 98.9125 +04/18 04:24:12 - mmengine - INFO - Iter(train) [ 13300/160000] lr: 9.2562e-03 eta: 22:22:21 time: 0.5493 data_time: 0.0062 memory: 7635 loss: 0.0206 decode.loss_ce: 0.0117 decode.acc_seg: 99.5731 aux.loss_ce: 0.0089 aux.acc_seg: 99.1207 +04/18 04:24:40 - mmengine - INFO - Iter(train) [ 13350/160000] lr: 9.2534e-03 eta: 22:21:54 time: 0.5500 data_time: 0.0068 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0106 decode.acc_seg: 99.5807 aux.loss_ce: 0.0085 aux.acc_seg: 99.0040 +04/18 04:25:07 - mmengine - INFO - Iter(train) [ 13400/160000] lr: 9.2506e-03 eta: 22:21:28 time: 0.5474 data_time: 0.0060 memory: 7635 loss: 0.0194 decode.loss_ce: 0.0107 decode.acc_seg: 99.5396 aux.loss_ce: 0.0087 aux.acc_seg: 98.9564 +04/18 04:25:35 - mmengine - INFO - Iter(train) [ 13450/160000] lr: 9.2478e-03 eta: 22:21:01 time: 0.5490 data_time: 0.0060 memory: 7635 loss: 0.0202 decode.loss_ce: 0.0113 decode.acc_seg: 99.6347 aux.loss_ce: 0.0089 aux.acc_seg: 99.2900 +04/18 04:26:02 - mmengine - INFO - Iter(train) [ 13500/160000] lr: 9.2450e-03 eta: 22:20:33 time: 0.5492 data_time: 0.0061 memory: 7635 loss: 0.0213 decode.loss_ce: 0.0120 decode.acc_seg: 99.7210 aux.loss_ce: 0.0093 aux.acc_seg: 99.3200 +04/18 04:26:30 - mmengine - INFO - Iter(train) [ 13550/160000] lr: 9.2422e-03 eta: 22:20:06 time: 0.5489 data_time: 0.0063 memory: 7635 loss: 0.0211 decode.loss_ce: 0.0118 decode.acc_seg: 99.6255 aux.loss_ce: 0.0093 aux.acc_seg: 99.3737 +04/18 04:26:57 - mmengine - INFO - Iter(train) [ 13600/160000] lr: 9.2394e-03 eta: 22:19:39 time: 0.5493 data_time: 0.0055 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0114 decode.acc_seg: 99.3721 aux.loss_ce: 0.0093 aux.acc_seg: 98.8181 +04/18 04:27:25 - mmengine - INFO - Iter(train) [ 13650/160000] lr: 9.2366e-03 eta: 22:19:11 time: 0.5491 data_time: 0.0064 memory: 7635 loss: 0.0205 decode.loss_ce: 0.0116 decode.acc_seg: 99.5900 aux.loss_ce: 0.0088 aux.acc_seg: 99.2604 +04/18 04:27:52 - mmengine - INFO - Iter(train) [ 13700/160000] lr: 9.2338e-03 eta: 22:18:44 time: 0.5496 data_time: 0.0060 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0117 decode.acc_seg: 99.6017 aux.loss_ce: 0.0092 aux.acc_seg: 99.2126 +04/18 04:28:20 - mmengine - INFO - Iter(train) [ 13750/160000] lr: 9.2309e-03 eta: 22:18:18 time: 0.5484 data_time: 0.0058 memory: 7635 loss: 0.0212 decode.loss_ce: 0.0120 decode.acc_seg: 99.3927 aux.loss_ce: 0.0092 aux.acc_seg: 98.9446 +04/18 04:28:47 - mmengine - INFO - Iter(train) [ 13800/160000] lr: 9.2281e-03 eta: 22:17:50 time: 0.5484 data_time: 0.0060 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0118 decode.acc_seg: 99.4184 aux.loss_ce: 0.0091 aux.acc_seg: 98.8906 +04/18 04:29:14 - mmengine - INFO - Iter(train) [ 13850/160000] lr: 9.2253e-03 eta: 22:17:23 time: 0.5491 data_time: 0.0058 memory: 7635 loss: 0.0206 decode.loss_ce: 0.0115 decode.acc_seg: 99.4781 aux.loss_ce: 0.0091 aux.acc_seg: 98.8622 +04/18 04:29:42 - mmengine - INFO - Iter(train) [ 13900/160000] lr: 9.2225e-03 eta: 22:16:56 time: 0.5485 data_time: 0.0059 memory: 7635 loss: 0.0203 decode.loss_ce: 0.0111 decode.acc_seg: 99.5853 aux.loss_ce: 0.0092 aux.acc_seg: 99.0317 +04/18 04:30:09 - mmengine - INFO - Iter(train) [ 13950/160000] lr: 9.2197e-03 eta: 22:16:28 time: 0.5497 data_time: 0.0065 memory: 7635 loss: 0.0234 decode.loss_ce: 0.0134 decode.acc_seg: 99.5074 aux.loss_ce: 0.0100 aux.acc_seg: 99.0853 +04/18 04:30:37 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 04:30:37 - mmengine - INFO - Iter(train) [ 14000/160000] lr: 9.2169e-03 eta: 22:16:01 time: 0.5499 data_time: 0.0065 memory: 7635 loss: 0.0210 decode.loss_ce: 0.0117 decode.acc_seg: 99.5838 aux.loss_ce: 0.0093 aux.acc_seg: 99.0235 +04/18 04:31:04 - mmengine - INFO - Iter(train) [ 14050/160000] lr: 9.2141e-03 eta: 22:15:33 time: 0.5495 data_time: 0.0069 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0120 decode.acc_seg: 99.5310 aux.loss_ce: 0.0090 aux.acc_seg: 99.1996 +04/18 04:31:32 - mmengine - INFO - Iter(train) [ 14100/160000] lr: 9.2113e-03 eta: 22:15:05 time: 0.5483 data_time: 0.0058 memory: 7635 loss: 0.0200 decode.loss_ce: 0.0109 decode.acc_seg: 99.5172 aux.loss_ce: 0.0090 aux.acc_seg: 99.0437 +04/18 04:31:59 - mmengine - INFO - Iter(train) [ 14150/160000] lr: 9.2085e-03 eta: 22:14:38 time: 0.5491 data_time: 0.0062 memory: 7635 loss: 0.0219 decode.loss_ce: 0.0122 decode.acc_seg: 99.6517 aux.loss_ce: 0.0096 aux.acc_seg: 99.2456 +04/18 04:32:27 - mmengine - INFO - Iter(train) [ 14200/160000] lr: 9.2057e-03 eta: 22:14:10 time: 0.5494 data_time: 0.0060 memory: 7635 loss: 0.0220 decode.loss_ce: 0.0123 decode.acc_seg: 99.5095 aux.loss_ce: 0.0097 aux.acc_seg: 99.1067 +04/18 04:32:54 - mmengine - INFO - Iter(train) [ 14250/160000] lr: 9.2028e-03 eta: 22:13:43 time: 0.5481 data_time: 0.0065 memory: 7635 loss: 0.0200 decode.loss_ce: 0.0113 decode.acc_seg: 99.6144 aux.loss_ce: 0.0087 aux.acc_seg: 99.2407 +04/18 04:33:22 - mmengine - INFO - Iter(train) [ 14300/160000] lr: 9.2000e-03 eta: 22:13:16 time: 0.5504 data_time: 0.0058 memory: 7635 loss: 0.0234 decode.loss_ce: 0.0133 decode.acc_seg: 99.5925 aux.loss_ce: 0.0101 aux.acc_seg: 99.0404 +04/18 04:33:49 - mmengine - INFO - Iter(train) [ 14350/160000] lr: 9.1972e-03 eta: 22:12:49 time: 0.5510 data_time: 0.0062 memory: 7635 loss: 0.0211 decode.loss_ce: 0.0119 decode.acc_seg: 99.5124 aux.loss_ce: 0.0092 aux.acc_seg: 99.0582 +04/18 04:34:17 - mmengine - INFO - Iter(train) [ 14400/160000] lr: 9.1944e-03 eta: 22:12:21 time: 0.5491 data_time: 0.0059 memory: 7635 loss: 0.0190 decode.loss_ce: 0.0105 decode.acc_seg: 99.4257 aux.loss_ce: 0.0085 aux.acc_seg: 98.9985 +04/18 04:34:44 - mmengine - INFO - Iter(train) [ 14450/160000] lr: 9.1916e-03 eta: 22:11:56 time: 0.5595 data_time: 0.0067 memory: 7635 loss: 0.0201 decode.loss_ce: 0.0112 decode.acc_seg: 99.6026 aux.loss_ce: 0.0089 aux.acc_seg: 99.0494 +04/18 04:35:12 - mmengine - INFO - Iter(train) [ 14500/160000] lr: 9.1888e-03 eta: 22:11:29 time: 0.5491 data_time: 0.0066 memory: 7635 loss: 0.0206 decode.loss_ce: 0.0117 decode.acc_seg: 99.3559 aux.loss_ce: 0.0089 aux.acc_seg: 98.6353 +04/18 04:35:39 - mmengine - INFO - Iter(train) [ 14550/160000] lr: 9.1860e-03 eta: 22:11:02 time: 0.5500 data_time: 0.0059 memory: 7635 loss: 0.0204 decode.loss_ce: 0.0115 decode.acc_seg: 99.5761 aux.loss_ce: 0.0089 aux.acc_seg: 99.1471 +04/18 04:36:07 - mmengine - INFO - Iter(train) [ 14600/160000] lr: 9.1832e-03 eta: 22:10:34 time: 0.5500 data_time: 0.0060 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0096 decode.acc_seg: 99.6475 aux.loss_ce: 0.0082 aux.acc_seg: 99.0929 +04/18 04:36:34 - mmengine - INFO - Iter(train) [ 14650/160000] lr: 9.1804e-03 eta: 22:10:07 time: 0.5498 data_time: 0.0065 memory: 7635 loss: 0.0205 decode.loss_ce: 0.0113 decode.acc_seg: 99.6323 aux.loss_ce: 0.0092 aux.acc_seg: 99.2687 +04/18 04:37:02 - mmengine - INFO - Iter(train) [ 14700/160000] lr: 9.1776e-03 eta: 22:09:40 time: 0.5487 data_time: 0.0062 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0104 decode.acc_seg: 99.6775 aux.loss_ce: 0.0083 aux.acc_seg: 99.3504 +04/18 04:37:29 - mmengine - INFO - Iter(train) [ 14750/160000] lr: 9.1747e-03 eta: 22:09:13 time: 0.5499 data_time: 0.0065 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0105 decode.acc_seg: 99.5389 aux.loss_ce: 0.0087 aux.acc_seg: 98.9362 +04/18 04:37:57 - mmengine - INFO - Iter(train) [ 14800/160000] lr: 9.1719e-03 eta: 22:08:46 time: 0.5510 data_time: 0.0068 memory: 7635 loss: 0.0206 decode.loss_ce: 0.0116 decode.acc_seg: 99.5256 aux.loss_ce: 0.0090 aux.acc_seg: 99.0493 +04/18 04:38:24 - mmengine - INFO - Iter(train) [ 14850/160000] lr: 9.1691e-03 eta: 22:08:20 time: 0.5493 data_time: 0.0062 memory: 7635 loss: 0.0205 decode.loss_ce: 0.0115 decode.acc_seg: 99.7200 aux.loss_ce: 0.0090 aux.acc_seg: 99.3581 +04/18 04:38:52 - mmengine - INFO - Iter(train) [ 14900/160000] lr: 9.1663e-03 eta: 22:07:52 time: 0.5493 data_time: 0.0065 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0107 decode.acc_seg: 99.6327 aux.loss_ce: 0.0082 aux.acc_seg: 99.1156 +04/18 04:39:19 - mmengine - INFO - Iter(train) [ 14950/160000] lr: 9.1635e-03 eta: 22:07:25 time: 0.5489 data_time: 0.0061 memory: 7635 loss: 0.0196 decode.loss_ce: 0.0106 decode.acc_seg: 99.5702 aux.loss_ce: 0.0090 aux.acc_seg: 98.9697 +04/18 04:39:47 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 04:39:47 - mmengine - INFO - Iter(train) [ 15000/160000] lr: 9.1607e-03 eta: 22:06:58 time: 0.5517 data_time: 0.0066 memory: 7635 loss: 0.0190 decode.loss_ce: 0.0106 decode.acc_seg: 99.6482 aux.loss_ce: 0.0084 aux.acc_seg: 99.2591 +04/18 04:40:14 - mmengine - INFO - Iter(train) [ 15050/160000] lr: 9.1579e-03 eta: 22:06:30 time: 0.5496 data_time: 0.0066 memory: 7635 loss: 0.0199 decode.loss_ce: 0.0111 decode.acc_seg: 99.5513 aux.loss_ce: 0.0087 aux.acc_seg: 99.0436 +04/18 04:40:41 - mmengine - INFO - Iter(train) [ 15100/160000] lr: 9.1551e-03 eta: 22:06:03 time: 0.5490 data_time: 0.0060 memory: 7635 loss: 0.0216 decode.loss_ce: 0.0122 decode.acc_seg: 99.5356 aux.loss_ce: 0.0095 aux.acc_seg: 99.0978 +04/18 04:41:09 - mmengine - INFO - Iter(train) [ 15150/160000] lr: 9.1522e-03 eta: 22:05:36 time: 0.5500 data_time: 0.0062 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0115 decode.acc_seg: 99.5110 aux.loss_ce: 0.0092 aux.acc_seg: 98.9477 +04/18 04:41:36 - mmengine - INFO - Iter(train) [ 15200/160000] lr: 9.1494e-03 eta: 22:05:09 time: 0.5495 data_time: 0.0066 memory: 7635 loss: 0.0200 decode.loss_ce: 0.0112 decode.acc_seg: 99.5566 aux.loss_ce: 0.0088 aux.acc_seg: 99.2162 +04/18 04:42:04 - mmengine - INFO - Iter(train) [ 15250/160000] lr: 9.1466e-03 eta: 22:04:41 time: 0.5492 data_time: 0.0065 memory: 7635 loss: 0.0200 decode.loss_ce: 0.0110 decode.acc_seg: 99.4129 aux.loss_ce: 0.0091 aux.acc_seg: 98.9281 +04/18 04:42:31 - mmengine - INFO - Iter(train) [ 15300/160000] lr: 9.1438e-03 eta: 22:04:14 time: 0.5493 data_time: 0.0063 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0107 decode.acc_seg: 99.5400 aux.loss_ce: 0.0089 aux.acc_seg: 98.9194 +04/18 04:42:59 - mmengine - INFO - Iter(train) [ 15350/160000] lr: 9.1410e-03 eta: 22:03:47 time: 0.5485 data_time: 0.0063 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0107 decode.acc_seg: 99.3506 aux.loss_ce: 0.0088 aux.acc_seg: 98.7174 +04/18 04:43:26 - mmengine - INFO - Iter(train) [ 15400/160000] lr: 9.1382e-03 eta: 22:03:19 time: 0.5493 data_time: 0.0060 memory: 7635 loss: 0.0203 decode.loss_ce: 0.0114 decode.acc_seg: 99.4677 aux.loss_ce: 0.0089 aux.acc_seg: 98.9253 +04/18 04:43:54 - mmengine - INFO - Iter(train) [ 15450/160000] lr: 9.1354e-03 eta: 22:02:52 time: 0.5500 data_time: 0.0068 memory: 7635 loss: 0.0210 decode.loss_ce: 0.0117 decode.acc_seg: 99.5312 aux.loss_ce: 0.0093 aux.acc_seg: 99.0074 +04/18 04:44:21 - mmengine - INFO - Iter(train) [ 15500/160000] lr: 9.1326e-03 eta: 22:02:25 time: 0.5504 data_time: 0.0069 memory: 7635 loss: 0.0212 decode.loss_ce: 0.0116 decode.acc_seg: 99.4061 aux.loss_ce: 0.0095 aux.acc_seg: 98.9144 +04/18 04:44:49 - mmengine - INFO - Iter(train) [ 15550/160000] lr: 9.1297e-03 eta: 22:02:00 time: 0.5478 data_time: 0.0062 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0109 decode.acc_seg: 99.6532 aux.loss_ce: 0.0086 aux.acc_seg: 99.2592 +04/18 04:45:16 - mmengine - INFO - Iter(train) [ 15600/160000] lr: 9.1269e-03 eta: 22:01:33 time: 0.5504 data_time: 0.0062 memory: 7635 loss: 0.0197 decode.loss_ce: 0.0106 decode.acc_seg: 99.5452 aux.loss_ce: 0.0092 aux.acc_seg: 98.8504 +04/18 04:45:44 - mmengine - INFO - Iter(train) [ 15650/160000] lr: 9.1241e-03 eta: 22:01:05 time: 0.5498 data_time: 0.0061 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0116 decode.acc_seg: 99.4576 aux.loss_ce: 0.0093 aux.acc_seg: 98.7809 +04/18 04:46:11 - mmengine - INFO - Iter(train) [ 15700/160000] lr: 9.1213e-03 eta: 22:00:38 time: 0.5502 data_time: 0.0057 memory: 7635 loss: 0.0246 decode.loss_ce: 0.0141 decode.acc_seg: 99.4235 aux.loss_ce: 0.0105 aux.acc_seg: 98.8319 +04/18 04:46:39 - mmengine - INFO - Iter(train) [ 15750/160000] lr: 9.1185e-03 eta: 22:00:11 time: 0.5491 data_time: 0.0069 memory: 7635 loss: 0.0206 decode.loss_ce: 0.0113 decode.acc_seg: 99.6474 aux.loss_ce: 0.0093 aux.acc_seg: 99.1494 +04/18 04:47:06 - mmengine - INFO - Iter(train) [ 15800/160000] lr: 9.1157e-03 eta: 21:59:44 time: 0.5513 data_time: 0.0058 memory: 7635 loss: 0.0190 decode.loss_ce: 0.0104 decode.acc_seg: 99.6280 aux.loss_ce: 0.0086 aux.acc_seg: 99.1859 +04/18 04:47:34 - mmengine - INFO - Iter(train) [ 15850/160000] lr: 9.1129e-03 eta: 21:59:17 time: 0.5495 data_time: 0.0060 memory: 7635 loss: 0.0193 decode.loss_ce: 0.0105 decode.acc_seg: 99.6848 aux.loss_ce: 0.0088 aux.acc_seg: 99.1300 +04/18 04:48:02 - mmengine - INFO - Iter(train) [ 15900/160000] lr: 9.1101e-03 eta: 21:58:51 time: 0.5592 data_time: 0.0056 memory: 7635 loss: 0.0208 decode.loss_ce: 0.0116 decode.acc_seg: 99.5984 aux.loss_ce: 0.0092 aux.acc_seg: 99.1928 +04/18 04:48:29 - mmengine - INFO - Iter(train) [ 15950/160000] lr: 9.1072e-03 eta: 21:58:24 time: 0.5505 data_time: 0.0065 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0117 decode.acc_seg: 99.6102 aux.loss_ce: 0.0090 aux.acc_seg: 98.9958 +04/18 04:48:57 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 04:48:57 - mmengine - INFO - Iter(train) [ 16000/160000] lr: 9.1044e-03 eta: 21:57:57 time: 0.5515 data_time: 0.0063 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0098 decode.acc_seg: 99.6333 aux.loss_ce: 0.0085 aux.acc_seg: 99.1935 +04/18 04:49:24 - mmengine - INFO - Iter(train) [ 16050/160000] lr: 9.1016e-03 eta: 21:57:30 time: 0.5509 data_time: 0.0062 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0092 decode.acc_seg: 99.6289 aux.loss_ce: 0.0075 aux.acc_seg: 99.4611 +04/18 04:49:52 - mmengine - INFO - Iter(train) [ 16100/160000] lr: 9.0988e-03 eta: 21:57:03 time: 0.5509 data_time: 0.0058 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0095 decode.acc_seg: 99.5643 aux.loss_ce: 0.0079 aux.acc_seg: 99.2123 +04/18 04:50:19 - mmengine - INFO - Iter(train) [ 16150/160000] lr: 9.0960e-03 eta: 21:56:37 time: 0.5510 data_time: 0.0064 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0106 decode.acc_seg: 99.3545 aux.loss_ce: 0.0086 aux.acc_seg: 98.8721 +04/18 04:50:47 - mmengine - INFO - Iter(train) [ 16200/160000] lr: 9.0932e-03 eta: 21:56:10 time: 0.5496 data_time: 0.0062 memory: 7635 loss: 0.0196 decode.loss_ce: 0.0108 decode.acc_seg: 99.6251 aux.loss_ce: 0.0088 aux.acc_seg: 99.1516 +04/18 04:51:14 - mmengine - INFO - Iter(train) [ 16250/160000] lr: 9.0904e-03 eta: 21:55:43 time: 0.5509 data_time: 0.0064 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0105 decode.acc_seg: 99.5217 aux.loss_ce: 0.0087 aux.acc_seg: 99.0500 +04/18 04:51:42 - mmengine - INFO - Iter(train) [ 16300/160000] lr: 9.0875e-03 eta: 21:55:16 time: 0.5493 data_time: 0.0063 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0108 decode.acc_seg: 99.4192 aux.loss_ce: 0.0084 aux.acc_seg: 99.1014 +04/18 04:52:09 - mmengine - INFO - Iter(train) [ 16350/160000] lr: 9.0847e-03 eta: 21:54:49 time: 0.5497 data_time: 0.0057 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0105 decode.acc_seg: 99.6097 aux.loss_ce: 0.0089 aux.acc_seg: 99.0688 +04/18 04:52:37 - mmengine - INFO - Iter(train) [ 16400/160000] lr: 9.0819e-03 eta: 21:54:22 time: 0.5507 data_time: 0.0061 memory: 7635 loss: 0.0218 decode.loss_ce: 0.0126 decode.acc_seg: 99.4523 aux.loss_ce: 0.0091 aux.acc_seg: 99.0592 +04/18 04:53:04 - mmengine - INFO - Iter(train) [ 16450/160000] lr: 9.0791e-03 eta: 21:53:55 time: 0.5515 data_time: 0.0068 memory: 7635 loss: 0.0181 decode.loss_ce: 0.0097 decode.acc_seg: 99.6587 aux.loss_ce: 0.0084 aux.acc_seg: 99.2566 +04/18 04:53:32 - mmengine - INFO - Iter(train) [ 16500/160000] lr: 9.0763e-03 eta: 21:53:27 time: 0.5503 data_time: 0.0057 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0093 decode.acc_seg: 99.5905 aux.loss_ce: 0.0077 aux.acc_seg: 99.2222 +04/18 04:53:59 - mmengine - INFO - Iter(train) [ 16550/160000] lr: 9.0735e-03 eta: 21:53:01 time: 0.5504 data_time: 0.0068 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0104 decode.acc_seg: 99.6407 aux.loss_ce: 0.0085 aux.acc_seg: 99.3345 +04/18 04:54:27 - mmengine - INFO - Iter(train) [ 16600/160000] lr: 9.0706e-03 eta: 21:52:36 time: 0.5699 data_time: 0.0060 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0100 decode.acc_seg: 99.7334 aux.loss_ce: 0.0082 aux.acc_seg: 99.4706 +04/18 04:54:55 - mmengine - INFO - Iter(train) [ 16650/160000] lr: 9.0678e-03 eta: 21:52:08 time: 0.5493 data_time: 0.0062 memory: 7635 loss: 0.0200 decode.loss_ce: 0.0112 decode.acc_seg: 99.5617 aux.loss_ce: 0.0088 aux.acc_seg: 99.1335 +04/18 04:55:22 - mmengine - INFO - Iter(train) [ 16700/160000] lr: 9.0650e-03 eta: 21:51:41 time: 0.5510 data_time: 0.0058 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0104 decode.acc_seg: 99.6317 aux.loss_ce: 0.0084 aux.acc_seg: 99.2899 +04/18 04:55:50 - mmengine - INFO - Iter(train) [ 16750/160000] lr: 9.0622e-03 eta: 21:51:14 time: 0.5502 data_time: 0.0069 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0111 decode.acc_seg: 99.6498 aux.loss_ce: 0.0084 aux.acc_seg: 99.2376 +04/18 04:56:17 - mmengine - INFO - Iter(train) [ 16800/160000] lr: 9.0594e-03 eta: 21:50:48 time: 0.5512 data_time: 0.0062 memory: 7635 loss: 0.0208 decode.loss_ce: 0.0117 decode.acc_seg: 99.6216 aux.loss_ce: 0.0091 aux.acc_seg: 99.2575 +04/18 04:56:45 - mmengine - INFO - Iter(train) [ 16850/160000] lr: 9.0566e-03 eta: 21:50:20 time: 0.5495 data_time: 0.0060 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0107 decode.acc_seg: 99.6372 aux.loss_ce: 0.0089 aux.acc_seg: 99.2227 +04/18 04:57:12 - mmengine - INFO - Iter(train) [ 16900/160000] lr: 9.0538e-03 eta: 21:49:53 time: 0.5494 data_time: 0.0058 memory: 7635 loss: 0.0181 decode.loss_ce: 0.0101 decode.acc_seg: 99.3711 aux.loss_ce: 0.0080 aux.acc_seg: 98.9654 +04/18 04:57:40 - mmengine - INFO - Iter(train) [ 16950/160000] lr: 9.0509e-03 eta: 21:49:27 time: 0.5595 data_time: 0.0060 memory: 7635 loss: 0.0202 decode.loss_ce: 0.0112 decode.acc_seg: 99.5635 aux.loss_ce: 0.0090 aux.acc_seg: 98.9095 +04/18 04:58:07 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 04:58:07 - mmengine - INFO - Iter(train) [ 17000/160000] lr: 9.0481e-03 eta: 21:49:00 time: 0.5507 data_time: 0.0058 memory: 7635 loss: 0.0223 decode.loss_ce: 0.0129 decode.acc_seg: 99.0251 aux.loss_ce: 0.0094 aux.acc_seg: 98.6746 +04/18 04:58:35 - mmengine - INFO - Iter(train) [ 17050/160000] lr: 9.0453e-03 eta: 21:48:33 time: 0.5500 data_time: 0.0062 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0104 decode.acc_seg: 99.5245 aux.loss_ce: 0.0084 aux.acc_seg: 99.0348 +04/18 04:59:02 - mmengine - INFO - Iter(train) [ 17100/160000] lr: 9.0425e-03 eta: 21:48:06 time: 0.5504 data_time: 0.0054 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0116 decode.acc_seg: 99.5734 aux.loss_ce: 0.0091 aux.acc_seg: 99.1453 +04/18 04:59:30 - mmengine - INFO - Iter(train) [ 17150/160000] lr: 9.0397e-03 eta: 21:47:40 time: 0.5507 data_time: 0.0059 memory: 7635 loss: 0.0204 decode.loss_ce: 0.0115 decode.acc_seg: 99.5340 aux.loss_ce: 0.0089 aux.acc_seg: 99.0964 +04/18 04:59:57 - mmengine - INFO - Iter(train) [ 17200/160000] lr: 9.0369e-03 eta: 21:47:13 time: 0.5511 data_time: 0.0070 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0104 decode.acc_seg: 99.5158 aux.loss_ce: 0.0086 aux.acc_seg: 99.1407 +04/18 05:00:25 - mmengine - INFO - Iter(train) [ 17250/160000] lr: 9.0340e-03 eta: 21:46:46 time: 0.5507 data_time: 0.0065 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0104 decode.acc_seg: 99.5226 aux.loss_ce: 0.0081 aux.acc_seg: 99.0396 +04/18 05:00:52 - mmengine - INFO - Iter(train) [ 17300/160000] lr: 9.0312e-03 eta: 21:46:19 time: 0.5496 data_time: 0.0058 memory: 7635 loss: 0.0197 decode.loss_ce: 0.0104 decode.acc_seg: 99.5439 aux.loss_ce: 0.0092 aux.acc_seg: 99.0997 +04/18 05:01:20 - mmengine - INFO - Iter(train) [ 17350/160000] lr: 9.0284e-03 eta: 21:45:52 time: 0.5498 data_time: 0.0068 memory: 7635 loss: 0.0196 decode.loss_ce: 0.0107 decode.acc_seg: 99.7268 aux.loss_ce: 0.0089 aux.acc_seg: 99.3399 +04/18 05:01:48 - mmengine - INFO - Iter(train) [ 17400/160000] lr: 9.0256e-03 eta: 21:45:25 time: 0.5511 data_time: 0.0071 memory: 7635 loss: 0.0194 decode.loss_ce: 0.0105 decode.acc_seg: 99.5956 aux.loss_ce: 0.0089 aux.acc_seg: 99.0271 +04/18 05:02:15 - mmengine - INFO - Iter(train) [ 17450/160000] lr: 9.0228e-03 eta: 21:44:58 time: 0.5498 data_time: 0.0058 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0103 decode.acc_seg: 99.6334 aux.loss_ce: 0.0080 aux.acc_seg: 99.0758 +04/18 05:02:43 - mmengine - INFO - Iter(train) [ 17500/160000] lr: 9.0200e-03 eta: 21:44:31 time: 0.5496 data_time: 0.0059 memory: 7635 loss: 0.0210 decode.loss_ce: 0.0116 decode.acc_seg: 99.4657 aux.loss_ce: 0.0094 aux.acc_seg: 99.0963 +04/18 05:03:10 - mmengine - INFO - Iter(train) [ 17550/160000] lr: 9.0171e-03 eta: 21:44:04 time: 0.5508 data_time: 0.0062 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0100 decode.acc_seg: 99.6763 aux.loss_ce: 0.0085 aux.acc_seg: 99.2926 +04/18 05:03:38 - mmengine - INFO - Iter(train) [ 17600/160000] lr: 9.0143e-03 eta: 21:43:37 time: 0.5513 data_time: 0.0066 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0101 decode.acc_seg: 99.6868 aux.loss_ce: 0.0084 aux.acc_seg: 99.3676 +04/18 05:04:05 - mmengine - INFO - Iter(train) [ 17650/160000] lr: 9.0115e-03 eta: 21:43:10 time: 0.5506 data_time: 0.0060 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0099 decode.acc_seg: 99.6650 aux.loss_ce: 0.0083 aux.acc_seg: 99.2867 +04/18 05:04:33 - mmengine - INFO - Iter(train) [ 17700/160000] lr: 9.0087e-03 eta: 21:42:45 time: 0.5503 data_time: 0.0065 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0105 decode.acc_seg: 99.4845 aux.loss_ce: 0.0086 aux.acc_seg: 98.9722 +04/18 05:05:00 - mmengine - INFO - Iter(train) [ 17750/160000] lr: 9.0059e-03 eta: 21:42:18 time: 0.5516 data_time: 0.0066 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0099 decode.acc_seg: 99.7449 aux.loss_ce: 0.0084 aux.acc_seg: 99.3885 +04/18 05:05:28 - mmengine - INFO - Iter(train) [ 17800/160000] lr: 9.0031e-03 eta: 21:41:51 time: 0.5514 data_time: 0.0060 memory: 7635 loss: 0.0197 decode.loss_ce: 0.0113 decode.acc_seg: 99.6144 aux.loss_ce: 0.0085 aux.acc_seg: 99.2468 +04/18 05:05:55 - mmengine - INFO - Iter(train) [ 17850/160000] lr: 9.0002e-03 eta: 21:41:24 time: 0.5495 data_time: 0.0061 memory: 7635 loss: 0.0203 decode.loss_ce: 0.0111 decode.acc_seg: 99.5973 aux.loss_ce: 0.0092 aux.acc_seg: 98.9994 +04/18 05:06:23 - mmengine - INFO - Iter(train) [ 17900/160000] lr: 8.9974e-03 eta: 21:40:57 time: 0.5513 data_time: 0.0063 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0103 decode.acc_seg: 99.5664 aux.loss_ce: 0.0085 aux.acc_seg: 99.0761 +04/18 05:06:51 - mmengine - INFO - Iter(train) [ 17950/160000] lr: 8.9946e-03 eta: 21:40:30 time: 0.5504 data_time: 0.0056 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0104 decode.acc_seg: 99.5418 aux.loss_ce: 0.0086 aux.acc_seg: 98.7377 +04/18 05:07:18 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 05:07:18 - mmengine - INFO - Iter(train) [ 18000/160000] lr: 8.9918e-03 eta: 21:40:03 time: 0.5504 data_time: 0.0069 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0102 decode.acc_seg: 99.4904 aux.loss_ce: 0.0085 aux.acc_seg: 98.9013 +04/18 05:07:46 - mmengine - INFO - Iter(train) [ 18050/160000] lr: 8.9890e-03 eta: 21:39:37 time: 0.5611 data_time: 0.0070 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0094 decode.acc_seg: 99.6426 aux.loss_ce: 0.0080 aux.acc_seg: 99.0942 +04/18 05:08:13 - mmengine - INFO - Iter(train) [ 18100/160000] lr: 8.9862e-03 eta: 21:39:10 time: 0.5509 data_time: 0.0060 memory: 7635 loss: 0.0198 decode.loss_ce: 0.0110 decode.acc_seg: 99.6600 aux.loss_ce: 0.0088 aux.acc_seg: 99.1778 +04/18 05:08:41 - mmengine - INFO - Iter(train) [ 18150/160000] lr: 8.9833e-03 eta: 21:38:43 time: 0.5515 data_time: 0.0063 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0104 decode.acc_seg: 99.5716 aux.loss_ce: 0.0085 aux.acc_seg: 99.0636 +04/18 05:09:08 - mmengine - INFO - Iter(train) [ 18200/160000] lr: 8.9805e-03 eta: 21:38:16 time: 0.5493 data_time: 0.0061 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0098 decode.acc_seg: 99.6521 aux.loss_ce: 0.0087 aux.acc_seg: 99.1826 +04/18 05:09:36 - mmengine - INFO - Iter(train) [ 18250/160000] lr: 8.9777e-03 eta: 21:37:49 time: 0.5515 data_time: 0.0057 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0105 decode.acc_seg: 99.6871 aux.loss_ce: 0.0086 aux.acc_seg: 99.2124 +04/18 05:10:03 - mmengine - INFO - Iter(train) [ 18300/160000] lr: 8.9749e-03 eta: 21:37:22 time: 0.5509 data_time: 0.0063 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0099 decode.acc_seg: 99.6368 aux.loss_ce: 0.0083 aux.acc_seg: 99.2934 +04/18 05:10:31 - mmengine - INFO - Iter(train) [ 18350/160000] lr: 8.9721e-03 eta: 21:36:55 time: 0.5511 data_time: 0.0058 memory: 7635 loss: 0.0199 decode.loss_ce: 0.0109 decode.acc_seg: 99.6661 aux.loss_ce: 0.0091 aux.acc_seg: 99.1489 +04/18 05:10:58 - mmengine - INFO - Iter(train) [ 18400/160000] lr: 8.9692e-03 eta: 21:36:28 time: 0.5509 data_time: 0.0062 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0097 decode.acc_seg: 99.6330 aux.loss_ce: 0.0082 aux.acc_seg: 99.2747 +04/18 05:11:26 - mmengine - INFO - Iter(train) [ 18450/160000] lr: 8.9664e-03 eta: 21:36:01 time: 0.5518 data_time: 0.0064 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0104 decode.acc_seg: 99.6861 aux.loss_ce: 0.0084 aux.acc_seg: 99.2772 +04/18 05:11:54 - mmengine - INFO - Iter(train) [ 18500/160000] lr: 8.9636e-03 eta: 21:35:35 time: 0.5521 data_time: 0.0065 memory: 7635 loss: 0.0193 decode.loss_ce: 0.0103 decode.acc_seg: 99.5523 aux.loss_ce: 0.0089 aux.acc_seg: 98.8003 +04/18 05:12:21 - mmengine - INFO - Iter(train) [ 18550/160000] lr: 8.9608e-03 eta: 21:35:08 time: 0.5509 data_time: 0.0065 memory: 7635 loss: 0.0202 decode.loss_ce: 0.0113 decode.acc_seg: 99.6337 aux.loss_ce: 0.0088 aux.acc_seg: 99.2847 +04/18 05:12:49 - mmengine - INFO - Iter(train) [ 18600/160000] lr: 8.9580e-03 eta: 21:34:41 time: 0.5507 data_time: 0.0067 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0096 decode.acc_seg: 99.6078 aux.loss_ce: 0.0083 aux.acc_seg: 99.1495 +04/18 05:13:16 - mmengine - INFO - Iter(train) [ 18650/160000] lr: 8.9551e-03 eta: 21:34:14 time: 0.5496 data_time: 0.0058 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0102 decode.acc_seg: 99.6562 aux.loss_ce: 0.0084 aux.acc_seg: 99.2815 +04/18 05:13:44 - mmengine - INFO - Iter(train) [ 18700/160000] lr: 8.9523e-03 eta: 21:33:46 time: 0.5494 data_time: 0.0060 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0100 decode.acc_seg: 99.6137 aux.loss_ce: 0.0088 aux.acc_seg: 99.2285 +04/18 05:14:11 - mmengine - INFO - Iter(train) [ 18750/160000] lr: 8.9495e-03 eta: 21:33:21 time: 0.5596 data_time: 0.0066 memory: 7635 loss: 0.0202 decode.loss_ce: 0.0113 decode.acc_seg: 99.4499 aux.loss_ce: 0.0090 aux.acc_seg: 98.6691 +04/18 05:14:39 - mmengine - INFO - Iter(train) [ 18800/160000] lr: 8.9467e-03 eta: 21:32:54 time: 0.5520 data_time: 0.0065 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0101 decode.acc_seg: 99.5760 aux.loss_ce: 0.0088 aux.acc_seg: 99.1145 +04/18 05:15:07 - mmengine - INFO - Iter(train) [ 18850/160000] lr: 8.9439e-03 eta: 21:32:28 time: 0.5528 data_time: 0.0071 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0101 decode.acc_seg: 99.6401 aux.loss_ce: 0.0088 aux.acc_seg: 99.1317 +04/18 05:15:34 - mmengine - INFO - Iter(train) [ 18900/160000] lr: 8.9411e-03 eta: 21:32:01 time: 0.5518 data_time: 0.0059 memory: 7635 loss: 0.0186 decode.loss_ce: 0.0102 decode.acc_seg: 99.5771 aux.loss_ce: 0.0085 aux.acc_seg: 98.9853 +04/18 05:16:02 - mmengine - INFO - Iter(train) [ 18950/160000] lr: 8.9382e-03 eta: 21:31:34 time: 0.5523 data_time: 0.0060 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0104 decode.acc_seg: 99.5743 aux.loss_ce: 0.0086 aux.acc_seg: 99.2046 +04/18 05:16:29 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 05:16:29 - mmengine - INFO - Iter(train) [ 19000/160000] lr: 8.9354e-03 eta: 21:31:07 time: 0.5514 data_time: 0.0054 memory: 7635 loss: 0.0194 decode.loss_ce: 0.0105 decode.acc_seg: 99.5222 aux.loss_ce: 0.0089 aux.acc_seg: 99.1184 +04/18 05:16:57 - mmengine - INFO - Iter(train) [ 19050/160000] lr: 8.9326e-03 eta: 21:30:40 time: 0.5505 data_time: 0.0065 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0096 decode.acc_seg: 99.6870 aux.loss_ce: 0.0078 aux.acc_seg: 99.2786 +04/18 05:17:24 - mmengine - INFO - Iter(train) [ 19100/160000] lr: 8.9298e-03 eta: 21:30:13 time: 0.5521 data_time: 0.0072 memory: 7635 loss: 0.0204 decode.loss_ce: 0.0112 decode.acc_seg: 99.5139 aux.loss_ce: 0.0092 aux.acc_seg: 98.9241 +04/18 05:17:52 - mmengine - INFO - Iter(train) [ 19150/160000] lr: 8.9270e-03 eta: 21:29:47 time: 0.5511 data_time: 0.0065 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0102 decode.acc_seg: 99.5030 aux.loss_ce: 0.0086 aux.acc_seg: 98.9733 +04/18 05:18:20 - mmengine - INFO - Iter(train) [ 19200/160000] lr: 8.9241e-03 eta: 21:29:21 time: 0.5534 data_time: 0.0062 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0105 decode.acc_seg: 99.6538 aux.loss_ce: 0.0087 aux.acc_seg: 99.2344 +04/18 05:18:47 - mmengine - INFO - Iter(train) [ 19250/160000] lr: 8.9213e-03 eta: 21:28:54 time: 0.5512 data_time: 0.0063 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0105 decode.acc_seg: 99.4306 aux.loss_ce: 0.0087 aux.acc_seg: 98.9574 +04/18 05:19:15 - mmengine - INFO - Iter(train) [ 19300/160000] lr: 8.9185e-03 eta: 21:28:27 time: 0.5517 data_time: 0.0066 memory: 7635 loss: 0.0202 decode.loss_ce: 0.0109 decode.acc_seg: 99.5718 aux.loss_ce: 0.0093 aux.acc_seg: 98.9916 +04/18 05:19:42 - mmengine - INFO - Iter(train) [ 19350/160000] lr: 8.9157e-03 eta: 21:28:01 time: 0.5517 data_time: 0.0066 memory: 7635 loss: 0.0193 decode.loss_ce: 0.0106 decode.acc_seg: 99.3697 aux.loss_ce: 0.0088 aux.acc_seg: 98.7476 +04/18 05:20:10 - mmengine - INFO - Iter(train) [ 19400/160000] lr: 8.9129e-03 eta: 21:27:34 time: 0.5510 data_time: 0.0062 memory: 7635 loss: 0.0190 decode.loss_ce: 0.0106 decode.acc_seg: 99.7349 aux.loss_ce: 0.0084 aux.acc_seg: 99.3744 +04/18 05:20:37 - mmengine - INFO - Iter(train) [ 19450/160000] lr: 8.9100e-03 eta: 21:27:07 time: 0.5523 data_time: 0.0065 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0106 decode.acc_seg: 99.6332 aux.loss_ce: 0.0085 aux.acc_seg: 99.2365 +04/18 05:21:05 - mmengine - INFO - Iter(train) [ 19500/160000] lr: 8.9072e-03 eta: 21:26:40 time: 0.5513 data_time: 0.0065 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0102 decode.acc_seg: 99.6510 aux.loss_ce: 0.0085 aux.acc_seg: 99.2619 +04/18 05:21:33 - mmengine - INFO - Iter(train) [ 19550/160000] lr: 8.9044e-03 eta: 21:26:13 time: 0.5508 data_time: 0.0071 memory: 7635 loss: 0.0186 decode.loss_ce: 0.0100 decode.acc_seg: 99.7253 aux.loss_ce: 0.0087 aux.acc_seg: 99.3011 +04/18 05:22:00 - mmengine - INFO - Iter(train) [ 19600/160000] lr: 8.9016e-03 eta: 21:25:47 time: 0.5517 data_time: 0.0061 memory: 7635 loss: 0.0190 decode.loss_ce: 0.0107 decode.acc_seg: 99.5919 aux.loss_ce: 0.0083 aux.acc_seg: 99.2030 +04/18 05:22:28 - mmengine - INFO - Iter(train) [ 19650/160000] lr: 8.8987e-03 eta: 21:25:20 time: 0.5516 data_time: 0.0066 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0092 decode.acc_seg: 99.6017 aux.loss_ce: 0.0079 aux.acc_seg: 99.0779 +04/18 05:22:55 - mmengine - INFO - Iter(train) [ 19700/160000] lr: 8.8959e-03 eta: 21:24:53 time: 0.5512 data_time: 0.0063 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0096 decode.acc_seg: 99.6183 aux.loss_ce: 0.0083 aux.acc_seg: 99.2425 +04/18 05:23:23 - mmengine - INFO - Iter(train) [ 19750/160000] lr: 8.8931e-03 eta: 21:24:26 time: 0.5523 data_time: 0.0059 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0105 decode.acc_seg: 99.6702 aux.loss_ce: 0.0084 aux.acc_seg: 99.3751 +04/18 05:23:50 - mmengine - INFO - Iter(train) [ 19800/160000] lr: 8.8903e-03 eta: 21:24:00 time: 0.5511 data_time: 0.0064 memory: 7635 loss: 0.0210 decode.loss_ce: 0.0119 decode.acc_seg: 99.6434 aux.loss_ce: 0.0091 aux.acc_seg: 99.2486 +04/18 05:24:18 - mmengine - INFO - Iter(train) [ 19850/160000] lr: 8.8875e-03 eta: 21:23:34 time: 0.5507 data_time: 0.0065 memory: 7635 loss: 0.0186 decode.loss_ce: 0.0103 decode.acc_seg: 99.5049 aux.loss_ce: 0.0083 aux.acc_seg: 99.0289 +04/18 05:24:46 - mmengine - INFO - Iter(train) [ 19900/160000] lr: 8.8846e-03 eta: 21:23:07 time: 0.5499 data_time: 0.0060 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0099 decode.acc_seg: 99.6104 aux.loss_ce: 0.0085 aux.acc_seg: 99.0854 +04/18 05:25:13 - mmengine - INFO - Iter(train) [ 19950/160000] lr: 8.8818e-03 eta: 21:22:41 time: 0.5517 data_time: 0.0063 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0104 decode.acc_seg: 99.5382 aux.loss_ce: 0.0088 aux.acc_seg: 98.8459 +04/18 05:25:41 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 05:25:41 - mmengine - INFO - Iter(train) [ 20000/160000] lr: 8.8790e-03 eta: 21:22:14 time: 0.5526 data_time: 0.0070 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0094 decode.acc_seg: 99.4759 aux.loss_ce: 0.0081 aux.acc_seg: 98.9945 +04/18 05:25:41 - mmengine - INFO - Saving checkpoint at 20000 iterations +04/18 05:25:45 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:07 time: 0.0464 data_time: 0.0013 memory: 1657 +04/18 05:25:47 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:04 time: 0.0463 data_time: 0.0013 memory: 1657 +04/18 05:25:50 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:02 time: 0.0468 data_time: 0.0015 memory: 1657 +04/18 05:25:52 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.0458 data_time: 0.0014 memory: 1657 +04/18 05:25:52 - mmengine - INFO - per class results: +04/18 05:25:52 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 99.05 | 99.56 | 99.52 | 99.49 | 99.56 | +| contrast | 79.18 | 87.62 | 88.38 | 89.15 | 87.62 | ++------------+-------+-------+--------+-----------+--------+ +04/18 05:25:52 - mmengine - INFO - Iter(val) [200/200] aAcc: 99.0800 mIoU: 89.1100 mAcc: 93.5900 mFscore: 93.9500 mPrecision: 94.3200 mRecall: 93.5900 data_time: 0.0014 time: 0.0465 +04/18 05:26:20 - mmengine - INFO - Iter(train) [ 20050/160000] lr: 8.8762e-03 eta: 21:21:48 time: 0.5514 data_time: 0.0057 memory: 7635 loss: 0.0210 decode.loss_ce: 0.0120 decode.acc_seg: 99.6273 aux.loss_ce: 0.0090 aux.acc_seg: 99.1736 +04/18 05:26:47 - mmengine - INFO - Iter(train) [ 20100/160000] lr: 8.8734e-03 eta: 21:21:21 time: 0.5521 data_time: 0.0061 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0096 decode.acc_seg: 99.5410 aux.loss_ce: 0.0083 aux.acc_seg: 99.0590 +04/18 05:27:15 - mmengine - INFO - Iter(train) [ 20150/160000] lr: 8.8705e-03 eta: 21:20:54 time: 0.5500 data_time: 0.0060 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0101 decode.acc_seg: 99.5425 aux.loss_ce: 0.0086 aux.acc_seg: 98.8522 +04/18 05:27:42 - mmengine - INFO - Iter(train) [ 20200/160000] lr: 8.8677e-03 eta: 21:20:27 time: 0.5511 data_time: 0.0059 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0095 decode.acc_seg: 99.7218 aux.loss_ce: 0.0082 aux.acc_seg: 99.2821 +04/18 05:28:10 - mmengine - INFO - Iter(train) [ 20250/160000] lr: 8.8649e-03 eta: 21:20:00 time: 0.5521 data_time: 0.0061 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0102 decode.acc_seg: 99.5185 aux.loss_ce: 0.0085 aux.acc_seg: 99.1319 +04/18 05:28:38 - mmengine - INFO - Iter(train) [ 20300/160000] lr: 8.8621e-03 eta: 21:19:34 time: 0.5510 data_time: 0.0057 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0094 decode.acc_seg: 99.6010 aux.loss_ce: 0.0085 aux.acc_seg: 98.9961 +04/18 05:29:05 - mmengine - INFO - Iter(train) [ 20350/160000] lr: 8.8592e-03 eta: 21:19:07 time: 0.5508 data_time: 0.0064 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0105 decode.acc_seg: 99.6302 aux.loss_ce: 0.0087 aux.acc_seg: 99.1408 +04/18 05:29:33 - mmengine - INFO - Iter(train) [ 20400/160000] lr: 8.8564e-03 eta: 21:18:40 time: 0.5528 data_time: 0.0057 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0106 decode.acc_seg: 99.6640 aux.loss_ce: 0.0089 aux.acc_seg: 99.3505 +04/18 05:30:00 - mmengine - INFO - Iter(train) [ 20450/160000] lr: 8.8536e-03 eta: 21:18:14 time: 0.5523 data_time: 0.0065 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0097 decode.acc_seg: 99.5840 aux.loss_ce: 0.0082 aux.acc_seg: 99.1034 +04/18 05:30:28 - mmengine - INFO - Iter(train) [ 20500/160000] lr: 8.8508e-03 eta: 21:17:48 time: 0.5521 data_time: 0.0067 memory: 7635 loss: 0.0181 decode.loss_ce: 0.0098 decode.acc_seg: 99.6780 aux.loss_ce: 0.0083 aux.acc_seg: 99.3616 +04/18 05:30:56 - mmengine - INFO - Iter(train) [ 20550/160000] lr: 8.8480e-03 eta: 21:17:21 time: 0.5506 data_time: 0.0061 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0117 decode.acc_seg: 99.4849 aux.loss_ce: 0.0092 aux.acc_seg: 98.8701 +04/18 05:31:23 - mmengine - INFO - Iter(train) [ 20600/160000] lr: 8.8451e-03 eta: 21:16:54 time: 0.5526 data_time: 0.0065 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0103 decode.acc_seg: 99.5954 aux.loss_ce: 0.0089 aux.acc_seg: 99.0963 +04/18 05:31:51 - mmengine - INFO - Iter(train) [ 20650/160000] lr: 8.8423e-03 eta: 21:16:27 time: 0.5537 data_time: 0.0063 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0096 decode.acc_seg: 99.7395 aux.loss_ce: 0.0083 aux.acc_seg: 99.3160 +04/18 05:32:18 - mmengine - INFO - Iter(train) [ 20700/160000] lr: 8.8395e-03 eta: 21:16:01 time: 0.5512 data_time: 0.0057 memory: 7635 loss: 0.0215 decode.loss_ce: 0.0122 decode.acc_seg: 99.4198 aux.loss_ce: 0.0093 aux.acc_seg: 98.7784 +04/18 05:32:46 - mmengine - INFO - Iter(train) [ 20750/160000] lr: 8.8367e-03 eta: 21:15:34 time: 0.5497 data_time: 0.0060 memory: 7635 loss: 0.0186 decode.loss_ce: 0.0101 decode.acc_seg: 99.6177 aux.loss_ce: 0.0085 aux.acc_seg: 99.2023 +04/18 05:33:14 - mmengine - INFO - Iter(train) [ 20800/160000] lr: 8.8338e-03 eta: 21:15:07 time: 0.5523 data_time: 0.0072 memory: 7635 loss: 0.0196 decode.loss_ce: 0.0106 decode.acc_seg: 99.5582 aux.loss_ce: 0.0089 aux.acc_seg: 99.0932 +04/18 05:33:41 - mmengine - INFO - Iter(train) [ 20850/160000] lr: 8.8310e-03 eta: 21:14:41 time: 0.5520 data_time: 0.0062 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0095 decode.acc_seg: 99.5618 aux.loss_ce: 0.0078 aux.acc_seg: 99.1218 +04/18 05:34:09 - mmengine - INFO - Iter(train) [ 20900/160000] lr: 8.8282e-03 eta: 21:14:15 time: 0.5509 data_time: 0.0064 memory: 7635 loss: 0.0193 decode.loss_ce: 0.0105 decode.acc_seg: 99.6472 aux.loss_ce: 0.0087 aux.acc_seg: 99.1203 +04/18 05:34:36 - mmengine - INFO - Iter(train) [ 20950/160000] lr: 8.8254e-03 eta: 21:13:48 time: 0.5526 data_time: 0.0068 memory: 7635 loss: 0.0200 decode.loss_ce: 0.0110 decode.acc_seg: 99.6499 aux.loss_ce: 0.0090 aux.acc_seg: 99.1760 +04/18 05:35:04 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 05:35:04 - mmengine - INFO - Iter(train) [ 21000/160000] lr: 8.8225e-03 eta: 21:13:21 time: 0.5513 data_time: 0.0058 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0097 decode.acc_seg: 99.5724 aux.loss_ce: 0.0079 aux.acc_seg: 99.1061 +04/18 05:35:32 - mmengine - INFO - Iter(train) [ 21050/160000] lr: 8.8197e-03 eta: 21:12:54 time: 0.5525 data_time: 0.0056 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0100 decode.acc_seg: 99.5926 aux.loss_ce: 0.0084 aux.acc_seg: 99.2459 +04/18 05:35:59 - mmengine - INFO - Iter(train) [ 21100/160000] lr: 8.8169e-03 eta: 21:12:28 time: 0.5527 data_time: 0.0062 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0100 decode.acc_seg: 99.5698 aux.loss_ce: 0.0083 aux.acc_seg: 99.0493 +04/18 05:36:27 - mmengine - INFO - Iter(train) [ 21150/160000] lr: 8.8141e-03 eta: 21:12:01 time: 0.5513 data_time: 0.0059 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0094 decode.acc_seg: 99.5793 aux.loss_ce: 0.0082 aux.acc_seg: 99.1617 +04/18 05:36:55 - mmengine - INFO - Iter(train) [ 21200/160000] lr: 8.8112e-03 eta: 21:11:34 time: 0.5520 data_time: 0.0065 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0090 decode.acc_seg: 99.5918 aux.loss_ce: 0.0078 aux.acc_seg: 99.0768 +04/18 05:37:22 - mmengine - INFO - Iter(train) [ 21250/160000] lr: 8.8084e-03 eta: 21:11:08 time: 0.5513 data_time: 0.0067 memory: 7635 loss: 0.0181 decode.loss_ce: 0.0095 decode.acc_seg: 99.5050 aux.loss_ce: 0.0086 aux.acc_seg: 98.8276 +04/18 05:37:50 - mmengine - INFO - Iter(train) [ 21300/160000] lr: 8.8056e-03 eta: 21:10:41 time: 0.5502 data_time: 0.0057 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0092 decode.acc_seg: 99.6953 aux.loss_ce: 0.0081 aux.acc_seg: 99.3996 +04/18 05:38:17 - mmengine - INFO - Iter(train) [ 21350/160000] lr: 8.8028e-03 eta: 21:10:14 time: 0.5533 data_time: 0.0067 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0094 decode.acc_seg: 99.4920 aux.loss_ce: 0.0080 aux.acc_seg: 98.9112 +04/18 05:38:45 - mmengine - INFO - Iter(train) [ 21400/160000] lr: 8.7999e-03 eta: 21:09:47 time: 0.5530 data_time: 0.0065 memory: 7635 loss: 0.0194 decode.loss_ce: 0.0110 decode.acc_seg: 99.5637 aux.loss_ce: 0.0084 aux.acc_seg: 99.2282 +04/18 05:39:13 - mmengine - INFO - Iter(train) [ 21450/160000] lr: 8.7971e-03 eta: 21:09:21 time: 0.5530 data_time: 0.0062 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0095 decode.acc_seg: 99.6919 aux.loss_ce: 0.0078 aux.acc_seg: 99.2809 +04/18 05:39:40 - mmengine - INFO - Iter(train) [ 21500/160000] lr: 8.7943e-03 eta: 21:08:54 time: 0.5519 data_time: 0.0063 memory: 7635 loss: 0.0224 decode.loss_ce: 0.0130 decode.acc_seg: 99.5067 aux.loss_ce: 0.0094 aux.acc_seg: 99.1303 +04/18 05:40:08 - mmengine - INFO - Iter(train) [ 21550/160000] lr: 8.7915e-03 eta: 21:08:28 time: 0.5595 data_time: 0.0060 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0092 decode.acc_seg: 99.6182 aux.loss_ce: 0.0080 aux.acc_seg: 99.1472 +04/18 05:40:35 - mmengine - INFO - Iter(train) [ 21600/160000] lr: 8.7886e-03 eta: 21:08:01 time: 0.5515 data_time: 0.0058 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0102 decode.acc_seg: 99.6891 aux.loss_ce: 0.0086 aux.acc_seg: 99.2013 +04/18 05:41:03 - mmengine - INFO - Iter(train) [ 21650/160000] lr: 8.7858e-03 eta: 21:07:34 time: 0.5515 data_time: 0.0060 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0100 decode.acc_seg: 99.5457 aux.loss_ce: 0.0083 aux.acc_seg: 98.9417 +04/18 05:41:31 - mmengine - INFO - Iter(train) [ 21700/160000] lr: 8.7830e-03 eta: 21:07:08 time: 0.5534 data_time: 0.0062 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0098 decode.acc_seg: 99.7236 aux.loss_ce: 0.0082 aux.acc_seg: 99.4304 +04/18 05:41:58 - mmengine - INFO - Iter(train) [ 21750/160000] lr: 8.7802e-03 eta: 21:06:42 time: 0.5539 data_time: 0.0070 memory: 7635 loss: 0.0208 decode.loss_ce: 0.0118 decode.acc_seg: 99.7230 aux.loss_ce: 0.0090 aux.acc_seg: 99.4003 +04/18 05:42:26 - mmengine - INFO - Iter(train) [ 21800/160000] lr: 8.7773e-03 eta: 21:06:15 time: 0.5525 data_time: 0.0064 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0104 decode.acc_seg: 99.6181 aux.loss_ce: 0.0086 aux.acc_seg: 99.2075 +04/18 05:42:54 - mmengine - INFO - Iter(train) [ 21850/160000] lr: 8.7745e-03 eta: 21:05:49 time: 0.5530 data_time: 0.0065 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0097 decode.acc_seg: 99.5230 aux.loss_ce: 0.0082 aux.acc_seg: 98.9681 +04/18 05:43:22 - mmengine - INFO - Iter(train) [ 21900/160000] lr: 8.7717e-03 eta: 21:05:23 time: 0.5621 data_time: 0.0071 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0096 decode.acc_seg: 99.5763 aux.loss_ce: 0.0084 aux.acc_seg: 99.0456 +04/18 05:43:49 - mmengine - INFO - Iter(train) [ 21950/160000] lr: 8.7689e-03 eta: 21:04:57 time: 0.5519 data_time: 0.0071 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0089 decode.acc_seg: 99.6490 aux.loss_ce: 0.0077 aux.acc_seg: 99.1749 +04/18 05:44:17 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 05:44:17 - mmengine - INFO - Iter(train) [ 22000/160000] lr: 8.7660e-03 eta: 21:04:30 time: 0.5518 data_time: 0.0057 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0105 decode.acc_seg: 99.6772 aux.loss_ce: 0.0085 aux.acc_seg: 99.2457 +04/18 05:44:45 - mmengine - INFO - Iter(train) [ 22050/160000] lr: 8.7632e-03 eta: 21:04:04 time: 0.5551 data_time: 0.0065 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0106 decode.acc_seg: 99.6129 aux.loss_ce: 0.0085 aux.acc_seg: 99.1654 +04/18 05:45:12 - mmengine - INFO - Iter(train) [ 22100/160000] lr: 8.7604e-03 eta: 21:03:37 time: 0.5533 data_time: 0.0061 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0099 decode.acc_seg: 99.5454 aux.loss_ce: 0.0083 aux.acc_seg: 99.1231 +04/18 05:45:40 - mmengine - INFO - Iter(train) [ 22150/160000] lr: 8.7576e-03 eta: 21:03:11 time: 0.5527 data_time: 0.0056 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0098 decode.acc_seg: 99.6480 aux.loss_ce: 0.0087 aux.acc_seg: 99.1308 +04/18 05:46:07 - mmengine - INFO - Iter(train) [ 22200/160000] lr: 8.7547e-03 eta: 21:02:45 time: 0.5541 data_time: 0.0061 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0104 decode.acc_seg: 99.5435 aux.loss_ce: 0.0085 aux.acc_seg: 98.9551 +04/18 05:46:35 - mmengine - INFO - Iter(train) [ 22250/160000] lr: 8.7519e-03 eta: 21:02:18 time: 0.5525 data_time: 0.0064 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0098 decode.acc_seg: 99.6762 aux.loss_ce: 0.0080 aux.acc_seg: 99.3376 +04/18 05:47:03 - mmengine - INFO - Iter(train) [ 22300/160000] lr: 8.7491e-03 eta: 21:01:52 time: 0.5527 data_time: 0.0060 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0101 decode.acc_seg: 99.4088 aux.loss_ce: 0.0087 aux.acc_seg: 98.6343 +04/18 05:47:30 - mmengine - INFO - Iter(train) [ 22350/160000] lr: 8.7463e-03 eta: 21:01:25 time: 0.5531 data_time: 0.0059 memory: 7635 loss: 0.0190 decode.loss_ce: 0.0104 decode.acc_seg: 99.5303 aux.loss_ce: 0.0086 aux.acc_seg: 98.9007 +04/18 05:47:58 - mmengine - INFO - Iter(train) [ 22400/160000] lr: 8.7434e-03 eta: 21:00:59 time: 0.5521 data_time: 0.0060 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0092 decode.acc_seg: 99.7247 aux.loss_ce: 0.0082 aux.acc_seg: 99.3110 +04/18 05:48:26 - mmengine - INFO - Iter(train) [ 22450/160000] lr: 8.7406e-03 eta: 21:00:32 time: 0.5526 data_time: 0.0067 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0103 decode.acc_seg: 99.5754 aux.loss_ce: 0.0089 aux.acc_seg: 98.9209 +04/18 05:48:53 - mmengine - INFO - Iter(train) [ 22500/160000] lr: 8.7378e-03 eta: 21:00:06 time: 0.5520 data_time: 0.0054 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0100 decode.acc_seg: 99.6193 aux.loss_ce: 0.0084 aux.acc_seg: 99.2662 +04/18 05:49:21 - mmengine - INFO - Iter(train) [ 22550/160000] lr: 8.7350e-03 eta: 20:59:39 time: 0.5530 data_time: 0.0060 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0100 decode.acc_seg: 99.6494 aux.loss_ce: 0.0085 aux.acc_seg: 99.2477 +04/18 05:49:49 - mmengine - INFO - Iter(train) [ 22600/160000] lr: 8.7321e-03 eta: 20:59:13 time: 0.5533 data_time: 0.0060 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0094 decode.acc_seg: 99.4904 aux.loss_ce: 0.0080 aux.acc_seg: 98.8156 +04/18 05:50:16 - mmengine - INFO - Iter(train) [ 22650/160000] lr: 8.7293e-03 eta: 20:58:47 time: 0.5545 data_time: 0.0056 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0102 decode.acc_seg: 99.6898 aux.loss_ce: 0.0089 aux.acc_seg: 99.2451 +04/18 05:50:44 - mmengine - INFO - Iter(train) [ 22700/160000] lr: 8.7265e-03 eta: 20:58:20 time: 0.5535 data_time: 0.0063 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0098 decode.acc_seg: 99.6656 aux.loss_ce: 0.0085 aux.acc_seg: 99.1815 +04/18 05:51:12 - mmengine - INFO - Iter(train) [ 22750/160000] lr: 8.7236e-03 eta: 20:57:53 time: 0.5536 data_time: 0.0064 memory: 7635 loss: 0.0186 decode.loss_ce: 0.0102 decode.acc_seg: 99.6284 aux.loss_ce: 0.0084 aux.acc_seg: 99.1419 +04/18 05:51:39 - mmengine - INFO - Iter(train) [ 22800/160000] lr: 8.7208e-03 eta: 20:57:27 time: 0.5536 data_time: 0.0070 memory: 7635 loss: 0.0218 decode.loss_ce: 0.0122 decode.acc_seg: 99.6552 aux.loss_ce: 0.0096 aux.acc_seg: 99.1392 +04/18 05:52:07 - mmengine - INFO - Iter(train) [ 22850/160000] lr: 8.7180e-03 eta: 20:57:00 time: 0.5536 data_time: 0.0061 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0107 decode.acc_seg: 99.5021 aux.loss_ce: 0.0085 aux.acc_seg: 99.2125 +04/18 05:52:35 - mmengine - INFO - Iter(train) [ 22900/160000] lr: 8.7152e-03 eta: 20:56:33 time: 0.5529 data_time: 0.0061 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0093 decode.acc_seg: 99.6344 aux.loss_ce: 0.0081 aux.acc_seg: 99.2249 +04/18 05:53:02 - mmengine - INFO - Iter(train) [ 22950/160000] lr: 8.7123e-03 eta: 20:56:07 time: 0.5539 data_time: 0.0060 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0092 decode.acc_seg: 99.5120 aux.loss_ce: 0.0080 aux.acc_seg: 99.1488 +04/18 05:53:30 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 05:53:30 - mmengine - INFO - Iter(train) [ 23000/160000] lr: 8.7095e-03 eta: 20:55:41 time: 0.5524 data_time: 0.0064 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0094 decode.acc_seg: 99.5515 aux.loss_ce: 0.0081 aux.acc_seg: 99.0876 +04/18 05:53:58 - mmengine - INFO - Iter(train) [ 23050/160000] lr: 8.7067e-03 eta: 20:55:15 time: 0.5530 data_time: 0.0063 memory: 7635 loss: 0.0181 decode.loss_ce: 0.0098 decode.acc_seg: 99.6166 aux.loss_ce: 0.0083 aux.acc_seg: 99.1988 +04/18 05:54:25 - mmengine - INFO - Iter(train) [ 23100/160000] lr: 8.7038e-03 eta: 20:54:48 time: 0.5527 data_time: 0.0062 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0095 decode.acc_seg: 99.7205 aux.loss_ce: 0.0082 aux.acc_seg: 99.4087 +04/18 05:54:53 - mmengine - INFO - Iter(train) [ 23150/160000] lr: 8.7010e-03 eta: 20:54:22 time: 0.5532 data_time: 0.0058 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0092 decode.acc_seg: 99.7180 aux.loss_ce: 0.0082 aux.acc_seg: 99.4081 +04/18 05:55:21 - mmengine - INFO - Iter(train) [ 23200/160000] lr: 8.6982e-03 eta: 20:53:55 time: 0.5530 data_time: 0.0065 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0100 decode.acc_seg: 99.6326 aux.loss_ce: 0.0084 aux.acc_seg: 99.2283 +04/18 05:55:48 - mmengine - INFO - Iter(train) [ 23250/160000] lr: 8.6954e-03 eta: 20:53:28 time: 0.5526 data_time: 0.0065 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0103 decode.acc_seg: 99.5256 aux.loss_ce: 0.0089 aux.acc_seg: 99.0128 +04/18 05:56:16 - mmengine - INFO - Iter(train) [ 23300/160000] lr: 8.6925e-03 eta: 20:53:01 time: 0.5530 data_time: 0.0065 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0092 decode.acc_seg: 99.6036 aux.loss_ce: 0.0078 aux.acc_seg: 99.1862 +04/18 05:56:44 - mmengine - INFO - Iter(train) [ 23350/160000] lr: 8.6897e-03 eta: 20:52:35 time: 0.5527 data_time: 0.0064 memory: 7635 loss: 0.0206 decode.loss_ce: 0.0116 decode.acc_seg: 99.6028 aux.loss_ce: 0.0090 aux.acc_seg: 99.2061 +04/18 05:57:11 - mmengine - INFO - Iter(train) [ 23400/160000] lr: 8.6869e-03 eta: 20:52:08 time: 0.5537 data_time: 0.0066 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0098 decode.acc_seg: 99.6199 aux.loss_ce: 0.0082 aux.acc_seg: 99.1998 +04/18 05:57:39 - mmengine - INFO - Iter(train) [ 23450/160000] lr: 8.6840e-03 eta: 20:51:42 time: 0.5533 data_time: 0.0062 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0091 decode.acc_seg: 99.6366 aux.loss_ce: 0.0086 aux.acc_seg: 98.9037 +04/18 05:58:07 - mmengine - INFO - Iter(train) [ 23500/160000] lr: 8.6812e-03 eta: 20:51:15 time: 0.5540 data_time: 0.0056 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0102 decode.acc_seg: 99.5785 aux.loss_ce: 0.0087 aux.acc_seg: 99.1714 +04/18 05:58:34 - mmengine - INFO - Iter(train) [ 23550/160000] lr: 8.6784e-03 eta: 20:50:49 time: 0.5528 data_time: 0.0071 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0094 decode.acc_seg: 99.6135 aux.loss_ce: 0.0080 aux.acc_seg: 99.1868 +04/18 05:59:02 - mmengine - INFO - Iter(train) [ 23600/160000] lr: 8.6756e-03 eta: 20:50:22 time: 0.5525 data_time: 0.0059 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0090 decode.acc_seg: 99.6444 aux.loss_ce: 0.0077 aux.acc_seg: 99.1253 +04/18 05:59:30 - mmengine - INFO - Iter(train) [ 23650/160000] lr: 8.6727e-03 eta: 20:49:56 time: 0.5540 data_time: 0.0066 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0088 decode.acc_seg: 99.6012 aux.loss_ce: 0.0077 aux.acc_seg: 99.2838 +04/18 05:59:58 - mmengine - INFO - Iter(train) [ 23700/160000] lr: 8.6699e-03 eta: 20:49:30 time: 0.5640 data_time: 0.0056 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0102 decode.acc_seg: 99.4806 aux.loss_ce: 0.0090 aux.acc_seg: 99.0386 +04/18 06:00:25 - mmengine - INFO - Iter(train) [ 23750/160000] lr: 8.6671e-03 eta: 20:49:04 time: 0.5543 data_time: 0.0064 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0089 decode.acc_seg: 99.6102 aux.loss_ce: 0.0080 aux.acc_seg: 99.1569 +04/18 06:00:53 - mmengine - INFO - Iter(train) [ 23800/160000] lr: 8.6642e-03 eta: 20:48:37 time: 0.5538 data_time: 0.0067 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0103 decode.acc_seg: 99.6158 aux.loss_ce: 0.0086 aux.acc_seg: 99.3257 +04/18 06:01:21 - mmengine - INFO - Iter(train) [ 23850/160000] lr: 8.6614e-03 eta: 20:48:10 time: 0.5531 data_time: 0.0060 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0105 decode.acc_seg: 99.5465 aux.loss_ce: 0.0086 aux.acc_seg: 99.0709 +04/18 06:01:48 - mmengine - INFO - Iter(train) [ 23900/160000] lr: 8.6586e-03 eta: 20:47:44 time: 0.5531 data_time: 0.0063 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0101 decode.acc_seg: 99.5972 aux.loss_ce: 0.0088 aux.acc_seg: 99.0875 +04/18 06:02:16 - mmengine - INFO - Iter(train) [ 23950/160000] lr: 8.6558e-03 eta: 20:47:17 time: 0.5530 data_time: 0.0058 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0101 decode.acc_seg: 99.3958 aux.loss_ce: 0.0082 aux.acc_seg: 98.8238 +04/18 06:02:44 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 06:02:44 - mmengine - INFO - Iter(train) [ 24000/160000] lr: 8.6529e-03 eta: 20:46:50 time: 0.5530 data_time: 0.0057 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0093 decode.acc_seg: 99.6724 aux.loss_ce: 0.0083 aux.acc_seg: 98.9887 +04/18 06:03:11 - mmengine - INFO - Iter(train) [ 24050/160000] lr: 8.6501e-03 eta: 20:46:24 time: 0.5532 data_time: 0.0067 memory: 7635 loss: 0.0223 decode.loss_ce: 0.0130 decode.acc_seg: 99.6555 aux.loss_ce: 0.0093 aux.acc_seg: 99.1264 +04/18 06:03:39 - mmengine - INFO - Iter(train) [ 24100/160000] lr: 8.6473e-03 eta: 20:45:58 time: 0.5545 data_time: 0.0057 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0105 decode.acc_seg: 99.4915 aux.loss_ce: 0.0085 aux.acc_seg: 98.9536 +04/18 06:04:07 - mmengine - INFO - Iter(train) [ 24150/160000] lr: 8.6444e-03 eta: 20:45:32 time: 0.5522 data_time: 0.0063 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0092 decode.acc_seg: 99.6381 aux.loss_ce: 0.0078 aux.acc_seg: 99.0611 +04/18 06:04:34 - mmengine - INFO - Iter(train) [ 24200/160000] lr: 8.6416e-03 eta: 20:45:05 time: 0.5539 data_time: 0.0060 memory: 7635 loss: 0.0209 decode.loss_ce: 0.0119 decode.acc_seg: 99.5026 aux.loss_ce: 0.0090 aux.acc_seg: 98.9678 +04/18 06:05:02 - mmengine - INFO - Iter(train) [ 24250/160000] lr: 8.6388e-03 eta: 20:44:38 time: 0.5525 data_time: 0.0059 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0092 decode.acc_seg: 99.5783 aux.loss_ce: 0.0078 aux.acc_seg: 99.0633 +04/18 06:05:30 - mmengine - INFO - Iter(train) [ 24300/160000] lr: 8.6359e-03 eta: 20:44:11 time: 0.5530 data_time: 0.0063 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0104 decode.acc_seg: 99.5662 aux.loss_ce: 0.0089 aux.acc_seg: 99.2458 +04/18 06:05:57 - mmengine - INFO - Iter(train) [ 24350/160000] lr: 8.6331e-03 eta: 20:43:45 time: 0.5536 data_time: 0.0068 memory: 7635 loss: 0.0197 decode.loss_ce: 0.0108 decode.acc_seg: 99.5570 aux.loss_ce: 0.0090 aux.acc_seg: 99.0466 +04/18 06:06:25 - mmengine - INFO - Iter(train) [ 24400/160000] lr: 8.6303e-03 eta: 20:43:18 time: 0.5525 data_time: 0.0070 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0083 decode.acc_seg: 99.6946 aux.loss_ce: 0.0073 aux.acc_seg: 99.3561 +04/18 06:06:53 - mmengine - INFO - Iter(train) [ 24450/160000] lr: 8.6275e-03 eta: 20:42:51 time: 0.5517 data_time: 0.0069 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0085 decode.acc_seg: 99.6692 aux.loss_ce: 0.0076 aux.acc_seg: 99.2901 +04/18 06:07:20 - mmengine - INFO - Iter(train) [ 24500/160000] lr: 8.6246e-03 eta: 20:42:24 time: 0.5531 data_time: 0.0062 memory: 7635 loss: 0.0181 decode.loss_ce: 0.0095 decode.acc_seg: 99.5336 aux.loss_ce: 0.0086 aux.acc_seg: 98.9326 +04/18 06:07:48 - mmengine - INFO - Iter(train) [ 24550/160000] lr: 8.6218e-03 eta: 20:41:58 time: 0.5536 data_time: 0.0062 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0101 decode.acc_seg: 99.7498 aux.loss_ce: 0.0083 aux.acc_seg: 99.3122 +04/18 06:08:16 - mmengine - INFO - Iter(train) [ 24600/160000] lr: 8.6190e-03 eta: 20:41:31 time: 0.5539 data_time: 0.0061 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0092 decode.acc_seg: 99.5259 aux.loss_ce: 0.0080 aux.acc_seg: 99.1107 +04/18 06:08:43 - mmengine - INFO - Iter(train) [ 24650/160000] lr: 8.6161e-03 eta: 20:41:05 time: 0.5535 data_time: 0.0063 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0098 decode.acc_seg: 99.2348 aux.loss_ce: 0.0085 aux.acc_seg: 98.5955 +04/18 06:09:11 - mmengine - INFO - Iter(train) [ 24700/160000] lr: 8.6133e-03 eta: 20:40:38 time: 0.5524 data_time: 0.0064 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0094 decode.acc_seg: 99.6115 aux.loss_ce: 0.0082 aux.acc_seg: 99.1424 +04/18 06:09:38 - mmengine - INFO - Iter(train) [ 24750/160000] lr: 8.6105e-03 eta: 20:40:11 time: 0.5536 data_time: 0.0060 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0092 decode.acc_seg: 99.5449 aux.loss_ce: 0.0078 aux.acc_seg: 99.0828 +04/18 06:10:06 - mmengine - INFO - Iter(train) [ 24800/160000] lr: 8.6076e-03 eta: 20:39:45 time: 0.5536 data_time: 0.0065 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0091 decode.acc_seg: 99.7059 aux.loss_ce: 0.0076 aux.acc_seg: 99.2213 +04/18 06:10:34 - mmengine - INFO - Iter(train) [ 24850/160000] lr: 8.6048e-03 eta: 20:39:18 time: 0.5524 data_time: 0.0063 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0089 decode.acc_seg: 99.6322 aux.loss_ce: 0.0079 aux.acc_seg: 99.3381 +04/18 06:11:01 - mmengine - INFO - Iter(train) [ 24900/160000] lr: 8.6020e-03 eta: 20:38:51 time: 0.5515 data_time: 0.0060 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0092 decode.acc_seg: 99.6635 aux.loss_ce: 0.0084 aux.acc_seg: 98.9992 +04/18 06:11:29 - mmengine - INFO - Iter(train) [ 24950/160000] lr: 8.5991e-03 eta: 20:38:24 time: 0.5543 data_time: 0.0067 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0091 decode.acc_seg: 99.5434 aux.loss_ce: 0.0083 aux.acc_seg: 98.9474 +04/18 06:11:57 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 06:11:57 - mmengine - INFO - Iter(train) [ 25000/160000] lr: 8.5963e-03 eta: 20:37:57 time: 0.5542 data_time: 0.0063 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0094 decode.acc_seg: 99.5576 aux.loss_ce: 0.0081 aux.acc_seg: 99.0066 +04/18 06:12:24 - mmengine - INFO - Iter(train) [ 25050/160000] lr: 8.5935e-03 eta: 20:37:30 time: 0.5531 data_time: 0.0060 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0090 decode.acc_seg: 99.6518 aux.loss_ce: 0.0082 aux.acc_seg: 99.0782 +04/18 06:12:52 - mmengine - INFO - Iter(train) [ 25100/160000] lr: 8.5906e-03 eta: 20:37:04 time: 0.5525 data_time: 0.0060 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0094 decode.acc_seg: 99.6090 aux.loss_ce: 0.0083 aux.acc_seg: 99.0969 +04/18 06:13:20 - mmengine - INFO - Iter(train) [ 25150/160000] lr: 8.5878e-03 eta: 20:36:38 time: 0.5543 data_time: 0.0066 memory: 7635 loss: 0.0234 decode.loss_ce: 0.0139 decode.acc_seg: 99.2514 aux.loss_ce: 0.0096 aux.acc_seg: 98.7681 +04/18 06:13:48 - mmengine - INFO - Iter(train) [ 25200/160000] lr: 8.5850e-03 eta: 20:36:11 time: 0.5537 data_time: 0.0062 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0105 decode.acc_seg: 99.5507 aux.loss_ce: 0.0087 aux.acc_seg: 99.0808 +04/18 06:14:15 - mmengine - INFO - Iter(train) [ 25250/160000] lr: 8.5821e-03 eta: 20:35:45 time: 0.5536 data_time: 0.0063 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0100 decode.acc_seg: 99.5605 aux.loss_ce: 0.0089 aux.acc_seg: 98.8799 +04/18 06:14:43 - mmengine - INFO - Iter(train) [ 25300/160000] lr: 8.5793e-03 eta: 20:35:18 time: 0.5530 data_time: 0.0060 memory: 7635 loss: 0.0215 decode.loss_ce: 0.0119 decode.acc_seg: 99.6689 aux.loss_ce: 0.0096 aux.acc_seg: 99.2425 +04/18 06:15:11 - mmengine - INFO - Iter(train) [ 25350/160000] lr: 8.5765e-03 eta: 20:34:52 time: 0.5532 data_time: 0.0062 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0091 decode.acc_seg: 99.7281 aux.loss_ce: 0.0076 aux.acc_seg: 99.4073 +04/18 06:15:38 - mmengine - INFO - Iter(train) [ 25400/160000] lr: 8.5736e-03 eta: 20:34:25 time: 0.5546 data_time: 0.0071 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0095 decode.acc_seg: 99.6294 aux.loss_ce: 0.0087 aux.acc_seg: 99.0431 +04/18 06:16:06 - mmengine - INFO - Iter(train) [ 25450/160000] lr: 8.5708e-03 eta: 20:33:58 time: 0.5524 data_time: 0.0063 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0102 decode.acc_seg: 99.6443 aux.loss_ce: 0.0087 aux.acc_seg: 99.1878 +04/18 06:16:34 - mmengine - INFO - Iter(train) [ 25500/160000] lr: 8.5680e-03 eta: 20:33:32 time: 0.5533 data_time: 0.0062 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0103 decode.acc_seg: 99.6306 aux.loss_ce: 0.0089 aux.acc_seg: 99.1586 +04/18 06:17:01 - mmengine - INFO - Iter(train) [ 25550/160000] lr: 8.5651e-03 eta: 20:33:05 time: 0.5539 data_time: 0.0062 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0106 decode.acc_seg: 99.6760 aux.loss_ce: 0.0085 aux.acc_seg: 99.2848 +04/18 06:17:29 - mmengine - INFO - Iter(train) [ 25600/160000] lr: 8.5623e-03 eta: 20:32:38 time: 0.5533 data_time: 0.0061 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0098 decode.acc_seg: 99.6841 aux.loss_ce: 0.0084 aux.acc_seg: 99.1635 +04/18 06:17:57 - mmengine - INFO - Iter(train) [ 25650/160000] lr: 8.5595e-03 eta: 20:32:12 time: 0.5549 data_time: 0.0061 memory: 7635 loss: 0.0194 decode.loss_ce: 0.0107 decode.acc_seg: 99.5690 aux.loss_ce: 0.0086 aux.acc_seg: 98.9656 +04/18 06:18:24 - mmengine - INFO - Iter(train) [ 25700/160000] lr: 8.5566e-03 eta: 20:31:45 time: 0.5554 data_time: 0.0069 memory: 7635 loss: 0.0193 decode.loss_ce: 0.0107 decode.acc_seg: 99.6641 aux.loss_ce: 0.0086 aux.acc_seg: 99.2069 +04/18 06:18:52 - mmengine - INFO - Iter(train) [ 25750/160000] lr: 8.5538e-03 eta: 20:31:19 time: 0.5546 data_time: 0.0060 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0090 decode.acc_seg: 99.5224 aux.loss_ce: 0.0083 aux.acc_seg: 98.9250 +04/18 06:19:20 - mmengine - INFO - Iter(train) [ 25800/160000] lr: 8.5510e-03 eta: 20:30:52 time: 0.5558 data_time: 0.0058 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0091 decode.acc_seg: 99.5760 aux.loss_ce: 0.0084 aux.acc_seg: 98.9920 +04/18 06:19:48 - mmengine - INFO - Iter(train) [ 25850/160000] lr: 8.5481e-03 eta: 20:30:26 time: 0.5557 data_time: 0.0065 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0100 decode.acc_seg: 99.6197 aux.loss_ce: 0.0085 aux.acc_seg: 99.2267 +04/18 06:20:15 - mmengine - INFO - Iter(train) [ 25900/160000] lr: 8.5453e-03 eta: 20:30:00 time: 0.5527 data_time: 0.0061 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0094 decode.acc_seg: 99.7183 aux.loss_ce: 0.0083 aux.acc_seg: 99.2774 +04/18 06:20:43 - mmengine - INFO - Iter(train) [ 25950/160000] lr: 8.5425e-03 eta: 20:29:33 time: 0.5543 data_time: 0.0056 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0098 decode.acc_seg: 99.6219 aux.loss_ce: 0.0090 aux.acc_seg: 99.0978 +04/18 06:21:11 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 06:21:11 - mmengine - INFO - Iter(train) [ 26000/160000] lr: 8.5396e-03 eta: 20:29:07 time: 0.5536 data_time: 0.0067 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0102 decode.acc_seg: 99.6169 aux.loss_ce: 0.0086 aux.acc_seg: 99.1028 +04/18 06:21:39 - mmengine - INFO - Iter(train) [ 26050/160000] lr: 8.5368e-03 eta: 20:28:40 time: 0.5528 data_time: 0.0073 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0100 decode.acc_seg: 99.6387 aux.loss_ce: 0.0083 aux.acc_seg: 99.1585 +04/18 06:22:06 - mmengine - INFO - Iter(train) [ 26100/160000] lr: 8.5340e-03 eta: 20:28:13 time: 0.5531 data_time: 0.0060 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0095 decode.acc_seg: 99.6833 aux.loss_ce: 0.0084 aux.acc_seg: 99.2471 +04/18 06:22:34 - mmengine - INFO - Iter(train) [ 26150/160000] lr: 8.5311e-03 eta: 20:27:47 time: 0.5529 data_time: 0.0067 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0096 decode.acc_seg: 99.6755 aux.loss_ce: 0.0092 aux.acc_seg: 98.9544 +04/18 06:23:02 - mmengine - INFO - Iter(train) [ 26200/160000] lr: 8.5283e-03 eta: 20:27:20 time: 0.5529 data_time: 0.0063 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0086 decode.acc_seg: 99.6167 aux.loss_ce: 0.0081 aux.acc_seg: 99.0394 +04/18 06:23:29 - mmengine - INFO - Iter(train) [ 26250/160000] lr: 8.5255e-03 eta: 20:26:54 time: 0.5622 data_time: 0.0070 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0097 decode.acc_seg: 99.5710 aux.loss_ce: 0.0085 aux.acc_seg: 98.9657 +04/18 06:23:57 - mmengine - INFO - Iter(train) [ 26300/160000] lr: 8.5226e-03 eta: 20:26:28 time: 0.5558 data_time: 0.0070 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0097 decode.acc_seg: 99.5872 aux.loss_ce: 0.0086 aux.acc_seg: 99.1512 +04/18 06:24:25 - mmengine - INFO - Iter(train) [ 26350/160000] lr: 8.5198e-03 eta: 20:26:01 time: 0.5544 data_time: 0.0065 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0091 decode.acc_seg: 99.5209 aux.loss_ce: 0.0078 aux.acc_seg: 99.1055 +04/18 06:24:53 - mmengine - INFO - Iter(train) [ 26400/160000] lr: 8.5170e-03 eta: 20:25:35 time: 0.5538 data_time: 0.0059 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0089 decode.acc_seg: 99.6354 aux.loss_ce: 0.0077 aux.acc_seg: 99.2619 +04/18 06:25:20 - mmengine - INFO - Iter(train) [ 26450/160000] lr: 8.5141e-03 eta: 20:25:08 time: 0.5545 data_time: 0.0063 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0082 decode.acc_seg: 99.6066 aux.loss_ce: 0.0073 aux.acc_seg: 99.0895 +04/18 06:25:48 - mmengine - INFO - Iter(train) [ 26500/160000] lr: 8.5113e-03 eta: 20:24:42 time: 0.5551 data_time: 0.0063 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0089 decode.acc_seg: 99.6694 aux.loss_ce: 0.0080 aux.acc_seg: 99.1294 +04/18 06:26:16 - mmengine - INFO - Iter(train) [ 26550/160000] lr: 8.5085e-03 eta: 20:24:15 time: 0.5547 data_time: 0.0063 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0089 decode.acc_seg: 99.6081 aux.loss_ce: 0.0078 aux.acc_seg: 99.1234 +04/18 06:26:44 - mmengine - INFO - Iter(train) [ 26600/160000] lr: 8.5056e-03 eta: 20:23:49 time: 0.5548 data_time: 0.0067 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0093 decode.acc_seg: 99.5845 aux.loss_ce: 0.0081 aux.acc_seg: 99.1042 +04/18 06:27:11 - mmengine - INFO - Iter(train) [ 26650/160000] lr: 8.5028e-03 eta: 20:23:23 time: 0.5546 data_time: 0.0071 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0097 decode.acc_seg: 99.5620 aux.loss_ce: 0.0082 aux.acc_seg: 99.1332 +04/18 06:27:39 - mmengine - INFO - Iter(train) [ 26700/160000] lr: 8.5000e-03 eta: 20:22:56 time: 0.5545 data_time: 0.0066 memory: 7635 loss: 0.0193 decode.loss_ce: 0.0105 decode.acc_seg: 99.4620 aux.loss_ce: 0.0089 aux.acc_seg: 98.9699 +04/18 06:28:07 - mmengine - INFO - Iter(train) [ 26750/160000] lr: 8.4971e-03 eta: 20:22:30 time: 0.5546 data_time: 0.0060 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0106 decode.acc_seg: 99.5955 aux.loss_ce: 0.0086 aux.acc_seg: 99.1364 +04/18 06:28:35 - mmengine - INFO - Iter(train) [ 26800/160000] lr: 8.4943e-03 eta: 20:22:03 time: 0.5546 data_time: 0.0059 memory: 7635 loss: 0.0190 decode.loss_ce: 0.0105 decode.acc_seg: 99.6692 aux.loss_ce: 0.0085 aux.acc_seg: 99.2268 +04/18 06:29:02 - mmengine - INFO - Iter(train) [ 26850/160000] lr: 8.4914e-03 eta: 20:21:37 time: 0.5531 data_time: 0.0066 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0102 decode.acc_seg: 99.5187 aux.loss_ce: 0.0086 aux.acc_seg: 99.0642 +04/18 06:29:30 - mmengine - INFO - Iter(train) [ 26900/160000] lr: 8.4886e-03 eta: 20:21:11 time: 0.5547 data_time: 0.0061 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0092 decode.acc_seg: 99.6379 aux.loss_ce: 0.0080 aux.acc_seg: 99.1110 +04/18 06:29:58 - mmengine - INFO - Iter(train) [ 26950/160000] lr: 8.4858e-03 eta: 20:20:45 time: 0.5555 data_time: 0.0068 memory: 7635 loss: 0.0181 decode.loss_ce: 0.0095 decode.acc_seg: 99.6460 aux.loss_ce: 0.0086 aux.acc_seg: 99.1088 +04/18 06:30:26 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 06:30:26 - mmengine - INFO - Iter(train) [ 27000/160000] lr: 8.4829e-03 eta: 20:20:18 time: 0.5542 data_time: 0.0074 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0101 decode.acc_seg: 99.6244 aux.loss_ce: 0.0086 aux.acc_seg: 99.0749 +04/18 06:30:53 - mmengine - INFO - Iter(train) [ 27050/160000] lr: 8.4801e-03 eta: 20:19:51 time: 0.5542 data_time: 0.0070 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0089 decode.acc_seg: 99.6664 aux.loss_ce: 0.0082 aux.acc_seg: 99.1160 +04/18 06:31:21 - mmengine - INFO - Iter(train) [ 27100/160000] lr: 8.4773e-03 eta: 20:19:25 time: 0.5546 data_time: 0.0066 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0099 decode.acc_seg: 99.4372 aux.loss_ce: 0.0082 aux.acc_seg: 98.8480 +04/18 06:31:49 - mmengine - INFO - Iter(train) [ 27150/160000] lr: 8.4744e-03 eta: 20:18:58 time: 0.5541 data_time: 0.0056 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0083 decode.acc_seg: 99.7565 aux.loss_ce: 0.0076 aux.acc_seg: 99.4385 +04/18 06:32:17 - mmengine - INFO - Iter(train) [ 27200/160000] lr: 8.4716e-03 eta: 20:18:32 time: 0.5542 data_time: 0.0063 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0088 decode.acc_seg: 99.5640 aux.loss_ce: 0.0081 aux.acc_seg: 98.9797 +04/18 06:32:44 - mmengine - INFO - Iter(train) [ 27250/160000] lr: 8.4688e-03 eta: 20:18:05 time: 0.5553 data_time: 0.0062 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0101 decode.acc_seg: 99.5849 aux.loss_ce: 0.0082 aux.acc_seg: 99.2076 +04/18 06:33:12 - mmengine - INFO - Iter(train) [ 27300/160000] lr: 8.4659e-03 eta: 20:17:39 time: 0.5558 data_time: 0.0065 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0096 decode.acc_seg: 99.4665 aux.loss_ce: 0.0082 aux.acc_seg: 99.2051 +04/18 06:33:40 - mmengine - INFO - Iter(train) [ 27350/160000] lr: 8.4631e-03 eta: 20:17:13 time: 0.5539 data_time: 0.0066 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0096 decode.acc_seg: 99.5988 aux.loss_ce: 0.0082 aux.acc_seg: 98.9725 +04/18 06:34:08 - mmengine - INFO - Iter(train) [ 27400/160000] lr: 8.4602e-03 eta: 20:16:47 time: 0.5553 data_time: 0.0066 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0094 decode.acc_seg: 99.6114 aux.loss_ce: 0.0078 aux.acc_seg: 99.2443 +04/18 06:34:35 - mmengine - INFO - Iter(train) [ 27450/160000] lr: 8.4574e-03 eta: 20:16:20 time: 0.5562 data_time: 0.0062 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0091 decode.acc_seg: 99.7229 aux.loss_ce: 0.0081 aux.acc_seg: 99.2241 +04/18 06:35:03 - mmengine - INFO - Iter(train) [ 27500/160000] lr: 8.4546e-03 eta: 20:15:54 time: 0.5559 data_time: 0.0057 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0098 decode.acc_seg: 99.5677 aux.loss_ce: 0.0083 aux.acc_seg: 99.1057 +04/18 06:35:31 - mmengine - INFO - Iter(train) [ 27550/160000] lr: 8.4517e-03 eta: 20:15:27 time: 0.5555 data_time: 0.0067 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0092 decode.acc_seg: 99.7557 aux.loss_ce: 0.0083 aux.acc_seg: 99.3876 +04/18 06:35:59 - mmengine - INFO - Iter(train) [ 27600/160000] lr: 8.4489e-03 eta: 20:15:01 time: 0.5540 data_time: 0.0062 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0094 decode.acc_seg: 99.7061 aux.loss_ce: 0.0083 aux.acc_seg: 99.2312 +04/18 06:36:26 - mmengine - INFO - Iter(train) [ 27650/160000] lr: 8.4461e-03 eta: 20:14:34 time: 0.5547 data_time: 0.0068 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0101 decode.acc_seg: 99.4464 aux.loss_ce: 0.0090 aux.acc_seg: 98.5444 +04/18 06:36:54 - mmengine - INFO - Iter(train) [ 27700/160000] lr: 8.4432e-03 eta: 20:14:08 time: 0.5551 data_time: 0.0062 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0094 decode.acc_seg: 99.6277 aux.loss_ce: 0.0081 aux.acc_seg: 99.2190 +04/18 06:37:22 - mmengine - INFO - Iter(train) [ 27750/160000] lr: 8.4404e-03 eta: 20:13:41 time: 0.5538 data_time: 0.0061 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0096 decode.acc_seg: 99.5312 aux.loss_ce: 0.0086 aux.acc_seg: 98.8728 +04/18 06:37:50 - mmengine - INFO - Iter(train) [ 27800/160000] lr: 8.4375e-03 eta: 20:13:14 time: 0.5547 data_time: 0.0061 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0098 decode.acc_seg: 99.6878 aux.loss_ce: 0.0086 aux.acc_seg: 99.2142 +04/18 06:38:17 - mmengine - INFO - Iter(train) [ 27850/160000] lr: 8.4347e-03 eta: 20:12:48 time: 0.5538 data_time: 0.0063 memory: 7635 loss: 0.0217 decode.loss_ce: 0.0124 decode.acc_seg: 99.5290 aux.loss_ce: 0.0093 aux.acc_seg: 99.2845 +04/18 06:38:45 - mmengine - INFO - Iter(train) [ 27900/160000] lr: 8.4319e-03 eta: 20:12:21 time: 0.5535 data_time: 0.0064 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0087 decode.acc_seg: 99.6517 aux.loss_ce: 0.0077 aux.acc_seg: 99.2413 +04/18 06:39:13 - mmengine - INFO - Iter(train) [ 27950/160000] lr: 8.4290e-03 eta: 20:11:54 time: 0.5532 data_time: 0.0061 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0102 decode.acc_seg: 99.5343 aux.loss_ce: 0.0086 aux.acc_seg: 99.1739 +04/18 06:39:40 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 06:39:40 - mmengine - INFO - Iter(train) [ 28000/160000] lr: 8.4262e-03 eta: 20:11:27 time: 0.5533 data_time: 0.0064 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0089 decode.acc_seg: 99.5786 aux.loss_ce: 0.0079 aux.acc_seg: 99.1202 +04/18 06:40:08 - mmengine - INFO - Iter(train) [ 28050/160000] lr: 8.4233e-03 eta: 20:11:01 time: 0.5534 data_time: 0.0062 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0098 decode.acc_seg: 99.7616 aux.loss_ce: 0.0082 aux.acc_seg: 99.3387 +04/18 06:40:36 - mmengine - INFO - Iter(train) [ 28100/160000] lr: 8.4205e-03 eta: 20:10:34 time: 0.5538 data_time: 0.0060 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0098 decode.acc_seg: 99.4782 aux.loss_ce: 0.0085 aux.acc_seg: 99.0201 +04/18 06:41:03 - mmengine - INFO - Iter(train) [ 28150/160000] lr: 8.4177e-03 eta: 20:10:07 time: 0.5535 data_time: 0.0063 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0097 decode.acc_seg: 99.6655 aux.loss_ce: 0.0083 aux.acc_seg: 99.2508 +04/18 06:41:31 - mmengine - INFO - Iter(train) [ 28200/160000] lr: 8.4148e-03 eta: 20:09:40 time: 0.5526 data_time: 0.0061 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0101 decode.acc_seg: 99.4870 aux.loss_ce: 0.0088 aux.acc_seg: 99.1462 +04/18 06:41:59 - mmengine - INFO - Iter(train) [ 28250/160000] lr: 8.4120e-03 eta: 20:09:13 time: 0.5547 data_time: 0.0063 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0094 decode.acc_seg: 99.5909 aux.loss_ce: 0.0082 aux.acc_seg: 99.0626 +04/18 06:42:27 - mmengine - INFO - Iter(train) [ 28300/160000] lr: 8.4092e-03 eta: 20:08:47 time: 0.5556 data_time: 0.0066 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0093 decode.acc_seg: 99.6944 aux.loss_ce: 0.0080 aux.acc_seg: 99.3399 +04/18 06:42:54 - mmengine - INFO - Iter(train) [ 28350/160000] lr: 8.4063e-03 eta: 20:08:20 time: 0.5536 data_time: 0.0059 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0093 decode.acc_seg: 99.7280 aux.loss_ce: 0.0083 aux.acc_seg: 99.4580 +04/18 06:43:22 - mmengine - INFO - Iter(train) [ 28400/160000] lr: 8.4035e-03 eta: 20:07:54 time: 0.5551 data_time: 0.0069 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0089 decode.acc_seg: 99.6412 aux.loss_ce: 0.0078 aux.acc_seg: 99.1178 +04/18 06:43:50 - mmengine - INFO - Iter(train) [ 28450/160000] lr: 8.4006e-03 eta: 20:07:28 time: 0.5532 data_time: 0.0059 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0097 decode.acc_seg: 99.5481 aux.loss_ce: 0.0085 aux.acc_seg: 99.0436 +04/18 06:44:18 - mmengine - INFO - Iter(train) [ 28500/160000] lr: 8.3978e-03 eta: 20:07:02 time: 0.5532 data_time: 0.0066 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0099 decode.acc_seg: 99.5734 aux.loss_ce: 0.0088 aux.acc_seg: 99.0263 +04/18 06:44:46 - mmengine - INFO - Iter(train) [ 28550/160000] lr: 8.3950e-03 eta: 20:06:35 time: 0.5546 data_time: 0.0060 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0091 decode.acc_seg: 99.6369 aux.loss_ce: 0.0081 aux.acc_seg: 99.3268 +04/18 06:45:13 - mmengine - INFO - Iter(train) [ 28600/160000] lr: 8.3921e-03 eta: 20:06:08 time: 0.5554 data_time: 0.0072 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0095 decode.acc_seg: 99.5734 aux.loss_ce: 0.0081 aux.acc_seg: 98.9872 +04/18 06:45:41 - mmengine - INFO - Iter(train) [ 28650/160000] lr: 8.3893e-03 eta: 20:05:42 time: 0.5537 data_time: 0.0061 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0095 decode.acc_seg: 99.7258 aux.loss_ce: 0.0082 aux.acc_seg: 99.3052 +04/18 06:46:09 - mmengine - INFO - Iter(train) [ 28700/160000] lr: 8.3864e-03 eta: 20:05:15 time: 0.5554 data_time: 0.0065 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0080 decode.acc_seg: 99.5783 aux.loss_ce: 0.0076 aux.acc_seg: 99.2734 +04/18 06:46:36 - mmengine - INFO - Iter(train) [ 28750/160000] lr: 8.3836e-03 eta: 20:04:48 time: 0.5526 data_time: 0.0068 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0103 decode.acc_seg: 99.5633 aux.loss_ce: 0.0090 aux.acc_seg: 99.0564 +04/18 06:47:04 - mmengine - INFO - Iter(train) [ 28800/160000] lr: 8.3808e-03 eta: 20:04:21 time: 0.5524 data_time: 0.0065 memory: 7635 loss: 0.0196 decode.loss_ce: 0.0109 decode.acc_seg: 99.5342 aux.loss_ce: 0.0088 aux.acc_seg: 99.1306 +04/18 06:47:32 - mmengine - INFO - Iter(train) [ 28850/160000] lr: 8.3779e-03 eta: 20:03:54 time: 0.5549 data_time: 0.0061 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0094 decode.acc_seg: 99.6271 aux.loss_ce: 0.0084 aux.acc_seg: 99.1467 +04/18 06:48:00 - mmengine - INFO - Iter(train) [ 28900/160000] lr: 8.3751e-03 eta: 20:03:27 time: 0.5540 data_time: 0.0065 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0097 decode.acc_seg: 99.5158 aux.loss_ce: 0.0085 aux.acc_seg: 98.9241 +04/18 06:48:27 - mmengine - INFO - Iter(train) [ 28950/160000] lr: 8.3722e-03 eta: 20:03:01 time: 0.5547 data_time: 0.0063 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0080 decode.acc_seg: 99.6894 aux.loss_ce: 0.0070 aux.acc_seg: 99.1712 +04/18 06:48:55 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 06:48:55 - mmengine - INFO - Iter(train) [ 29000/160000] lr: 8.3694e-03 eta: 20:02:34 time: 0.5523 data_time: 0.0068 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0096 decode.acc_seg: 99.5784 aux.loss_ce: 0.0083 aux.acc_seg: 99.1201 +04/18 06:49:23 - mmengine - INFO - Iter(train) [ 29050/160000] lr: 8.3666e-03 eta: 20:02:07 time: 0.5536 data_time: 0.0062 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0092 decode.acc_seg: 99.6448 aux.loss_ce: 0.0083 aux.acc_seg: 99.1606 +04/18 06:49:50 - mmengine - INFO - Iter(train) [ 29100/160000] lr: 8.3637e-03 eta: 20:01:40 time: 0.5520 data_time: 0.0066 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0092 decode.acc_seg: 99.5769 aux.loss_ce: 0.0080 aux.acc_seg: 99.1119 +04/18 06:50:18 - mmengine - INFO - Iter(train) [ 29150/160000] lr: 8.3609e-03 eta: 20:01:13 time: 0.5529 data_time: 0.0066 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0087 decode.acc_seg: 99.6526 aux.loss_ce: 0.0078 aux.acc_seg: 99.1117 +04/18 06:50:46 - mmengine - INFO - Iter(train) [ 29200/160000] lr: 8.3580e-03 eta: 20:00:46 time: 0.5541 data_time: 0.0058 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0080 decode.acc_seg: 99.6968 aux.loss_ce: 0.0073 aux.acc_seg: 99.3466 +04/18 06:51:13 - mmengine - INFO - Iter(train) [ 29250/160000] lr: 8.3552e-03 eta: 20:00:20 time: 0.5542 data_time: 0.0061 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0095 decode.acc_seg: 99.6399 aux.loss_ce: 0.0080 aux.acc_seg: 99.0940 +04/18 06:51:41 - mmengine - INFO - Iter(train) [ 29300/160000] lr: 8.3524e-03 eta: 19:59:53 time: 0.5537 data_time: 0.0070 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0089 decode.acc_seg: 99.6080 aux.loss_ce: 0.0080 aux.acc_seg: 98.9253 +04/18 06:52:09 - mmengine - INFO - Iter(train) [ 29350/160000] lr: 8.3495e-03 eta: 19:59:26 time: 0.5539 data_time: 0.0070 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0095 decode.acc_seg: 99.6967 aux.loss_ce: 0.0081 aux.acc_seg: 99.2575 +04/18 06:52:36 - mmengine - INFO - Iter(train) [ 29400/160000] lr: 8.3467e-03 eta: 19:58:58 time: 0.5512 data_time: 0.0062 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0097 decode.acc_seg: 99.6092 aux.loss_ce: 0.0085 aux.acc_seg: 99.0195 +04/18 06:53:04 - mmengine - INFO - Iter(train) [ 29450/160000] lr: 8.3438e-03 eta: 19:58:32 time: 0.5548 data_time: 0.0058 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0088 decode.acc_seg: 99.6880 aux.loss_ce: 0.0078 aux.acc_seg: 99.3057 +04/18 06:53:32 - mmengine - INFO - Iter(train) [ 29500/160000] lr: 8.3410e-03 eta: 19:58:05 time: 0.5525 data_time: 0.0065 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0094 decode.acc_seg: 99.7135 aux.loss_ce: 0.0081 aux.acc_seg: 99.4377 +04/18 06:54:00 - mmengine - INFO - Iter(train) [ 29550/160000] lr: 8.3381e-03 eta: 19:57:38 time: 0.5548 data_time: 0.0067 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0092 decode.acc_seg: 99.5240 aux.loss_ce: 0.0082 aux.acc_seg: 99.0030 +04/18 06:54:27 - mmengine - INFO - Iter(train) [ 29600/160000] lr: 8.3353e-03 eta: 19:57:11 time: 0.5544 data_time: 0.0068 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0097 decode.acc_seg: 99.7782 aux.loss_ce: 0.0088 aux.acc_seg: 99.4354 +04/18 06:54:55 - mmengine - INFO - Iter(train) [ 29650/160000] lr: 8.3325e-03 eta: 19:56:45 time: 0.5536 data_time: 0.0060 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0090 decode.acc_seg: 99.7258 aux.loss_ce: 0.0082 aux.acc_seg: 99.3110 +04/18 06:55:23 - mmengine - INFO - Iter(train) [ 29700/160000] lr: 8.3296e-03 eta: 19:56:18 time: 0.5535 data_time: 0.0059 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0100 decode.acc_seg: 99.3491 aux.loss_ce: 0.0084 aux.acc_seg: 98.8539 +04/18 06:55:50 - mmengine - INFO - Iter(train) [ 29750/160000] lr: 8.3268e-03 eta: 19:55:51 time: 0.5531 data_time: 0.0061 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0089 decode.acc_seg: 99.5722 aux.loss_ce: 0.0078 aux.acc_seg: 99.1220 +04/18 06:56:18 - mmengine - INFO - Iter(train) [ 29800/160000] lr: 8.3239e-03 eta: 19:55:24 time: 0.5539 data_time: 0.0066 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0103 decode.acc_seg: 99.6004 aux.loss_ce: 0.0080 aux.acc_seg: 99.2584 +04/18 06:56:46 - mmengine - INFO - Iter(train) [ 29850/160000] lr: 8.3211e-03 eta: 19:54:57 time: 0.5534 data_time: 0.0060 memory: 7635 loss: 0.0197 decode.loss_ce: 0.0108 decode.acc_seg: 99.6187 aux.loss_ce: 0.0089 aux.acc_seg: 99.1012 +04/18 06:57:13 - mmengine - INFO - Iter(train) [ 29900/160000] lr: 8.3182e-03 eta: 19:54:30 time: 0.5540 data_time: 0.0059 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0091 decode.acc_seg: 99.6822 aux.loss_ce: 0.0084 aux.acc_seg: 99.1999 +04/18 06:57:41 - mmengine - INFO - Iter(train) [ 29950/160000] lr: 8.3154e-03 eta: 19:54:03 time: 0.5533 data_time: 0.0064 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0084 decode.acc_seg: 99.6488 aux.loss_ce: 0.0076 aux.acc_seg: 99.1222 +04/18 06:58:09 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 06:58:09 - mmengine - INFO - Iter(train) [ 30000/160000] lr: 8.3126e-03 eta: 19:53:37 time: 0.5532 data_time: 0.0060 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0098 decode.acc_seg: 99.6754 aux.loss_ce: 0.0083 aux.acc_seg: 99.3052 +04/18 06:58:09 - mmengine - INFO - Saving checkpoint at 30000 iterations +04/18 06:58:13 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:06 time: 0.0462 data_time: 0.0014 memory: 1657 +04/18 06:58:15 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:04 time: 0.0475 data_time: 0.0014 memory: 1657 +04/18 06:58:18 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:02 time: 0.0463 data_time: 0.0014 memory: 1657 +04/18 06:58:20 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.0457 data_time: 0.0013 memory: 1657 +04/18 06:58:20 - mmengine - INFO - per class results: +04/18 06:58:20 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 99.06 | 99.47 | 99.53 | 99.58 | 99.47 | +| contrast | 79.8 | 89.93 | 88.77 | 87.63 | 89.93 | ++------------+-------+-------+--------+-----------+--------+ +04/18 06:58:20 - mmengine - INFO - Iter(val) [200/200] aAcc: 99.0900 mIoU: 89.4300 mAcc: 94.7000 mFscore: 94.1500 mPrecision: 93.6000 mRecall: 94.7000 data_time: 0.0015 time: 0.0464 +04/18 06:58:48 - mmengine - INFO - Iter(train) [ 30050/160000] lr: 8.3097e-03 eta: 19:53:11 time: 0.5539 data_time: 0.0065 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0090 decode.acc_seg: 99.6932 aux.loss_ce: 0.0085 aux.acc_seg: 99.2504 +04/18 06:59:16 - mmengine - INFO - Iter(train) [ 30100/160000] lr: 8.3069e-03 eta: 19:52:44 time: 0.5542 data_time: 0.0066 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0092 decode.acc_seg: 99.6229 aux.loss_ce: 0.0082 aux.acc_seg: 99.0599 +04/18 06:59:43 - mmengine - INFO - Iter(train) [ 30150/160000] lr: 8.3040e-03 eta: 19:52:17 time: 0.5537 data_time: 0.0066 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0097 decode.acc_seg: 99.7074 aux.loss_ce: 0.0081 aux.acc_seg: 99.2696 +04/18 07:00:11 - mmengine - INFO - Iter(train) [ 30200/160000] lr: 8.3012e-03 eta: 19:51:50 time: 0.5526 data_time: 0.0061 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0101 decode.acc_seg: 99.6636 aux.loss_ce: 0.0083 aux.acc_seg: 99.2550 +04/18 07:00:39 - mmengine - INFO - Iter(train) [ 30250/160000] lr: 8.2983e-03 eta: 19:51:24 time: 0.5532 data_time: 0.0063 memory: 7635 loss: 0.0181 decode.loss_ce: 0.0098 decode.acc_seg: 99.6634 aux.loss_ce: 0.0083 aux.acc_seg: 99.2562 +04/18 07:01:06 - mmengine - INFO - Iter(train) [ 30300/160000] lr: 8.2955e-03 eta: 19:50:57 time: 0.5539 data_time: 0.0061 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0092 decode.acc_seg: 99.6403 aux.loss_ce: 0.0079 aux.acc_seg: 99.2498 +04/18 07:01:34 - mmengine - INFO - Iter(train) [ 30350/160000] lr: 8.2927e-03 eta: 19:50:30 time: 0.5537 data_time: 0.0066 memory: 7635 loss: 0.0181 decode.loss_ce: 0.0094 decode.acc_seg: 99.6887 aux.loss_ce: 0.0086 aux.acc_seg: 99.3160 +04/18 07:02:02 - mmengine - INFO - Iter(train) [ 30400/160000] lr: 8.2898e-03 eta: 19:50:03 time: 0.5544 data_time: 0.0063 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0080 decode.acc_seg: 99.6344 aux.loss_ce: 0.0077 aux.acc_seg: 99.1701 +04/18 07:02:29 - mmengine - INFO - Iter(train) [ 30450/160000] lr: 8.2870e-03 eta: 19:49:36 time: 0.5528 data_time: 0.0057 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0099 decode.acc_seg: 99.6485 aux.loss_ce: 0.0084 aux.acc_seg: 99.3006 +04/18 07:02:57 - mmengine - INFO - Iter(train) [ 30500/160000] lr: 8.2841e-03 eta: 19:49:10 time: 0.5638 data_time: 0.0068 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0094 decode.acc_seg: 99.6327 aux.loss_ce: 0.0081 aux.acc_seg: 99.2662 +04/18 07:03:25 - mmengine - INFO - Iter(train) [ 30550/160000] lr: 8.2813e-03 eta: 19:48:43 time: 0.5538 data_time: 0.0063 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0099 decode.acc_seg: 99.6390 aux.loss_ce: 0.0081 aux.acc_seg: 99.1124 +04/18 07:03:53 - mmengine - INFO - Iter(train) [ 30600/160000] lr: 8.2784e-03 eta: 19:48:16 time: 0.5540 data_time: 0.0067 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0093 decode.acc_seg: 99.6712 aux.loss_ce: 0.0087 aux.acc_seg: 99.2329 +04/18 07:04:20 - mmengine - INFO - Iter(train) [ 30650/160000] lr: 8.2756e-03 eta: 19:47:49 time: 0.5539 data_time: 0.0065 memory: 7635 loss: 0.0198 decode.loss_ce: 0.0106 decode.acc_seg: 99.4262 aux.loss_ce: 0.0093 aux.acc_seg: 99.0763 +04/18 07:04:48 - mmengine - INFO - Iter(train) [ 30700/160000] lr: 8.2728e-03 eta: 19:47:22 time: 0.5541 data_time: 0.0067 memory: 7635 loss: 0.0197 decode.loss_ce: 0.0103 decode.acc_seg: 99.4432 aux.loss_ce: 0.0094 aux.acc_seg: 98.5764 +04/18 07:05:16 - mmengine - INFO - Iter(train) [ 30750/160000] lr: 8.2699e-03 eta: 19:46:55 time: 0.5530 data_time: 0.0077 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0096 decode.acc_seg: 99.5905 aux.loss_ce: 0.0081 aux.acc_seg: 99.1282 +04/18 07:05:43 - mmengine - INFO - Iter(train) [ 30800/160000] lr: 8.2671e-03 eta: 19:46:28 time: 0.5539 data_time: 0.0061 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0088 decode.acc_seg: 99.6326 aux.loss_ce: 0.0080 aux.acc_seg: 99.1488 +04/18 07:06:11 - mmengine - INFO - Iter(train) [ 30850/160000] lr: 8.2642e-03 eta: 19:46:01 time: 0.5541 data_time: 0.0065 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0082 decode.acc_seg: 99.5945 aux.loss_ce: 0.0074 aux.acc_seg: 99.1685 +04/18 07:06:39 - mmengine - INFO - Iter(train) [ 30900/160000] lr: 8.2614e-03 eta: 19:45:34 time: 0.5548 data_time: 0.0064 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0095 decode.acc_seg: 99.7173 aux.loss_ce: 0.0081 aux.acc_seg: 99.3325 +04/18 07:07:07 - mmengine - INFO - Iter(train) [ 30950/160000] lr: 8.2585e-03 eta: 19:45:07 time: 0.5541 data_time: 0.0059 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0094 decode.acc_seg: 99.6530 aux.loss_ce: 0.0080 aux.acc_seg: 99.1598 +04/18 07:07:34 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 07:07:34 - mmengine - INFO - Iter(train) [ 31000/160000] lr: 8.2557e-03 eta: 19:44:40 time: 0.5538 data_time: 0.0060 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0090 decode.acc_seg: 99.6911 aux.loss_ce: 0.0084 aux.acc_seg: 99.1821 +04/18 07:08:02 - mmengine - INFO - Iter(train) [ 31050/160000] lr: 8.2528e-03 eta: 19:44:14 time: 0.5544 data_time: 0.0065 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0102 decode.acc_seg: 99.5356 aux.loss_ce: 0.0087 aux.acc_seg: 99.1998 +04/18 07:08:30 - mmengine - INFO - Iter(train) [ 31100/160000] lr: 8.2500e-03 eta: 19:43:47 time: 0.5540 data_time: 0.0065 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0089 decode.acc_seg: 99.5303 aux.loss_ce: 0.0076 aux.acc_seg: 99.0437 +04/18 07:08:58 - mmengine - INFO - Iter(train) [ 31150/160000] lr: 8.2471e-03 eta: 19:43:20 time: 0.5542 data_time: 0.0069 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0090 decode.acc_seg: 99.7702 aux.loss_ce: 0.0083 aux.acc_seg: 99.3576 +04/18 07:09:25 - mmengine - INFO - Iter(train) [ 31200/160000] lr: 8.2443e-03 eta: 19:42:53 time: 0.5550 data_time: 0.0062 memory: 7635 loss: 0.0181 decode.loss_ce: 0.0094 decode.acc_seg: 99.6806 aux.loss_ce: 0.0087 aux.acc_seg: 99.3296 +04/18 07:09:53 - mmengine - INFO - Iter(train) [ 31250/160000] lr: 8.2415e-03 eta: 19:42:26 time: 0.5541 data_time: 0.0063 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0095 decode.acc_seg: 99.5939 aux.loss_ce: 0.0082 aux.acc_seg: 99.2475 +04/18 07:10:21 - mmengine - INFO - Iter(train) [ 31300/160000] lr: 8.2386e-03 eta: 19:42:00 time: 0.5541 data_time: 0.0060 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0090 decode.acc_seg: 99.6605 aux.loss_ce: 0.0077 aux.acc_seg: 99.1757 +04/18 07:10:48 - mmengine - INFO - Iter(train) [ 31350/160000] lr: 8.2358e-03 eta: 19:41:33 time: 0.5548 data_time: 0.0064 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0088 decode.acc_seg: 99.6571 aux.loss_ce: 0.0082 aux.acc_seg: 99.0974 +04/18 07:11:16 - mmengine - INFO - Iter(train) [ 31400/160000] lr: 8.2329e-03 eta: 19:41:06 time: 0.5538 data_time: 0.0069 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0090 decode.acc_seg: 99.6479 aux.loss_ce: 0.0081 aux.acc_seg: 98.9635 +04/18 07:11:44 - mmengine - INFO - Iter(train) [ 31450/160000] lr: 8.2301e-03 eta: 19:40:39 time: 0.5542 data_time: 0.0063 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0089 decode.acc_seg: 99.6761 aux.loss_ce: 0.0079 aux.acc_seg: 99.3599 +04/18 07:12:12 - mmengine - INFO - Iter(train) [ 31500/160000] lr: 8.2272e-03 eta: 19:40:12 time: 0.5548 data_time: 0.0062 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0086 decode.acc_seg: 99.7297 aux.loss_ce: 0.0081 aux.acc_seg: 99.3115 +04/18 07:12:39 - mmengine - INFO - Iter(train) [ 31550/160000] lr: 8.2244e-03 eta: 19:39:45 time: 0.5535 data_time: 0.0068 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0090 decode.acc_seg: 99.6945 aux.loss_ce: 0.0082 aux.acc_seg: 99.3441 +04/18 07:13:07 - mmengine - INFO - Iter(train) [ 31600/160000] lr: 8.2215e-03 eta: 19:39:19 time: 0.5717 data_time: 0.0058 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0090 decode.acc_seg: 99.5731 aux.loss_ce: 0.0080 aux.acc_seg: 99.0817 +04/18 07:13:35 - mmengine - INFO - Iter(train) [ 31650/160000] lr: 8.2187e-03 eta: 19:38:52 time: 0.5542 data_time: 0.0068 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0095 decode.acc_seg: 99.6840 aux.loss_ce: 0.0085 aux.acc_seg: 99.1050 +04/18 07:14:03 - mmengine - INFO - Iter(train) [ 31700/160000] lr: 8.2158e-03 eta: 19:38:25 time: 0.5544 data_time: 0.0059 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0083 decode.acc_seg: 99.6806 aux.loss_ce: 0.0077 aux.acc_seg: 99.2702 +04/18 07:14:30 - mmengine - INFO - Iter(train) [ 31750/160000] lr: 8.2130e-03 eta: 19:37:58 time: 0.5534 data_time: 0.0061 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0084 decode.acc_seg: 99.6800 aux.loss_ce: 0.0078 aux.acc_seg: 99.0911 +04/18 07:14:58 - mmengine - INFO - Iter(train) [ 31800/160000] lr: 8.2101e-03 eta: 19:37:31 time: 0.5537 data_time: 0.0065 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0080 decode.acc_seg: 99.6931 aux.loss_ce: 0.0073 aux.acc_seg: 99.2317 +04/18 07:15:26 - mmengine - INFO - Iter(train) [ 31850/160000] lr: 8.2073e-03 eta: 19:37:04 time: 0.5537 data_time: 0.0061 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0108 decode.acc_seg: 99.6500 aux.loss_ce: 0.0087 aux.acc_seg: 99.3303 +04/18 07:15:53 - mmengine - INFO - Iter(train) [ 31900/160000] lr: 8.2045e-03 eta: 19:36:37 time: 0.5531 data_time: 0.0062 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0099 decode.acc_seg: 99.6323 aux.loss_ce: 0.0084 aux.acc_seg: 99.2227 +04/18 07:16:21 - mmengine - INFO - Iter(train) [ 31950/160000] lr: 8.2016e-03 eta: 19:36:10 time: 0.5551 data_time: 0.0063 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0088 decode.acc_seg: 99.6806 aux.loss_ce: 0.0079 aux.acc_seg: 99.1612 +04/18 07:16:49 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 07:16:49 - mmengine - INFO - Iter(train) [ 32000/160000] lr: 8.1988e-03 eta: 19:35:43 time: 0.5542 data_time: 0.0066 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0080 decode.acc_seg: 99.6737 aux.loss_ce: 0.0075 aux.acc_seg: 99.1329 +04/18 07:17:16 - mmengine - INFO - Iter(train) [ 32050/160000] lr: 8.1959e-03 eta: 19:35:16 time: 0.5535 data_time: 0.0064 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0083 decode.acc_seg: 99.6179 aux.loss_ce: 0.0078 aux.acc_seg: 99.2297 +04/18 07:17:44 - mmengine - INFO - Iter(train) [ 32100/160000] lr: 8.1931e-03 eta: 19:34:49 time: 0.5544 data_time: 0.0057 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0085 decode.acc_seg: 99.6239 aux.loss_ce: 0.0071 aux.acc_seg: 99.1700 +04/18 07:18:12 - mmengine - INFO - Iter(train) [ 32150/160000] lr: 8.1902e-03 eta: 19:34:22 time: 0.5551 data_time: 0.0058 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0103 decode.acc_seg: 99.5864 aux.loss_ce: 0.0083 aux.acc_seg: 99.2222 +04/18 07:18:40 - mmengine - INFO - Iter(train) [ 32200/160000] lr: 8.1874e-03 eta: 19:33:55 time: 0.5546 data_time: 0.0065 memory: 7635 loss: 0.0199 decode.loss_ce: 0.0108 decode.acc_seg: 99.4581 aux.loss_ce: 0.0091 aux.acc_seg: 98.7620 +04/18 07:19:07 - mmengine - INFO - Iter(train) [ 32250/160000] lr: 8.1845e-03 eta: 19:33:28 time: 0.5551 data_time: 0.0069 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0093 decode.acc_seg: 99.5605 aux.loss_ce: 0.0079 aux.acc_seg: 99.0670 +04/18 07:19:35 - mmengine - INFO - Iter(train) [ 32300/160000] lr: 8.1817e-03 eta: 19:33:02 time: 0.5533 data_time: 0.0064 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0103 decode.acc_seg: 99.6199 aux.loss_ce: 0.0088 aux.acc_seg: 99.1172 +04/18 07:20:03 - mmengine - INFO - Iter(train) [ 32350/160000] lr: 8.1788e-03 eta: 19:32:35 time: 0.5535 data_time: 0.0066 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0089 decode.acc_seg: 99.6283 aux.loss_ce: 0.0080 aux.acc_seg: 99.1041 +04/18 07:20:31 - mmengine - INFO - Iter(train) [ 32400/160000] lr: 8.1760e-03 eta: 19:32:08 time: 0.5545 data_time: 0.0071 memory: 7635 loss: 0.0213 decode.loss_ce: 0.0117 decode.acc_seg: 99.4681 aux.loss_ce: 0.0096 aux.acc_seg: 98.9940 +04/18 07:20:58 - mmengine - INFO - Iter(train) [ 32450/160000] lr: 8.1731e-03 eta: 19:31:41 time: 0.5551 data_time: 0.0070 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0086 decode.acc_seg: 99.6467 aux.loss_ce: 0.0077 aux.acc_seg: 99.0788 +04/18 07:21:26 - mmengine - INFO - Iter(train) [ 32500/160000] lr: 8.1703e-03 eta: 19:31:14 time: 0.5550 data_time: 0.0063 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0120 decode.acc_seg: 99.3916 aux.loss_ce: 0.0087 aux.acc_seg: 99.2522 +04/18 07:21:54 - mmengine - INFO - Iter(train) [ 32550/160000] lr: 8.1674e-03 eta: 19:30:47 time: 0.5538 data_time: 0.0069 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0095 decode.acc_seg: 99.6855 aux.loss_ce: 0.0083 aux.acc_seg: 99.2477 +04/18 07:22:21 - mmengine - INFO - Iter(train) [ 32600/160000] lr: 8.1646e-03 eta: 19:30:20 time: 0.5532 data_time: 0.0060 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0085 decode.acc_seg: 99.6750 aux.loss_ce: 0.0075 aux.acc_seg: 99.2866 +04/18 07:22:49 - mmengine - INFO - Iter(train) [ 32650/160000] lr: 8.1617e-03 eta: 19:29:53 time: 0.5539 data_time: 0.0070 memory: 7635 loss: 0.0186 decode.loss_ce: 0.0100 decode.acc_seg: 99.4937 aux.loss_ce: 0.0086 aux.acc_seg: 98.9815 +04/18 07:23:17 - mmengine - INFO - Iter(train) [ 32700/160000] lr: 8.1589e-03 eta: 19:29:27 time: 0.5529 data_time: 0.0061 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0091 decode.acc_seg: 99.6044 aux.loss_ce: 0.0082 aux.acc_seg: 99.1128 +04/18 07:23:45 - mmengine - INFO - Iter(train) [ 32750/160000] lr: 8.1560e-03 eta: 19:29:00 time: 0.5539 data_time: 0.0061 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0089 decode.acc_seg: 99.6833 aux.loss_ce: 0.0080 aux.acc_seg: 99.1595 +04/18 07:24:13 - mmengine - INFO - Iter(train) [ 32800/160000] lr: 8.1532e-03 eta: 19:28:33 time: 0.5539 data_time: 0.0059 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0085 decode.acc_seg: 99.6199 aux.loss_ce: 0.0076 aux.acc_seg: 99.2431 +04/18 07:24:40 - mmengine - INFO - Iter(train) [ 32850/160000] lr: 8.1503e-03 eta: 19:28:06 time: 0.5543 data_time: 0.0065 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0089 decode.acc_seg: 99.6441 aux.loss_ce: 0.0082 aux.acc_seg: 99.1406 +04/18 07:25:08 - mmengine - INFO - Iter(train) [ 32900/160000] lr: 8.1475e-03 eta: 19:27:39 time: 0.5537 data_time: 0.0060 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0099 decode.acc_seg: 99.4864 aux.loss_ce: 0.0086 aux.acc_seg: 99.0148 +04/18 07:25:36 - mmengine - INFO - Iter(train) [ 32950/160000] lr: 8.1446e-03 eta: 19:27:12 time: 0.5534 data_time: 0.0058 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0082 decode.acc_seg: 99.5735 aux.loss_ce: 0.0071 aux.acc_seg: 99.2791 +04/18 07:26:03 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 07:26:03 - mmengine - INFO - Iter(train) [ 33000/160000] lr: 8.1418e-03 eta: 19:26:45 time: 0.5538 data_time: 0.0066 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0082 decode.acc_seg: 99.6367 aux.loss_ce: 0.0076 aux.acc_seg: 99.0615 +04/18 07:26:31 - mmengine - INFO - Iter(train) [ 33050/160000] lr: 8.1389e-03 eta: 19:26:17 time: 0.5534 data_time: 0.0056 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0083 decode.acc_seg: 99.4673 aux.loss_ce: 0.0074 aux.acc_seg: 99.0739 +04/18 07:26:59 - mmengine - INFO - Iter(train) [ 33100/160000] lr: 8.1361e-03 eta: 19:25:50 time: 0.5523 data_time: 0.0065 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0089 decode.acc_seg: 99.6405 aux.loss_ce: 0.0085 aux.acc_seg: 99.1338 +04/18 07:27:26 - mmengine - INFO - Iter(train) [ 33150/160000] lr: 8.1332e-03 eta: 19:25:23 time: 0.5527 data_time: 0.0064 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0104 decode.acc_seg: 99.5017 aux.loss_ce: 0.0087 aux.acc_seg: 99.0799 +04/18 07:27:54 - mmengine - INFO - Iter(train) [ 33200/160000] lr: 8.1304e-03 eta: 19:24:56 time: 0.5534 data_time: 0.0071 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0092 decode.acc_seg: 99.5747 aux.loss_ce: 0.0081 aux.acc_seg: 99.1100 +04/18 07:28:22 - mmengine - INFO - Iter(train) [ 33250/160000] lr: 8.1275e-03 eta: 19:24:29 time: 0.5536 data_time: 0.0063 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0085 decode.acc_seg: 99.7563 aux.loss_ce: 0.0076 aux.acc_seg: 99.3136 +04/18 07:28:49 - mmengine - INFO - Iter(train) [ 33300/160000] lr: 8.1247e-03 eta: 19:24:02 time: 0.5539 data_time: 0.0065 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0083 decode.acc_seg: 99.7503 aux.loss_ce: 0.0079 aux.acc_seg: 99.3759 +04/18 07:29:17 - mmengine - INFO - Iter(train) [ 33350/160000] lr: 8.1218e-03 eta: 19:23:35 time: 0.5532 data_time: 0.0061 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0087 decode.acc_seg: 99.6707 aux.loss_ce: 0.0079 aux.acc_seg: 99.0377 +04/18 07:29:45 - mmengine - INFO - Iter(train) [ 33400/160000] lr: 8.1190e-03 eta: 19:23:08 time: 0.5534 data_time: 0.0069 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0077 decode.acc_seg: 99.5971 aux.loss_ce: 0.0075 aux.acc_seg: 98.8136 +04/18 07:30:12 - mmengine - INFO - Iter(train) [ 33450/160000] lr: 8.1161e-03 eta: 19:22:41 time: 0.5537 data_time: 0.0064 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0092 decode.acc_seg: 99.6467 aux.loss_ce: 0.0084 aux.acc_seg: 99.2325 +04/18 07:30:40 - mmengine - INFO - Iter(train) [ 33500/160000] lr: 8.1133e-03 eta: 19:22:14 time: 0.5527 data_time: 0.0063 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0085 decode.acc_seg: 99.6486 aux.loss_ce: 0.0077 aux.acc_seg: 99.1936 +04/18 07:31:08 - mmengine - INFO - Iter(train) [ 33550/160000] lr: 8.1104e-03 eta: 19:21:46 time: 0.5521 data_time: 0.0064 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0087 decode.acc_seg: 99.6886 aux.loss_ce: 0.0075 aux.acc_seg: 99.3878 +04/18 07:31:35 - mmengine - INFO - Iter(train) [ 33600/160000] lr: 8.1076e-03 eta: 19:21:19 time: 0.5530 data_time: 0.0066 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0076 decode.acc_seg: 99.6814 aux.loss_ce: 0.0073 aux.acc_seg: 99.2887 +04/18 07:32:03 - mmengine - INFO - Iter(train) [ 33650/160000] lr: 8.1047e-03 eta: 19:20:52 time: 0.5537 data_time: 0.0065 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0087 decode.acc_seg: 99.6370 aux.loss_ce: 0.0081 aux.acc_seg: 99.2607 +04/18 07:32:31 - mmengine - INFO - Iter(train) [ 33700/160000] lr: 8.1019e-03 eta: 19:20:25 time: 0.5624 data_time: 0.0066 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0096 decode.acc_seg: 99.6526 aux.loss_ce: 0.0081 aux.acc_seg: 99.2915 +04/18 07:32:59 - mmengine - INFO - Iter(train) [ 33750/160000] lr: 8.0990e-03 eta: 19:19:58 time: 0.5529 data_time: 0.0065 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0102 decode.acc_seg: 99.6324 aux.loss_ce: 0.0085 aux.acc_seg: 99.3615 +04/18 07:33:26 - mmengine - INFO - Iter(train) [ 33800/160000] lr: 8.0962e-03 eta: 19:19:31 time: 0.5521 data_time: 0.0062 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0093 decode.acc_seg: 99.6343 aux.loss_ce: 0.0080 aux.acc_seg: 99.2024 +04/18 07:33:54 - mmengine - INFO - Iter(train) [ 33850/160000] lr: 8.0933e-03 eta: 19:19:04 time: 0.5543 data_time: 0.0065 memory: 7635 loss: 0.0186 decode.loss_ce: 0.0102 decode.acc_seg: 99.7551 aux.loss_ce: 0.0084 aux.acc_seg: 99.4288 +04/18 07:34:22 - mmengine - INFO - Iter(train) [ 33900/160000] lr: 8.0905e-03 eta: 19:18:37 time: 0.5521 data_time: 0.0061 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0084 decode.acc_seg: 99.5685 aux.loss_ce: 0.0079 aux.acc_seg: 99.0632 +04/18 07:34:49 - mmengine - INFO - Iter(train) [ 33950/160000] lr: 8.0876e-03 eta: 19:18:09 time: 0.5523 data_time: 0.0062 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0091 decode.acc_seg: 99.6301 aux.loss_ce: 0.0083 aux.acc_seg: 99.0987 +04/18 07:35:17 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 07:35:17 - mmengine - INFO - Iter(train) [ 34000/160000] lr: 8.0848e-03 eta: 19:17:42 time: 0.5528 data_time: 0.0066 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0098 decode.acc_seg: 99.6192 aux.loss_ce: 0.0085 aux.acc_seg: 98.8934 +04/18 07:35:45 - mmengine - INFO - Iter(train) [ 34050/160000] lr: 8.0819e-03 eta: 19:17:15 time: 0.5515 data_time: 0.0064 memory: 7635 loss: 0.0217 decode.loss_ce: 0.0127 decode.acc_seg: 99.1635 aux.loss_ce: 0.0090 aux.acc_seg: 98.8476 +04/18 07:36:12 - mmengine - INFO - Iter(train) [ 34100/160000] lr: 8.0791e-03 eta: 19:16:47 time: 0.5527 data_time: 0.0062 memory: 7635 loss: 0.0216 decode.loss_ce: 0.0124 decode.acc_seg: 99.4928 aux.loss_ce: 0.0092 aux.acc_seg: 99.1305 +04/18 07:36:40 - mmengine - INFO - Iter(train) [ 34150/160000] lr: 8.0762e-03 eta: 19:16:20 time: 0.5552 data_time: 0.0062 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0095 decode.acc_seg: 99.6648 aux.loss_ce: 0.0082 aux.acc_seg: 99.2626 +04/18 07:37:08 - mmengine - INFO - Iter(train) [ 34200/160000] lr: 8.0734e-03 eta: 19:15:53 time: 0.5536 data_time: 0.0062 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0093 decode.acc_seg: 99.6975 aux.loss_ce: 0.0083 aux.acc_seg: 99.3648 +04/18 07:37:35 - mmengine - INFO - Iter(train) [ 34250/160000] lr: 8.0705e-03 eta: 19:15:26 time: 0.5548 data_time: 0.0064 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0096 decode.acc_seg: 99.6301 aux.loss_ce: 0.0086 aux.acc_seg: 99.1169 +04/18 07:38:03 - mmengine - INFO - Iter(train) [ 34300/160000] lr: 8.0677e-03 eta: 19:14:59 time: 0.5546 data_time: 0.0063 memory: 7635 loss: 0.0201 decode.loss_ce: 0.0108 decode.acc_seg: 99.7034 aux.loss_ce: 0.0093 aux.acc_seg: 99.1738 +04/18 07:38:31 - mmengine - INFO - Iter(train) [ 34350/160000] lr: 8.0648e-03 eta: 19:14:33 time: 0.5553 data_time: 0.0064 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0092 decode.acc_seg: 99.6557 aux.loss_ce: 0.0080 aux.acc_seg: 99.2526 +04/18 07:38:59 - mmengine - INFO - Iter(train) [ 34400/160000] lr: 8.0620e-03 eta: 19:14:06 time: 0.5548 data_time: 0.0067 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0100 decode.acc_seg: 99.4610 aux.loss_ce: 0.0086 aux.acc_seg: 98.9637 +04/18 07:39:26 - mmengine - INFO - Iter(train) [ 34450/160000] lr: 8.0591e-03 eta: 19:13:39 time: 0.5554 data_time: 0.0057 memory: 7635 loss: 0.0186 decode.loss_ce: 0.0098 decode.acc_seg: 99.6735 aux.loss_ce: 0.0088 aux.acc_seg: 99.1503 +04/18 07:39:54 - mmengine - INFO - Iter(train) [ 34500/160000] lr: 8.0563e-03 eta: 19:13:12 time: 0.5548 data_time: 0.0063 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0098 decode.acc_seg: 99.5838 aux.loss_ce: 0.0085 aux.acc_seg: 99.2102 +04/18 07:40:22 - mmengine - INFO - Iter(train) [ 34550/160000] lr: 8.0534e-03 eta: 19:12:45 time: 0.5552 data_time: 0.0060 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0081 decode.acc_seg: 99.5566 aux.loss_ce: 0.0075 aux.acc_seg: 99.0582 +04/18 07:40:50 - mmengine - INFO - Iter(train) [ 34600/160000] lr: 8.0506e-03 eta: 19:12:19 time: 0.5556 data_time: 0.0070 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0084 decode.acc_seg: 99.6641 aux.loss_ce: 0.0078 aux.acc_seg: 99.1263 +04/18 07:41:18 - mmengine - INFO - Iter(train) [ 34650/160000] lr: 8.0477e-03 eta: 19:11:52 time: 0.5545 data_time: 0.0063 memory: 7635 loss: 0.0186 decode.loss_ce: 0.0100 decode.acc_seg: 99.5957 aux.loss_ce: 0.0086 aux.acc_seg: 99.1197 +04/18 07:41:45 - mmengine - INFO - Iter(train) [ 34700/160000] lr: 8.0448e-03 eta: 19:11:25 time: 0.5540 data_time: 0.0060 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0082 decode.acc_seg: 99.5585 aux.loss_ce: 0.0078 aux.acc_seg: 99.0187 +04/18 07:42:13 - mmengine - INFO - Iter(train) [ 34750/160000] lr: 8.0420e-03 eta: 19:10:58 time: 0.5547 data_time: 0.0058 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0085 decode.acc_seg: 99.6587 aux.loss_ce: 0.0081 aux.acc_seg: 99.2609 +04/18 07:42:41 - mmengine - INFO - Iter(train) [ 34800/160000] lr: 8.0391e-03 eta: 19:10:32 time: 0.5561 data_time: 0.0071 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0086 decode.acc_seg: 99.6828 aux.loss_ce: 0.0076 aux.acc_seg: 99.2761 +04/18 07:43:09 - mmengine - INFO - Iter(train) [ 34850/160000] lr: 8.0363e-03 eta: 19:10:05 time: 0.5544 data_time: 0.0061 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0100 decode.acc_seg: 99.6472 aux.loss_ce: 0.0093 aux.acc_seg: 99.2065 +04/18 07:43:37 - mmengine - INFO - Iter(train) [ 34900/160000] lr: 8.0334e-03 eta: 19:09:38 time: 0.5544 data_time: 0.0059 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0086 decode.acc_seg: 99.7450 aux.loss_ce: 0.0079 aux.acc_seg: 99.3112 +04/18 07:44:04 - mmengine - INFO - Iter(train) [ 34950/160000] lr: 8.0306e-03 eta: 19:09:11 time: 0.5557 data_time: 0.0063 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0076 decode.acc_seg: 99.6366 aux.loss_ce: 0.0074 aux.acc_seg: 99.2044 +04/18 07:44:32 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 07:44:32 - mmengine - INFO - Iter(train) [ 35000/160000] lr: 8.0277e-03 eta: 19:08:45 time: 0.5542 data_time: 0.0061 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0088 decode.acc_seg: 99.6536 aux.loss_ce: 0.0081 aux.acc_seg: 98.9813 +04/18 07:45:00 - mmengine - INFO - Iter(train) [ 35050/160000] lr: 8.0249e-03 eta: 19:08:18 time: 0.5559 data_time: 0.0065 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0081 decode.acc_seg: 99.5843 aux.loss_ce: 0.0075 aux.acc_seg: 99.2393 +04/18 07:45:28 - mmengine - INFO - Iter(train) [ 35100/160000] lr: 8.0220e-03 eta: 19:07:51 time: 0.5569 data_time: 0.0060 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0086 decode.acc_seg: 99.6353 aux.loss_ce: 0.0078 aux.acc_seg: 99.0710 +04/18 07:45:56 - mmengine - INFO - Iter(train) [ 35150/160000] lr: 8.0192e-03 eta: 19:07:24 time: 0.5550 data_time: 0.0060 memory: 7635 loss: 0.0190 decode.loss_ce: 0.0102 decode.acc_seg: 99.5936 aux.loss_ce: 0.0088 aux.acc_seg: 98.9582 +04/18 07:46:23 - mmengine - INFO - Iter(train) [ 35200/160000] lr: 8.0163e-03 eta: 19:06:57 time: 0.5543 data_time: 0.0064 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0094 decode.acc_seg: 99.5790 aux.loss_ce: 0.0083 aux.acc_seg: 99.0090 +04/18 07:46:51 - mmengine - INFO - Iter(train) [ 35250/160000] lr: 8.0135e-03 eta: 19:06:30 time: 0.5546 data_time: 0.0064 memory: 7635 loss: 0.0197 decode.loss_ce: 0.0109 decode.acc_seg: 99.6765 aux.loss_ce: 0.0088 aux.acc_seg: 99.2990 +04/18 07:47:19 - mmengine - INFO - Iter(train) [ 35300/160000] lr: 8.0106e-03 eta: 19:06:04 time: 0.5561 data_time: 0.0062 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0089 decode.acc_seg: 99.6948 aux.loss_ce: 0.0079 aux.acc_seg: 99.3765 +04/18 07:47:47 - mmengine - INFO - Iter(train) [ 35350/160000] lr: 8.0077e-03 eta: 19:05:37 time: 0.5554 data_time: 0.0062 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0099 decode.acc_seg: 99.5642 aux.loss_ce: 0.0080 aux.acc_seg: 99.1477 +04/18 07:48:14 - mmengine - INFO - Iter(train) [ 35400/160000] lr: 8.0049e-03 eta: 19:05:10 time: 0.5551 data_time: 0.0067 memory: 7635 loss: 0.0181 decode.loss_ce: 0.0098 decode.acc_seg: 99.5631 aux.loss_ce: 0.0083 aux.acc_seg: 98.8854 +04/18 07:48:42 - mmengine - INFO - Iter(train) [ 35450/160000] lr: 8.0020e-03 eta: 19:04:43 time: 0.5545 data_time: 0.0059 memory: 7635 loss: 0.0197 decode.loss_ce: 0.0108 decode.acc_seg: 99.5046 aux.loss_ce: 0.0089 aux.acc_seg: 98.9935 +04/18 07:49:10 - mmengine - INFO - Iter(train) [ 35500/160000] lr: 7.9992e-03 eta: 19:04:16 time: 0.5546 data_time: 0.0065 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0086 decode.acc_seg: 99.6613 aux.loss_ce: 0.0074 aux.acc_seg: 99.2374 +04/18 07:49:38 - mmengine - INFO - Iter(train) [ 35550/160000] lr: 7.9963e-03 eta: 19:03:49 time: 0.5559 data_time: 0.0073 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0096 decode.acc_seg: 99.7468 aux.loss_ce: 0.0082 aux.acc_seg: 99.4647 +04/18 07:50:06 - mmengine - INFO - Iter(train) [ 35600/160000] lr: 7.9935e-03 eta: 19:03:22 time: 0.5548 data_time: 0.0062 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0093 decode.acc_seg: 99.5835 aux.loss_ce: 0.0089 aux.acc_seg: 99.1254 +04/18 07:50:33 - mmengine - INFO - Iter(train) [ 35650/160000] lr: 7.9906e-03 eta: 19:02:55 time: 0.5532 data_time: 0.0058 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0087 decode.acc_seg: 99.6332 aux.loss_ce: 0.0079 aux.acc_seg: 99.2542 +04/18 07:51:01 - mmengine - INFO - Iter(train) [ 35700/160000] lr: 7.9878e-03 eta: 19:02:29 time: 0.5561 data_time: 0.0058 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0093 decode.acc_seg: 99.5657 aux.loss_ce: 0.0077 aux.acc_seg: 99.2030 +04/18 07:51:29 - mmengine - INFO - Iter(train) [ 35750/160000] lr: 7.9849e-03 eta: 19:02:02 time: 0.5547 data_time: 0.0067 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0099 decode.acc_seg: 99.7052 aux.loss_ce: 0.0087 aux.acc_seg: 99.2721 +04/18 07:51:57 - mmengine - INFO - Iter(train) [ 35800/160000] lr: 7.9820e-03 eta: 19:01:35 time: 0.5581 data_time: 0.0071 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0087 decode.acc_seg: 99.5608 aux.loss_ce: 0.0080 aux.acc_seg: 99.0271 +04/18 07:52:25 - mmengine - INFO - Iter(train) [ 35850/160000] lr: 7.9792e-03 eta: 19:01:08 time: 0.5644 data_time: 0.0064 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0095 decode.acc_seg: 99.5785 aux.loss_ce: 0.0086 aux.acc_seg: 98.9521 +04/18 07:52:52 - mmengine - INFO - Iter(train) [ 35900/160000] lr: 7.9763e-03 eta: 19:00:42 time: 0.5556 data_time: 0.0064 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0097 decode.acc_seg: 99.5304 aux.loss_ce: 0.0083 aux.acc_seg: 99.1608 +04/18 07:53:20 - mmengine - INFO - Iter(train) [ 35950/160000] lr: 7.9735e-03 eta: 19:00:15 time: 0.5558 data_time: 0.0069 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0095 decode.acc_seg: 99.6070 aux.loss_ce: 0.0082 aux.acc_seg: 99.2854 +04/18 07:53:48 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 07:53:48 - mmengine - INFO - Iter(train) [ 36000/160000] lr: 7.9706e-03 eta: 18:59:49 time: 0.5550 data_time: 0.0065 memory: 7635 loss: 0.0181 decode.loss_ce: 0.0098 decode.acc_seg: 99.7172 aux.loss_ce: 0.0082 aux.acc_seg: 99.4210 +04/18 07:54:16 - mmengine - INFO - Iter(train) [ 36050/160000] lr: 7.9678e-03 eta: 18:59:22 time: 0.5554 data_time: 0.0066 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0096 decode.acc_seg: 99.4981 aux.loss_ce: 0.0089 aux.acc_seg: 98.9285 +04/18 07:54:44 - mmengine - INFO - Iter(train) [ 36100/160000] lr: 7.9649e-03 eta: 18:58:55 time: 0.5553 data_time: 0.0062 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0088 decode.acc_seg: 99.6424 aux.loss_ce: 0.0079 aux.acc_seg: 99.1971 +04/18 07:55:12 - mmengine - INFO - Iter(train) [ 36150/160000] lr: 7.9621e-03 eta: 18:58:28 time: 0.5571 data_time: 0.0064 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0084 decode.acc_seg: 99.6611 aux.loss_ce: 0.0080 aux.acc_seg: 99.1581 +04/18 07:55:39 - mmengine - INFO - Iter(train) [ 36200/160000] lr: 7.9592e-03 eta: 18:58:01 time: 0.5560 data_time: 0.0066 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0086 decode.acc_seg: 99.6945 aux.loss_ce: 0.0079 aux.acc_seg: 99.3343 +04/18 07:56:07 - mmengine - INFO - Iter(train) [ 36250/160000] lr: 7.9563e-03 eta: 18:57:34 time: 0.5548 data_time: 0.0062 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0092 decode.acc_seg: 99.5932 aux.loss_ce: 0.0080 aux.acc_seg: 99.1081 +04/18 07:56:35 - mmengine - INFO - Iter(train) [ 36300/160000] lr: 7.9535e-03 eta: 18:57:08 time: 0.5556 data_time: 0.0068 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0091 decode.acc_seg: 99.7787 aux.loss_ce: 0.0080 aux.acc_seg: 99.4762 +04/18 07:57:03 - mmengine - INFO - Iter(train) [ 36350/160000] lr: 7.9506e-03 eta: 18:56:41 time: 0.5554 data_time: 0.0063 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0084 decode.acc_seg: 99.6775 aux.loss_ce: 0.0079 aux.acc_seg: 99.3327 +04/18 07:57:31 - mmengine - INFO - Iter(train) [ 36400/160000] lr: 7.9478e-03 eta: 18:56:14 time: 0.5567 data_time: 0.0060 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0089 decode.acc_seg: 99.5826 aux.loss_ce: 0.0078 aux.acc_seg: 99.0872 +04/18 07:57:58 - mmengine - INFO - Iter(train) [ 36450/160000] lr: 7.9449e-03 eta: 18:55:47 time: 0.5553 data_time: 0.0065 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0088 decode.acc_seg: 99.6811 aux.loss_ce: 0.0082 aux.acc_seg: 99.2513 +04/18 07:58:26 - mmengine - INFO - Iter(train) [ 36500/160000] lr: 7.9421e-03 eta: 18:55:20 time: 0.5542 data_time: 0.0063 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0087 decode.acc_seg: 99.7375 aux.loss_ce: 0.0079 aux.acc_seg: 99.3763 +04/18 07:58:54 - mmengine - INFO - Iter(train) [ 36550/160000] lr: 7.9392e-03 eta: 18:54:53 time: 0.5528 data_time: 0.0059 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0084 decode.acc_seg: 99.7747 aux.loss_ce: 0.0076 aux.acc_seg: 99.4019 +04/18 07:59:22 - mmengine - INFO - Iter(train) [ 36600/160000] lr: 7.9363e-03 eta: 18:54:26 time: 0.5534 data_time: 0.0068 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0087 decode.acc_seg: 99.6769 aux.loss_ce: 0.0080 aux.acc_seg: 99.1664 +04/18 07:59:49 - mmengine - INFO - Iter(train) [ 36650/160000] lr: 7.9335e-03 eta: 18:53:59 time: 0.5552 data_time: 0.0063 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0092 decode.acc_seg: 99.5338 aux.loss_ce: 0.0084 aux.acc_seg: 99.1665 +04/18 08:00:17 - mmengine - INFO - Iter(train) [ 36700/160000] lr: 7.9306e-03 eta: 18:53:32 time: 0.5555 data_time: 0.0063 memory: 7635 loss: 0.0205 decode.loss_ce: 0.0109 decode.acc_seg: 99.6617 aux.loss_ce: 0.0095 aux.acc_seg: 99.2718 +04/18 08:00:45 - mmengine - INFO - Iter(train) [ 36750/160000] lr: 7.9278e-03 eta: 18:53:05 time: 0.5557 data_time: 0.0060 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0101 decode.acc_seg: 99.6224 aux.loss_ce: 0.0086 aux.acc_seg: 99.1976 +04/18 08:01:13 - mmengine - INFO - Iter(train) [ 36800/160000] lr: 7.9249e-03 eta: 18:52:38 time: 0.5542 data_time: 0.0060 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0099 decode.acc_seg: 99.7107 aux.loss_ce: 0.0086 aux.acc_seg: 99.1987 +04/18 08:01:40 - mmengine - INFO - Iter(train) [ 36850/160000] lr: 7.9221e-03 eta: 18:52:11 time: 0.5554 data_time: 0.0064 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0084 decode.acc_seg: 99.7684 aux.loss_ce: 0.0073 aux.acc_seg: 99.3653 +04/18 08:02:08 - mmengine - INFO - Iter(train) [ 36900/160000] lr: 7.9192e-03 eta: 18:51:44 time: 0.5554 data_time: 0.0061 memory: 7635 loss: 0.0181 decode.loss_ce: 0.0097 decode.acc_seg: 99.6262 aux.loss_ce: 0.0084 aux.acc_seg: 99.1335 +04/18 08:02:36 - mmengine - INFO - Iter(train) [ 36950/160000] lr: 7.9163e-03 eta: 18:51:17 time: 0.5564 data_time: 0.0059 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0087 decode.acc_seg: 99.6028 aux.loss_ce: 0.0081 aux.acc_seg: 99.0002 +04/18 08:03:04 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 08:03:04 - mmengine - INFO - Iter(train) [ 37000/160000] lr: 7.9135e-03 eta: 18:50:51 time: 0.5559 data_time: 0.0069 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0089 decode.acc_seg: 99.7021 aux.loss_ce: 0.0077 aux.acc_seg: 99.1801 +04/18 08:03:32 - mmengine - INFO - Iter(train) [ 37050/160000] lr: 7.9106e-03 eta: 18:50:24 time: 0.5552 data_time: 0.0066 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0106 decode.acc_seg: 99.6867 aux.loss_ce: 0.0082 aux.acc_seg: 99.2112 +04/18 08:03:59 - mmengine - INFO - Iter(train) [ 37100/160000] lr: 7.9078e-03 eta: 18:49:57 time: 0.5555 data_time: 0.0062 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0099 decode.acc_seg: 99.6346 aux.loss_ce: 0.0084 aux.acc_seg: 99.0377 +04/18 08:04:27 - mmengine - INFO - Iter(train) [ 37150/160000] lr: 7.9049e-03 eta: 18:49:30 time: 0.5554 data_time: 0.0061 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0093 decode.acc_seg: 99.5802 aux.loss_ce: 0.0084 aux.acc_seg: 99.2209 +04/18 08:04:55 - mmengine - INFO - Iter(train) [ 37200/160000] lr: 7.9020e-03 eta: 18:49:03 time: 0.5567 data_time: 0.0067 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0093 decode.acc_seg: 99.6939 aux.loss_ce: 0.0080 aux.acc_seg: 99.2533 +04/18 08:05:23 - mmengine - INFO - Iter(train) [ 37250/160000] lr: 7.8992e-03 eta: 18:48:36 time: 0.5557 data_time: 0.0063 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0092 decode.acc_seg: 99.6069 aux.loss_ce: 0.0086 aux.acc_seg: 98.9867 +04/18 08:05:50 - mmengine - INFO - Iter(train) [ 37300/160000] lr: 7.8963e-03 eta: 18:48:09 time: 0.5555 data_time: 0.0070 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0087 decode.acc_seg: 99.4745 aux.loss_ce: 0.0082 aux.acc_seg: 99.0009 +04/18 08:06:18 - mmengine - INFO - Iter(train) [ 37350/160000] lr: 7.8935e-03 eta: 18:47:42 time: 0.5573 data_time: 0.0067 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0084 decode.acc_seg: 99.7446 aux.loss_ce: 0.0077 aux.acc_seg: 99.3535 +04/18 08:06:46 - mmengine - INFO - Iter(train) [ 37400/160000] lr: 7.8906e-03 eta: 18:47:15 time: 0.5537 data_time: 0.0063 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0084 decode.acc_seg: 99.6960 aux.loss_ce: 0.0073 aux.acc_seg: 99.2872 +04/18 08:07:14 - mmengine - INFO - Iter(train) [ 37450/160000] lr: 7.8877e-03 eta: 18:46:48 time: 0.5541 data_time: 0.0065 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0092 decode.acc_seg: 99.6699 aux.loss_ce: 0.0084 aux.acc_seg: 99.2398 +04/18 08:07:41 - mmengine - INFO - Iter(train) [ 37500/160000] lr: 7.8849e-03 eta: 18:46:20 time: 0.5549 data_time: 0.0064 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0089 decode.acc_seg: 99.6231 aux.loss_ce: 0.0084 aux.acc_seg: 99.0577 +04/18 08:08:09 - mmengine - INFO - Iter(train) [ 37550/160000] lr: 7.8820e-03 eta: 18:45:53 time: 0.5546 data_time: 0.0071 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0088 decode.acc_seg: 99.7350 aux.loss_ce: 0.0084 aux.acc_seg: 99.3443 +04/18 08:08:37 - mmengine - INFO - Iter(train) [ 37600/160000] lr: 7.8792e-03 eta: 18:45:26 time: 0.5532 data_time: 0.0064 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0079 decode.acc_seg: 99.6883 aux.loss_ce: 0.0071 aux.acc_seg: 99.3621 +04/18 08:09:05 - mmengine - INFO - Iter(train) [ 37650/160000] lr: 7.8763e-03 eta: 18:44:59 time: 0.5553 data_time: 0.0065 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0080 decode.acc_seg: 99.6845 aux.loss_ce: 0.0074 aux.acc_seg: 98.9697 +04/18 08:09:32 - mmengine - INFO - Iter(train) [ 37700/160000] lr: 7.8734e-03 eta: 18:44:32 time: 0.5538 data_time: 0.0067 memory: 7635 loss: 0.0199 decode.loss_ce: 0.0112 decode.acc_seg: 99.4838 aux.loss_ce: 0.0088 aux.acc_seg: 99.1152 +04/18 08:10:00 - mmengine - INFO - Iter(train) [ 37750/160000] lr: 7.8706e-03 eta: 18:44:05 time: 0.5533 data_time: 0.0062 memory: 7635 loss: 0.0206 decode.loss_ce: 0.0108 decode.acc_seg: 99.6160 aux.loss_ce: 0.0098 aux.acc_seg: 99.0363 +04/18 08:10:28 - mmengine - INFO - Iter(train) [ 37800/160000] lr: 7.8677e-03 eta: 18:43:38 time: 0.5540 data_time: 0.0065 memory: 7635 loss: 0.0190 decode.loss_ce: 0.0105 decode.acc_seg: 99.2393 aux.loss_ce: 0.0085 aux.acc_seg: 98.8298 +04/18 08:10:56 - mmengine - INFO - Iter(train) [ 37850/160000] lr: 7.8649e-03 eta: 18:43:11 time: 0.5554 data_time: 0.0060 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0090 decode.acc_seg: 99.6859 aux.loss_ce: 0.0080 aux.acc_seg: 99.2764 +04/18 08:11:23 - mmengine - INFO - Iter(train) [ 37900/160000] lr: 7.8620e-03 eta: 18:42:44 time: 0.5531 data_time: 0.0065 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0108 decode.acc_seg: 99.5868 aux.loss_ce: 0.0087 aux.acc_seg: 99.1085 +04/18 08:11:51 - mmengine - INFO - Iter(train) [ 37950/160000] lr: 7.8591e-03 eta: 18:42:16 time: 0.5546 data_time: 0.0061 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0101 decode.acc_seg: 99.6949 aux.loss_ce: 0.0086 aux.acc_seg: 99.2368 +04/18 08:12:19 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 08:12:19 - mmengine - INFO - Iter(train) [ 38000/160000] lr: 7.8563e-03 eta: 18:41:49 time: 0.5533 data_time: 0.0065 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0085 decode.acc_seg: 99.6777 aux.loss_ce: 0.0079 aux.acc_seg: 99.1671 +04/18 08:12:47 - mmengine - INFO - Iter(train) [ 38050/160000] lr: 7.8534e-03 eta: 18:41:23 time: 0.5631 data_time: 0.0062 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0073 decode.acc_seg: 99.5980 aux.loss_ce: 0.0070 aux.acc_seg: 99.1495 +04/18 08:13:14 - mmengine - INFO - Iter(train) [ 38100/160000] lr: 7.8506e-03 eta: 18:40:56 time: 0.5528 data_time: 0.0061 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0088 decode.acc_seg: 99.6484 aux.loss_ce: 0.0083 aux.acc_seg: 98.9731 +04/18 08:13:42 - mmengine - INFO - Iter(train) [ 38150/160000] lr: 7.8477e-03 eta: 18:40:28 time: 0.5538 data_time: 0.0067 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0083 decode.acc_seg: 99.6856 aux.loss_ce: 0.0077 aux.acc_seg: 99.0777 +04/18 08:14:10 - mmengine - INFO - Iter(train) [ 38200/160000] lr: 7.8448e-03 eta: 18:40:01 time: 0.5559 data_time: 0.0057 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0083 decode.acc_seg: 99.6652 aux.loss_ce: 0.0077 aux.acc_seg: 99.2684 +04/18 08:14:38 - mmengine - INFO - Iter(train) [ 38250/160000] lr: 7.8420e-03 eta: 18:39:34 time: 0.5537 data_time: 0.0071 memory: 7635 loss: 0.0199 decode.loss_ce: 0.0107 decode.acc_seg: 99.4609 aux.loss_ce: 0.0092 aux.acc_seg: 98.8550 +04/18 08:15:05 - mmengine - INFO - Iter(train) [ 38300/160000] lr: 7.8391e-03 eta: 18:39:07 time: 0.5519 data_time: 0.0060 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0086 decode.acc_seg: 99.5614 aux.loss_ce: 0.0081 aux.acc_seg: 99.1699 +04/18 08:15:33 - mmengine - INFO - Iter(train) [ 38350/160000] lr: 7.8363e-03 eta: 18:38:40 time: 0.5548 data_time: 0.0061 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0086 decode.acc_seg: 99.5927 aux.loss_ce: 0.0080 aux.acc_seg: 99.0862 +04/18 08:16:01 - mmengine - INFO - Iter(train) [ 38400/160000] lr: 7.8334e-03 eta: 18:38:13 time: 0.5544 data_time: 0.0064 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0092 decode.acc_seg: 99.7965 aux.loss_ce: 0.0082 aux.acc_seg: 99.5151 +04/18 08:16:28 - mmengine - INFO - Iter(train) [ 38450/160000] lr: 7.8305e-03 eta: 18:37:45 time: 0.5540 data_time: 0.0064 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0083 decode.acc_seg: 99.6346 aux.loss_ce: 0.0073 aux.acc_seg: 99.1762 +04/18 08:16:56 - mmengine - INFO - Iter(train) [ 38500/160000] lr: 7.8277e-03 eta: 18:37:18 time: 0.5554 data_time: 0.0072 memory: 7635 loss: 0.0186 decode.loss_ce: 0.0096 decode.acc_seg: 99.5115 aux.loss_ce: 0.0090 aux.acc_seg: 98.9328 +04/18 08:17:24 - mmengine - INFO - Iter(train) [ 38550/160000] lr: 7.8248e-03 eta: 18:36:51 time: 0.5544 data_time: 0.0064 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0094 decode.acc_seg: 99.6286 aux.loss_ce: 0.0083 aux.acc_seg: 99.1783 +04/18 08:17:52 - mmengine - INFO - Iter(train) [ 38600/160000] lr: 7.8219e-03 eta: 18:36:24 time: 0.5539 data_time: 0.0059 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0090 decode.acc_seg: 99.6795 aux.loss_ce: 0.0081 aux.acc_seg: 99.2602 +04/18 08:18:19 - mmengine - INFO - Iter(train) [ 38650/160000] lr: 7.8191e-03 eta: 18:35:57 time: 0.5525 data_time: 0.0060 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0095 decode.acc_seg: 99.5373 aux.loss_ce: 0.0087 aux.acc_seg: 98.7154 +04/18 08:18:47 - mmengine - INFO - Iter(train) [ 38700/160000] lr: 7.8162e-03 eta: 18:35:29 time: 0.5533 data_time: 0.0062 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0084 decode.acc_seg: 99.6734 aux.loss_ce: 0.0082 aux.acc_seg: 99.0753 +04/18 08:19:15 - mmengine - INFO - Iter(train) [ 38750/160000] lr: 7.8134e-03 eta: 18:35:02 time: 0.5524 data_time: 0.0062 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0082 decode.acc_seg: 99.6527 aux.loss_ce: 0.0077 aux.acc_seg: 99.2320 +04/18 08:19:42 - mmengine - INFO - Iter(train) [ 38800/160000] lr: 7.8105e-03 eta: 18:34:35 time: 0.5532 data_time: 0.0060 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0092 decode.acc_seg: 99.5894 aux.loss_ce: 0.0083 aux.acc_seg: 99.0193 +04/18 08:20:10 - mmengine - INFO - Iter(train) [ 38850/160000] lr: 7.8076e-03 eta: 18:34:07 time: 0.5535 data_time: 0.0067 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0089 decode.acc_seg: 99.6009 aux.loss_ce: 0.0087 aux.acc_seg: 98.7836 +04/18 08:20:38 - mmengine - INFO - Iter(train) [ 38900/160000] lr: 7.8048e-03 eta: 18:33:40 time: 0.5526 data_time: 0.0068 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0094 decode.acc_seg: 99.6873 aux.loss_ce: 0.0080 aux.acc_seg: 99.2844 +04/18 08:21:05 - mmengine - INFO - Iter(train) [ 38950/160000] lr: 7.8019e-03 eta: 18:33:13 time: 0.5547 data_time: 0.0062 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0102 decode.acc_seg: 99.6688 aux.loss_ce: 0.0087 aux.acc_seg: 99.1081 +04/18 08:21:33 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 08:21:33 - mmengine - INFO - Iter(train) [ 39000/160000] lr: 7.7990e-03 eta: 18:32:45 time: 0.5533 data_time: 0.0065 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0090 decode.acc_seg: 99.6576 aux.loss_ce: 0.0080 aux.acc_seg: 99.2667 +04/18 08:22:01 - mmengine - INFO - Iter(train) [ 39050/160000] lr: 7.7962e-03 eta: 18:32:18 time: 0.5548 data_time: 0.0071 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0092 decode.acc_seg: 99.6274 aux.loss_ce: 0.0081 aux.acc_seg: 99.0817 +04/18 08:22:29 - mmengine - INFO - Iter(train) [ 39100/160000] lr: 7.7933e-03 eta: 18:31:51 time: 0.5544 data_time: 0.0059 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0099 decode.acc_seg: 99.5752 aux.loss_ce: 0.0081 aux.acc_seg: 99.3050 +04/18 08:22:56 - mmengine - INFO - Iter(train) [ 39150/160000] lr: 7.7904e-03 eta: 18:31:24 time: 0.5536 data_time: 0.0059 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0088 decode.acc_seg: 99.5658 aux.loss_ce: 0.0083 aux.acc_seg: 99.1587 +04/18 08:23:24 - mmengine - INFO - Iter(train) [ 39200/160000] lr: 7.7876e-03 eta: 18:30:57 time: 0.5538 data_time: 0.0063 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0081 decode.acc_seg: 99.5652 aux.loss_ce: 0.0077 aux.acc_seg: 98.9863 +04/18 08:23:52 - mmengine - INFO - Iter(train) [ 39250/160000] lr: 7.7847e-03 eta: 18:30:30 time: 0.5512 data_time: 0.0065 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0098 decode.acc_seg: 99.7126 aux.loss_ce: 0.0081 aux.acc_seg: 99.4448 +04/18 08:24:19 - mmengine - INFO - Iter(train) [ 39300/160000] lr: 7.7819e-03 eta: 18:30:02 time: 0.5540 data_time: 0.0062 memory: 7635 loss: 0.0210 decode.loss_ce: 0.0116 decode.acc_seg: 99.6580 aux.loss_ce: 0.0094 aux.acc_seg: 99.0866 +04/18 08:24:47 - mmengine - INFO - Iter(train) [ 39350/160000] lr: 7.7790e-03 eta: 18:29:35 time: 0.5529 data_time: 0.0069 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0082 decode.acc_seg: 99.7506 aux.loss_ce: 0.0077 aux.acc_seg: 99.2886 +04/18 08:25:15 - mmengine - INFO - Iter(train) [ 39400/160000] lr: 7.7761e-03 eta: 18:29:08 time: 0.5525 data_time: 0.0062 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0096 decode.acc_seg: 99.6969 aux.loss_ce: 0.0085 aux.acc_seg: 99.2848 +04/18 08:25:42 - mmengine - INFO - Iter(train) [ 39450/160000] lr: 7.7733e-03 eta: 18:28:40 time: 0.5536 data_time: 0.0070 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0095 decode.acc_seg: 99.5288 aux.loss_ce: 0.0082 aux.acc_seg: 99.0371 +04/18 08:26:10 - mmengine - INFO - Iter(train) [ 39500/160000] lr: 7.7704e-03 eta: 18:28:13 time: 0.5525 data_time: 0.0067 memory: 7635 loss: 0.0202 decode.loss_ce: 0.0109 decode.acc_seg: 99.4141 aux.loss_ce: 0.0093 aux.acc_seg: 98.8440 +04/18 08:26:38 - mmengine - INFO - Iter(train) [ 39550/160000] lr: 7.7675e-03 eta: 18:27:45 time: 0.5533 data_time: 0.0057 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0092 decode.acc_seg: 99.7376 aux.loss_ce: 0.0085 aux.acc_seg: 99.2864 +04/18 08:27:05 - mmengine - INFO - Iter(train) [ 39600/160000] lr: 7.7647e-03 eta: 18:27:18 time: 0.5536 data_time: 0.0071 memory: 7635 loss: 0.0196 decode.loss_ce: 0.0105 decode.acc_seg: 99.5536 aux.loss_ce: 0.0091 aux.acc_seg: 99.0755 +04/18 08:27:33 - mmengine - INFO - Iter(train) [ 39650/160000] lr: 7.7618e-03 eta: 18:26:51 time: 0.5541 data_time: 0.0066 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0083 decode.acc_seg: 99.6278 aux.loss_ce: 0.0077 aux.acc_seg: 99.2082 +04/18 08:28:01 - mmengine - INFO - Iter(train) [ 39700/160000] lr: 7.7589e-03 eta: 18:26:23 time: 0.5530 data_time: 0.0059 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0093 decode.acc_seg: 99.4593 aux.loss_ce: 0.0086 aux.acc_seg: 98.7346 +04/18 08:28:28 - mmengine - INFO - Iter(train) [ 39750/160000] lr: 7.7561e-03 eta: 18:25:56 time: 0.5533 data_time: 0.0061 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0094 decode.acc_seg: 99.6967 aux.loss_ce: 0.0086 aux.acc_seg: 99.1289 +04/18 08:28:56 - mmengine - INFO - Iter(train) [ 39800/160000] lr: 7.7532e-03 eta: 18:25:29 time: 0.5540 data_time: 0.0065 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0086 decode.acc_seg: 99.6865 aux.loss_ce: 0.0081 aux.acc_seg: 99.2378 +04/18 08:29:24 - mmengine - INFO - Iter(train) [ 39850/160000] lr: 7.7503e-03 eta: 18:25:01 time: 0.5521 data_time: 0.0062 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0095 decode.acc_seg: 99.6639 aux.loss_ce: 0.0083 aux.acc_seg: 99.2950 +04/18 08:29:51 - mmengine - INFO - Iter(train) [ 39900/160000] lr: 7.7475e-03 eta: 18:24:34 time: 0.5523 data_time: 0.0071 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0089 decode.acc_seg: 99.5162 aux.loss_ce: 0.0084 aux.acc_seg: 98.6649 +04/18 08:30:19 - mmengine - INFO - Iter(train) [ 39950/160000] lr: 7.7446e-03 eta: 18:24:06 time: 0.5514 data_time: 0.0060 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0080 decode.acc_seg: 99.7220 aux.loss_ce: 0.0071 aux.acc_seg: 99.3871 +04/18 08:30:47 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 08:30:47 - mmengine - INFO - Iter(train) [ 40000/160000] lr: 7.7417e-03 eta: 18:23:38 time: 0.5515 data_time: 0.0061 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0082 decode.acc_seg: 99.6464 aux.loss_ce: 0.0077 aux.acc_seg: 99.1118 +04/18 08:30:47 - mmengine - INFO - Saving checkpoint at 40000 iterations +04/18 08:30:51 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:06 time: 0.0462 data_time: 0.0014 memory: 1657 +04/18 08:30:53 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:04 time: 0.0463 data_time: 0.0015 memory: 1657 +04/18 08:30:55 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:02 time: 0.0464 data_time: 0.0014 memory: 1657 +04/18 08:30:57 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.0456 data_time: 0.0014 memory: 1657 +04/18 08:30:58 - mmengine - INFO - per class results: +04/18 08:30:58 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 99.06 | 99.47 | 99.53 | 99.58 | 99.47 | +| contrast | 79.93 | 90.03 | 88.85 | 87.69 | 90.03 | ++------------+-------+-------+--------+-----------+--------+ +04/18 08:30:58 - mmengine - INFO - Iter(val) [200/200] aAcc: 99.1000 mIoU: 89.5000 mAcc: 94.7500 mFscore: 94.1900 mPrecision: 93.6400 mRecall: 94.7500 data_time: 0.0015 time: 0.0463 +04/18 08:31:25 - mmengine - INFO - Iter(train) [ 40050/160000] lr: 7.7389e-03 eta: 18:23:11 time: 0.5502 data_time: 0.0066 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0086 decode.acc_seg: 99.6489 aux.loss_ce: 0.0078 aux.acc_seg: 99.2826 +04/18 08:31:53 - mmengine - INFO - Iter(train) [ 40100/160000] lr: 7.7360e-03 eta: 18:22:43 time: 0.5498 data_time: 0.0061 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0076 decode.acc_seg: 99.6363 aux.loss_ce: 0.0073 aux.acc_seg: 99.0268 +04/18 08:32:20 - mmengine - INFO - Iter(train) [ 40150/160000] lr: 7.7332e-03 eta: 18:22:16 time: 0.5515 data_time: 0.0062 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0084 decode.acc_seg: 99.6901 aux.loss_ce: 0.0076 aux.acc_seg: 99.2650 +04/18 08:32:48 - mmengine - INFO - Iter(train) [ 40200/160000] lr: 7.7303e-03 eta: 18:21:48 time: 0.5500 data_time: 0.0062 memory: 7635 loss: 0.0214 decode.loss_ce: 0.0116 decode.acc_seg: 99.6566 aux.loss_ce: 0.0098 aux.acc_seg: 99.1296 +04/18 08:33:15 - mmengine - INFO - Iter(train) [ 40250/160000] lr: 7.7274e-03 eta: 18:21:20 time: 0.5496 data_time: 0.0062 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0106 decode.acc_seg: 99.5221 aux.loss_ce: 0.0085 aux.acc_seg: 99.0771 +04/18 08:33:43 - mmengine - INFO - Iter(train) [ 40300/160000] lr: 7.7246e-03 eta: 18:20:53 time: 0.5524 data_time: 0.0067 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0090 decode.acc_seg: 99.6153 aux.loss_ce: 0.0079 aux.acc_seg: 99.1534 +04/18 08:34:10 - mmengine - INFO - Iter(train) [ 40350/160000] lr: 7.7217e-03 eta: 18:20:25 time: 0.5520 data_time: 0.0069 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0086 decode.acc_seg: 99.7101 aux.loss_ce: 0.0074 aux.acc_seg: 99.4150 +04/18 08:34:38 - mmengine - INFO - Iter(train) [ 40400/160000] lr: 7.7188e-03 eta: 18:19:57 time: 0.5489 data_time: 0.0062 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0087 decode.acc_seg: 99.6420 aux.loss_ce: 0.0079 aux.acc_seg: 99.0813 +04/18 08:35:06 - mmengine - INFO - Iter(train) [ 40450/160000] lr: 7.7160e-03 eta: 18:19:29 time: 0.5513 data_time: 0.0070 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0081 decode.acc_seg: 99.6397 aux.loss_ce: 0.0074 aux.acc_seg: 99.1655 +04/18 08:35:33 - mmengine - INFO - Iter(train) [ 40500/160000] lr: 7.7131e-03 eta: 18:19:02 time: 0.5514 data_time: 0.0060 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0090 decode.acc_seg: 99.6161 aux.loss_ce: 0.0078 aux.acc_seg: 99.1043 +04/18 08:36:01 - mmengine - INFO - Iter(train) [ 40550/160000] lr: 7.7102e-03 eta: 18:18:34 time: 0.5513 data_time: 0.0067 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0088 decode.acc_seg: 99.5161 aux.loss_ce: 0.0080 aux.acc_seg: 98.9918 +04/18 08:36:28 - mmengine - INFO - Iter(train) [ 40600/160000] lr: 7.7074e-03 eta: 18:18:06 time: 0.5497 data_time: 0.0069 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0085 decode.acc_seg: 99.6079 aux.loss_ce: 0.0080 aux.acc_seg: 99.0249 +04/18 08:36:56 - mmengine - INFO - Iter(train) [ 40650/160000] lr: 7.7045e-03 eta: 18:17:39 time: 0.5511 data_time: 0.0061 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0088 decode.acc_seg: 99.5480 aux.loss_ce: 0.0082 aux.acc_seg: 99.1893 +04/18 08:37:23 - mmengine - INFO - Iter(train) [ 40700/160000] lr: 7.7016e-03 eta: 18:17:11 time: 0.5499 data_time: 0.0060 memory: 7635 loss: 0.0201 decode.loss_ce: 0.0106 decode.acc_seg: 99.5810 aux.loss_ce: 0.0095 aux.acc_seg: 99.2526 +04/18 08:37:51 - mmengine - INFO - Iter(train) [ 40750/160000] lr: 7.6988e-03 eta: 18:16:43 time: 0.5503 data_time: 0.0067 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0090 decode.acc_seg: 99.6409 aux.loss_ce: 0.0081 aux.acc_seg: 99.0313 +04/18 08:38:18 - mmengine - INFO - Iter(train) [ 40800/160000] lr: 7.6959e-03 eta: 18:16:15 time: 0.5496 data_time: 0.0067 memory: 7635 loss: 0.0197 decode.loss_ce: 0.0108 decode.acc_seg: 99.6670 aux.loss_ce: 0.0088 aux.acc_seg: 99.2678 +04/18 08:38:46 - mmengine - INFO - Iter(train) [ 40850/160000] lr: 7.6930e-03 eta: 18:15:48 time: 0.5506 data_time: 0.0064 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0087 decode.acc_seg: 99.5914 aux.loss_ce: 0.0077 aux.acc_seg: 99.2438 +04/18 08:39:13 - mmengine - INFO - Iter(train) [ 40900/160000] lr: 7.6901e-03 eta: 18:15:20 time: 0.5510 data_time: 0.0063 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0087 decode.acc_seg: 99.6388 aux.loss_ce: 0.0077 aux.acc_seg: 99.3010 +04/18 08:39:41 - mmengine - INFO - Iter(train) [ 40950/160000] lr: 7.6873e-03 eta: 18:14:52 time: 0.5518 data_time: 0.0066 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0080 decode.acc_seg: 99.7011 aux.loss_ce: 0.0078 aux.acc_seg: 99.4434 +04/18 08:40:09 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 08:40:09 - mmengine - INFO - Iter(train) [ 41000/160000] lr: 7.6844e-03 eta: 18:14:25 time: 0.5506 data_time: 0.0065 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0093 decode.acc_seg: 99.6187 aux.loss_ce: 0.0085 aux.acc_seg: 98.9475 +04/18 08:40:36 - mmengine - INFO - Iter(train) [ 41050/160000] lr: 7.6815e-03 eta: 18:13:57 time: 0.5507 data_time: 0.0066 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0090 decode.acc_seg: 99.6560 aux.loss_ce: 0.0081 aux.acc_seg: 99.3476 +04/18 08:41:04 - mmengine - INFO - Iter(train) [ 41100/160000] lr: 7.6787e-03 eta: 18:13:29 time: 0.5513 data_time: 0.0060 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0081 decode.acc_seg: 99.6568 aux.loss_ce: 0.0077 aux.acc_seg: 99.0934 +04/18 08:41:31 - mmengine - INFO - Iter(train) [ 41150/160000] lr: 7.6758e-03 eta: 18:13:01 time: 0.5521 data_time: 0.0067 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0087 decode.acc_seg: 99.7006 aux.loss_ce: 0.0076 aux.acc_seg: 99.2484 +04/18 08:41:59 - mmengine - INFO - Iter(train) [ 41200/160000] lr: 7.6729e-03 eta: 18:12:34 time: 0.5503 data_time: 0.0064 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0097 decode.acc_seg: 99.5850 aux.loss_ce: 0.0087 aux.acc_seg: 99.0829 +04/18 08:42:26 - mmengine - INFO - Iter(train) [ 41250/160000] lr: 7.6701e-03 eta: 18:12:06 time: 0.5498 data_time: 0.0062 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0092 decode.acc_seg: 99.6686 aux.loss_ce: 0.0086 aux.acc_seg: 98.9568 +04/18 08:42:54 - mmengine - INFO - Iter(train) [ 41300/160000] lr: 7.6672e-03 eta: 18:11:39 time: 0.5518 data_time: 0.0070 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0088 decode.acc_seg: 99.6409 aux.loss_ce: 0.0080 aux.acc_seg: 99.1268 +04/18 08:43:22 - mmengine - INFO - Iter(train) [ 41350/160000] lr: 7.6643e-03 eta: 18:11:11 time: 0.5498 data_time: 0.0061 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0081 decode.acc_seg: 99.6580 aux.loss_ce: 0.0075 aux.acc_seg: 99.0657 +04/18 08:43:49 - mmengine - INFO - Iter(train) [ 41400/160000] lr: 7.6615e-03 eta: 18:10:43 time: 0.5502 data_time: 0.0059 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0086 decode.acc_seg: 99.6943 aux.loss_ce: 0.0076 aux.acc_seg: 99.2686 +04/18 08:44:17 - mmengine - INFO - Iter(train) [ 41450/160000] lr: 7.6586e-03 eta: 18:10:15 time: 0.5504 data_time: 0.0062 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0089 decode.acc_seg: 99.5800 aux.loss_ce: 0.0084 aux.acc_seg: 99.1014 +04/18 08:44:44 - mmengine - INFO - Iter(train) [ 41500/160000] lr: 7.6557e-03 eta: 18:09:48 time: 0.5502 data_time: 0.0061 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0081 decode.acc_seg: 99.6276 aux.loss_ce: 0.0077 aux.acc_seg: 99.1901 +04/18 08:45:12 - mmengine - INFO - Iter(train) [ 41550/160000] lr: 7.6529e-03 eta: 18:09:20 time: 0.5602 data_time: 0.0078 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0080 decode.acc_seg: 99.6048 aux.loss_ce: 0.0076 aux.acc_seg: 99.1132 +04/18 08:45:39 - mmengine - INFO - Iter(train) [ 41600/160000] lr: 7.6500e-03 eta: 18:08:53 time: 0.5518 data_time: 0.0066 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0090 decode.acc_seg: 99.6599 aux.loss_ce: 0.0085 aux.acc_seg: 99.1570 +04/18 08:46:07 - mmengine - INFO - Iter(train) [ 41650/160000] lr: 7.6471e-03 eta: 18:08:25 time: 0.5505 data_time: 0.0061 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0083 decode.acc_seg: 99.6052 aux.loss_ce: 0.0079 aux.acc_seg: 99.1905 +04/18 08:46:35 - mmengine - INFO - Iter(train) [ 41700/160000] lr: 7.6442e-03 eta: 18:07:57 time: 0.5510 data_time: 0.0064 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0094 decode.acc_seg: 99.6144 aux.loss_ce: 0.0083 aux.acc_seg: 99.2826 +04/18 08:47:02 - mmengine - INFO - Iter(train) [ 41750/160000] lr: 7.6414e-03 eta: 18:07:29 time: 0.5497 data_time: 0.0063 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0086 decode.acc_seg: 99.6094 aux.loss_ce: 0.0080 aux.acc_seg: 99.2477 +04/18 08:47:30 - mmengine - INFO - Iter(train) [ 41800/160000] lr: 7.6385e-03 eta: 18:07:02 time: 0.5517 data_time: 0.0066 memory: 7635 loss: 0.0203 decode.loss_ce: 0.0111 decode.acc_seg: 99.6124 aux.loss_ce: 0.0092 aux.acc_seg: 99.1284 +04/18 08:47:57 - mmengine - INFO - Iter(train) [ 41850/160000] lr: 7.6356e-03 eta: 18:06:34 time: 0.5502 data_time: 0.0059 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0086 decode.acc_seg: 99.5851 aux.loss_ce: 0.0079 aux.acc_seg: 98.9687 +04/18 08:48:25 - mmengine - INFO - Iter(train) [ 41900/160000] lr: 7.6328e-03 eta: 18:06:07 time: 0.5513 data_time: 0.0059 memory: 7635 loss: 0.0201 decode.loss_ce: 0.0113 decode.acc_seg: 99.1785 aux.loss_ce: 0.0088 aux.acc_seg: 98.7809 +04/18 08:48:52 - mmengine - INFO - Iter(train) [ 41950/160000] lr: 7.6299e-03 eta: 18:05:39 time: 0.5517 data_time: 0.0065 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0093 decode.acc_seg: 99.7153 aux.loss_ce: 0.0086 aux.acc_seg: 99.2450 +04/18 08:49:20 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 08:49:20 - mmengine - INFO - Iter(train) [ 42000/160000] lr: 7.6270e-03 eta: 18:05:11 time: 0.5518 data_time: 0.0068 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0101 decode.acc_seg: 99.7354 aux.loss_ce: 0.0086 aux.acc_seg: 99.2950 +04/18 08:49:48 - mmengine - INFO - Iter(train) [ 42050/160000] lr: 7.6242e-03 eta: 18:04:44 time: 0.5514 data_time: 0.0069 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0093 decode.acc_seg: 99.6522 aux.loss_ce: 0.0082 aux.acc_seg: 99.2332 +04/18 08:50:15 - mmengine - INFO - Iter(train) [ 42100/160000] lr: 7.6213e-03 eta: 18:04:16 time: 0.5517 data_time: 0.0066 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0089 decode.acc_seg: 99.6117 aux.loss_ce: 0.0082 aux.acc_seg: 98.8979 +04/18 08:50:43 - mmengine - INFO - Iter(train) [ 42150/160000] lr: 7.6184e-03 eta: 18:03:48 time: 0.5510 data_time: 0.0067 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0092 decode.acc_seg: 99.4720 aux.loss_ce: 0.0081 aux.acc_seg: 99.0637 +04/18 08:51:10 - mmengine - INFO - Iter(train) [ 42200/160000] lr: 7.6155e-03 eta: 18:03:21 time: 0.5513 data_time: 0.0068 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0094 decode.acc_seg: 99.6223 aux.loss_ce: 0.0085 aux.acc_seg: 99.1876 +04/18 08:51:38 - mmengine - INFO - Iter(train) [ 42250/160000] lr: 7.6127e-03 eta: 18:02:53 time: 0.5508 data_time: 0.0059 memory: 7635 loss: 0.0186 decode.loss_ce: 0.0098 decode.acc_seg: 99.6596 aux.loss_ce: 0.0088 aux.acc_seg: 99.2562 +04/18 08:52:05 - mmengine - INFO - Iter(train) [ 42300/160000] lr: 7.6098e-03 eta: 18:02:25 time: 0.5529 data_time: 0.0075 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0095 decode.acc_seg: 99.6082 aux.loss_ce: 0.0087 aux.acc_seg: 99.1274 +04/18 08:52:33 - mmengine - INFO - Iter(train) [ 42350/160000] lr: 7.6069e-03 eta: 18:01:58 time: 0.5608 data_time: 0.0065 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0080 decode.acc_seg: 99.6665 aux.loss_ce: 0.0077 aux.acc_seg: 99.0905 +04/18 08:53:01 - mmengine - INFO - Iter(train) [ 42400/160000] lr: 7.6041e-03 eta: 18:01:31 time: 0.5522 data_time: 0.0064 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0087 decode.acc_seg: 99.7198 aux.loss_ce: 0.0081 aux.acc_seg: 99.2658 +04/18 08:53:28 - mmengine - INFO - Iter(train) [ 42450/160000] lr: 7.6012e-03 eta: 18:01:03 time: 0.5516 data_time: 0.0059 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0078 decode.acc_seg: 99.7198 aux.loss_ce: 0.0074 aux.acc_seg: 99.4457 +04/18 08:53:56 - mmengine - INFO - Iter(train) [ 42500/160000] lr: 7.5983e-03 eta: 18:00:36 time: 0.5513 data_time: 0.0067 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0080 decode.acc_seg: 99.7454 aux.loss_ce: 0.0076 aux.acc_seg: 99.2842 +04/18 08:54:24 - mmengine - INFO - Iter(train) [ 42550/160000] lr: 7.5954e-03 eta: 18:00:08 time: 0.5515 data_time: 0.0068 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0087 decode.acc_seg: 99.5263 aux.loss_ce: 0.0079 aux.acc_seg: 99.0045 +04/18 08:54:51 - mmengine - INFO - Iter(train) [ 42600/160000] lr: 7.5926e-03 eta: 17:59:40 time: 0.5525 data_time: 0.0064 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0084 decode.acc_seg: 99.5923 aux.loss_ce: 0.0079 aux.acc_seg: 98.9909 +04/18 08:55:19 - mmengine - INFO - Iter(train) [ 42650/160000] lr: 7.5897e-03 eta: 17:59:13 time: 0.5536 data_time: 0.0074 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0099 decode.acc_seg: 99.6208 aux.loss_ce: 0.0082 aux.acc_seg: 99.2052 +04/18 08:55:47 - mmengine - INFO - Iter(train) [ 42700/160000] lr: 7.5868e-03 eta: 17:58:46 time: 0.5515 data_time: 0.0064 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0100 decode.acc_seg: 99.7200 aux.loss_ce: 0.0082 aux.acc_seg: 99.3009 +04/18 08:56:14 - mmengine - INFO - Iter(train) [ 42750/160000] lr: 7.5840e-03 eta: 17:58:18 time: 0.5506 data_time: 0.0065 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0091 decode.acc_seg: 99.6605 aux.loss_ce: 0.0084 aux.acc_seg: 99.1475 +04/18 08:56:42 - mmengine - INFO - Iter(train) [ 42800/160000] lr: 7.5811e-03 eta: 17:57:50 time: 0.5513 data_time: 0.0064 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0073 decode.acc_seg: 99.7220 aux.loss_ce: 0.0070 aux.acc_seg: 99.3004 +04/18 08:57:09 - mmengine - INFO - Iter(train) [ 42850/160000] lr: 7.5782e-03 eta: 17:57:23 time: 0.5531 data_time: 0.0067 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0083 decode.acc_seg: 99.7429 aux.loss_ce: 0.0077 aux.acc_seg: 99.2773 +04/18 08:57:37 - mmengine - INFO - Iter(train) [ 42900/160000] lr: 7.5753e-03 eta: 17:56:55 time: 0.5513 data_time: 0.0063 memory: 7635 loss: 0.0219 decode.loss_ce: 0.0122 decode.acc_seg: 99.5868 aux.loss_ce: 0.0097 aux.acc_seg: 99.0719 +04/18 08:58:04 - mmengine - INFO - Iter(train) [ 42950/160000] lr: 7.5725e-03 eta: 17:56:27 time: 0.5522 data_time: 0.0069 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0085 decode.acc_seg: 99.6851 aux.loss_ce: 0.0079 aux.acc_seg: 99.1350 +04/18 08:58:32 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 08:58:32 - mmengine - INFO - Iter(train) [ 43000/160000] lr: 7.5696e-03 eta: 17:56:00 time: 0.5515 data_time: 0.0061 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0096 decode.acc_seg: 99.5961 aux.loss_ce: 0.0088 aux.acc_seg: 99.0003 +04/18 08:59:00 - mmengine - INFO - Iter(train) [ 43050/160000] lr: 7.5667e-03 eta: 17:55:32 time: 0.5530 data_time: 0.0069 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0099 decode.acc_seg: 99.4495 aux.loss_ce: 0.0085 aux.acc_seg: 98.6427 +04/18 08:59:27 - mmengine - INFO - Iter(train) [ 43100/160000] lr: 7.5638e-03 eta: 17:55:05 time: 0.5527 data_time: 0.0061 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0097 decode.acc_seg: 99.7053 aux.loss_ce: 0.0085 aux.acc_seg: 99.2823 +04/18 08:59:55 - mmengine - INFO - Iter(train) [ 43150/160000] lr: 7.5610e-03 eta: 17:54:37 time: 0.5523 data_time: 0.0061 memory: 7635 loss: 0.0194 decode.loss_ce: 0.0102 decode.acc_seg: 99.6934 aux.loss_ce: 0.0092 aux.acc_seg: 99.2806 +04/18 09:00:22 - mmengine - INFO - Iter(train) [ 43200/160000] lr: 7.5581e-03 eta: 17:54:09 time: 0.5502 data_time: 0.0060 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0086 decode.acc_seg: 99.6169 aux.loss_ce: 0.0084 aux.acc_seg: 99.1678 +04/18 09:00:50 - mmengine - INFO - Iter(train) [ 43250/160000] lr: 7.5552e-03 eta: 17:53:42 time: 0.5511 data_time: 0.0070 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0084 decode.acc_seg: 99.8030 aux.loss_ce: 0.0081 aux.acc_seg: 99.2701 +04/18 09:01:18 - mmengine - INFO - Iter(train) [ 43300/160000] lr: 7.5524e-03 eta: 17:53:14 time: 0.5518 data_time: 0.0064 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0081 decode.acc_seg: 99.7141 aux.loss_ce: 0.0076 aux.acc_seg: 99.3835 +04/18 09:01:45 - mmengine - INFO - Iter(train) [ 43350/160000] lr: 7.5495e-03 eta: 17:52:47 time: 0.5523 data_time: 0.0068 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0087 decode.acc_seg: 99.6002 aux.loss_ce: 0.0079 aux.acc_seg: 99.1570 +04/18 09:02:13 - mmengine - INFO - Iter(train) [ 43400/160000] lr: 7.5466e-03 eta: 17:52:19 time: 0.5535 data_time: 0.0069 memory: 7635 loss: 0.0181 decode.loss_ce: 0.0096 decode.acc_seg: 99.7343 aux.loss_ce: 0.0085 aux.acc_seg: 99.3286 +04/18 09:02:41 - mmengine - INFO - Iter(train) [ 43450/160000] lr: 7.5437e-03 eta: 17:51:52 time: 0.5505 data_time: 0.0058 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0096 decode.acc_seg: 99.6410 aux.loss_ce: 0.0082 aux.acc_seg: 99.2811 +04/18 09:03:08 - mmengine - INFO - Iter(train) [ 43500/160000] lr: 7.5409e-03 eta: 17:51:24 time: 0.5511 data_time: 0.0071 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0086 decode.acc_seg: 99.6983 aux.loss_ce: 0.0078 aux.acc_seg: 99.2352 +04/18 09:03:36 - mmengine - INFO - Iter(train) [ 43550/160000] lr: 7.5380e-03 eta: 17:50:57 time: 0.5500 data_time: 0.0059 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0091 decode.acc_seg: 99.5979 aux.loss_ce: 0.0083 aux.acc_seg: 99.1764 +04/18 09:04:03 - mmengine - INFO - Iter(train) [ 43600/160000] lr: 7.5351e-03 eta: 17:50:29 time: 0.5506 data_time: 0.0060 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0090 decode.acc_seg: 99.7025 aux.loss_ce: 0.0082 aux.acc_seg: 99.3903 +04/18 09:04:31 - mmengine - INFO - Iter(train) [ 43650/160000] lr: 7.5322e-03 eta: 17:50:02 time: 0.5519 data_time: 0.0063 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0104 decode.acc_seg: 99.5419 aux.loss_ce: 0.0085 aux.acc_seg: 99.0887 +04/18 09:04:59 - mmengine - INFO - Iter(train) [ 43700/160000] lr: 7.5294e-03 eta: 17:49:34 time: 0.5517 data_time: 0.0058 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0092 decode.acc_seg: 99.7018 aux.loss_ce: 0.0086 aux.acc_seg: 99.1895 +04/18 09:05:26 - mmengine - INFO - Iter(train) [ 43750/160000] lr: 7.5265e-03 eta: 17:49:07 time: 0.5528 data_time: 0.0061 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0095 decode.acc_seg: 99.6678 aux.loss_ce: 0.0084 aux.acc_seg: 99.2646 +04/18 09:05:54 - mmengine - INFO - Iter(train) [ 43800/160000] lr: 7.5236e-03 eta: 17:48:39 time: 0.5531 data_time: 0.0064 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0089 decode.acc_seg: 99.6360 aux.loss_ce: 0.0082 aux.acc_seg: 98.9284 +04/18 09:06:21 - mmengine - INFO - Iter(train) [ 43850/160000] lr: 7.5207e-03 eta: 17:48:11 time: 0.5518 data_time: 0.0061 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0087 decode.acc_seg: 99.7040 aux.loss_ce: 0.0078 aux.acc_seg: 99.3532 +04/18 09:06:49 - mmengine - INFO - Iter(train) [ 43900/160000] lr: 7.5179e-03 eta: 17:47:44 time: 0.5518 data_time: 0.0067 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0090 decode.acc_seg: 99.6299 aux.loss_ce: 0.0081 aux.acc_seg: 99.2554 +04/18 09:07:17 - mmengine - INFO - Iter(train) [ 43950/160000] lr: 7.5150e-03 eta: 17:47:16 time: 0.5502 data_time: 0.0063 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0093 decode.acc_seg: 99.5738 aux.loss_ce: 0.0082 aux.acc_seg: 99.0312 +04/18 09:07:44 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 09:07:44 - mmengine - INFO - Iter(train) [ 44000/160000] lr: 7.5121e-03 eta: 17:46:49 time: 0.5526 data_time: 0.0065 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0090 decode.acc_seg: 99.5694 aux.loss_ce: 0.0082 aux.acc_seg: 99.0113 +04/18 09:08:12 - mmengine - INFO - Iter(train) [ 44050/160000] lr: 7.5092e-03 eta: 17:46:21 time: 0.5525 data_time: 0.0067 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0090 decode.acc_seg: 99.7020 aux.loss_ce: 0.0083 aux.acc_seg: 99.2927 +04/18 09:08:39 - mmengine - INFO - Iter(train) [ 44100/160000] lr: 7.5064e-03 eta: 17:45:54 time: 0.5527 data_time: 0.0062 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0085 decode.acc_seg: 99.7297 aux.loss_ce: 0.0081 aux.acc_seg: 99.3670 +04/18 09:09:07 - mmengine - INFO - Iter(train) [ 44150/160000] lr: 7.5035e-03 eta: 17:45:26 time: 0.5522 data_time: 0.0063 memory: 7635 loss: 0.0186 decode.loss_ce: 0.0095 decode.acc_seg: 99.6385 aux.loss_ce: 0.0090 aux.acc_seg: 99.0378 +04/18 09:09:35 - mmengine - INFO - Iter(train) [ 44200/160000] lr: 7.5006e-03 eta: 17:44:59 time: 0.5520 data_time: 0.0074 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0094 decode.acc_seg: 99.6935 aux.loss_ce: 0.0085 aux.acc_seg: 99.0993 +04/18 09:10:02 - mmengine - INFO - Iter(train) [ 44250/160000] lr: 7.4977e-03 eta: 17:44:31 time: 0.5524 data_time: 0.0062 memory: 7635 loss: 0.0196 decode.loss_ce: 0.0103 decode.acc_seg: 99.6043 aux.loss_ce: 0.0093 aux.acc_seg: 98.8139 +04/18 09:10:30 - mmengine - INFO - Iter(train) [ 44300/160000] lr: 7.4949e-03 eta: 17:44:03 time: 0.5511 data_time: 0.0058 memory: 7635 loss: 0.0197 decode.loss_ce: 0.0107 decode.acc_seg: 99.6945 aux.loss_ce: 0.0090 aux.acc_seg: 99.2631 +04/18 09:10:57 - mmengine - INFO - Iter(train) [ 44350/160000] lr: 7.4920e-03 eta: 17:43:36 time: 0.5506 data_time: 0.0063 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0085 decode.acc_seg: 99.6607 aux.loss_ce: 0.0079 aux.acc_seg: 99.0323 +04/18 09:11:25 - mmengine - INFO - Iter(train) [ 44400/160000] lr: 7.4891e-03 eta: 17:43:08 time: 0.5532 data_time: 0.0067 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0094 decode.acc_seg: 99.6331 aux.loss_ce: 0.0085 aux.acc_seg: 99.1103 +04/18 09:11:53 - mmengine - INFO - Iter(train) [ 44450/160000] lr: 7.4862e-03 eta: 17:42:41 time: 0.5526 data_time: 0.0064 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0089 decode.acc_seg: 99.5473 aux.loss_ce: 0.0082 aux.acc_seg: 99.2693 +04/18 09:12:20 - mmengine - INFO - Iter(train) [ 44500/160000] lr: 7.4833e-03 eta: 17:42:14 time: 0.5713 data_time: 0.0069 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0096 decode.acc_seg: 99.7069 aux.loss_ce: 0.0089 aux.acc_seg: 99.3418 +04/18 09:12:48 - mmengine - INFO - Iter(train) [ 44550/160000] lr: 7.4805e-03 eta: 17:41:46 time: 0.5519 data_time: 0.0060 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0089 decode.acc_seg: 99.6710 aux.loss_ce: 0.0083 aux.acc_seg: 99.2764 +04/18 09:13:16 - mmengine - INFO - Iter(train) [ 44600/160000] lr: 7.4776e-03 eta: 17:41:19 time: 0.5518 data_time: 0.0060 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0100 decode.acc_seg: 99.6036 aux.loss_ce: 0.0087 aux.acc_seg: 99.1800 +04/18 09:13:43 - mmengine - INFO - Iter(train) [ 44650/160000] lr: 7.4747e-03 eta: 17:40:51 time: 0.5531 data_time: 0.0062 memory: 7635 loss: 0.0201 decode.loss_ce: 0.0108 decode.acc_seg: 99.5868 aux.loss_ce: 0.0093 aux.acc_seg: 99.1148 +04/18 09:14:11 - mmengine - INFO - Iter(train) [ 44700/160000] lr: 7.4718e-03 eta: 17:40:23 time: 0.5532 data_time: 0.0065 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0083 decode.acc_seg: 99.7260 aux.loss_ce: 0.0076 aux.acc_seg: 99.3047 +04/18 09:14:39 - mmengine - INFO - Iter(train) [ 44750/160000] lr: 7.4690e-03 eta: 17:39:56 time: 0.5532 data_time: 0.0061 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0094 decode.acc_seg: 99.6604 aux.loss_ce: 0.0094 aux.acc_seg: 99.2559 +04/18 09:15:06 - mmengine - INFO - Iter(train) [ 44800/160000] lr: 7.4661e-03 eta: 17:39:29 time: 0.5505 data_time: 0.0065 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0081 decode.acc_seg: 99.7090 aux.loss_ce: 0.0075 aux.acc_seg: 99.2836 +04/18 09:15:34 - mmengine - INFO - Iter(train) [ 44850/160000] lr: 7.4632e-03 eta: 17:39:01 time: 0.5527 data_time: 0.0063 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0093 decode.acc_seg: 99.7059 aux.loss_ce: 0.0083 aux.acc_seg: 99.1607 +04/18 09:16:01 - mmengine - INFO - Iter(train) [ 44900/160000] lr: 7.4603e-03 eta: 17:38:33 time: 0.5514 data_time: 0.0064 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0088 decode.acc_seg: 99.6642 aux.loss_ce: 0.0080 aux.acc_seg: 99.3046 +04/18 09:16:29 - mmengine - INFO - Iter(train) [ 44950/160000] lr: 7.4575e-03 eta: 17:38:06 time: 0.5534 data_time: 0.0066 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0095 decode.acc_seg: 99.5405 aux.loss_ce: 0.0080 aux.acc_seg: 99.0601 +04/18 09:16:57 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 09:16:57 - mmengine - INFO - Iter(train) [ 45000/160000] lr: 7.4546e-03 eta: 17:37:38 time: 0.5516 data_time: 0.0058 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0087 decode.acc_seg: 99.6920 aux.loss_ce: 0.0082 aux.acc_seg: 99.1603 +04/18 09:17:24 - mmengine - INFO - Iter(train) [ 45050/160000] lr: 7.4517e-03 eta: 17:37:11 time: 0.5528 data_time: 0.0059 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0093 decode.acc_seg: 99.7298 aux.loss_ce: 0.0084 aux.acc_seg: 99.3738 +04/18 09:17:52 - mmengine - INFO - Iter(train) [ 45100/160000] lr: 7.4488e-03 eta: 17:36:43 time: 0.5535 data_time: 0.0062 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0104 decode.acc_seg: 99.5881 aux.loss_ce: 0.0088 aux.acc_seg: 99.1511 +04/18 09:18:20 - mmengine - INFO - Iter(train) [ 45150/160000] lr: 7.4459e-03 eta: 17:36:16 time: 0.5501 data_time: 0.0067 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0087 decode.acc_seg: 99.6174 aux.loss_ce: 0.0080 aux.acc_seg: 99.2241 +04/18 09:18:47 - mmengine - INFO - Iter(train) [ 45200/160000] lr: 7.4431e-03 eta: 17:35:48 time: 0.5519 data_time: 0.0065 memory: 7635 loss: 0.0193 decode.loss_ce: 0.0106 decode.acc_seg: 99.2593 aux.loss_ce: 0.0086 aux.acc_seg: 98.8551 +04/18 09:19:15 - mmengine - INFO - Iter(train) [ 45250/160000] lr: 7.4402e-03 eta: 17:35:21 time: 0.5512 data_time: 0.0066 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0088 decode.acc_seg: 99.6578 aux.loss_ce: 0.0082 aux.acc_seg: 99.1809 +04/18 09:19:42 - mmengine - INFO - Iter(train) [ 45300/160000] lr: 7.4373e-03 eta: 17:34:53 time: 0.5518 data_time: 0.0065 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0088 decode.acc_seg: 99.6301 aux.loss_ce: 0.0082 aux.acc_seg: 99.1590 +04/18 09:20:10 - mmengine - INFO - Iter(train) [ 45350/160000] lr: 7.4344e-03 eta: 17:34:26 time: 0.5525 data_time: 0.0058 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0096 decode.acc_seg: 99.2689 aux.loss_ce: 0.0088 aux.acc_seg: 98.5504 +04/18 09:20:38 - mmengine - INFO - Iter(train) [ 45400/160000] lr: 7.4316e-03 eta: 17:33:58 time: 0.5512 data_time: 0.0065 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0092 decode.acc_seg: 99.6595 aux.loss_ce: 0.0085 aux.acc_seg: 99.1810 +04/18 09:21:05 - mmengine - INFO - Iter(train) [ 45450/160000] lr: 7.4287e-03 eta: 17:33:31 time: 0.5519 data_time: 0.0073 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0084 decode.acc_seg: 99.5793 aux.loss_ce: 0.0075 aux.acc_seg: 99.0155 +04/18 09:21:33 - mmengine - INFO - Iter(train) [ 45500/160000] lr: 7.4258e-03 eta: 17:33:03 time: 0.5536 data_time: 0.0067 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0086 decode.acc_seg: 99.6611 aux.loss_ce: 0.0080 aux.acc_seg: 99.2641 +04/18 09:22:00 - mmengine - INFO - Iter(train) [ 45550/160000] lr: 7.4229e-03 eta: 17:32:36 time: 0.5537 data_time: 0.0072 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0082 decode.acc_seg: 99.6822 aux.loss_ce: 0.0075 aux.acc_seg: 99.3224 +04/18 09:22:28 - mmengine - INFO - Iter(train) [ 45600/160000] lr: 7.4200e-03 eta: 17:32:08 time: 0.5517 data_time: 0.0063 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0083 decode.acc_seg: 99.6726 aux.loss_ce: 0.0076 aux.acc_seg: 99.3276 +04/18 09:22:56 - mmengine - INFO - Iter(train) [ 45650/160000] lr: 7.4172e-03 eta: 17:31:41 time: 0.5516 data_time: 0.0068 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0079 decode.acc_seg: 99.6734 aux.loss_ce: 0.0073 aux.acc_seg: 99.1992 +04/18 09:23:23 - mmengine - INFO - Iter(train) [ 45700/160000] lr: 7.4143e-03 eta: 17:31:13 time: 0.5521 data_time: 0.0059 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0078 decode.acc_seg: 99.6832 aux.loss_ce: 0.0076 aux.acc_seg: 99.1064 +04/18 09:23:51 - mmengine - INFO - Iter(train) [ 45750/160000] lr: 7.4114e-03 eta: 17:30:46 time: 0.5528 data_time: 0.0062 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0085 decode.acc_seg: 99.6417 aux.loss_ce: 0.0082 aux.acc_seg: 98.9037 +04/18 09:24:19 - mmengine - INFO - Iter(train) [ 45800/160000] lr: 7.4085e-03 eta: 17:30:18 time: 0.5516 data_time: 0.0063 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0086 decode.acc_seg: 99.7640 aux.loss_ce: 0.0077 aux.acc_seg: 99.3542 +04/18 09:24:46 - mmengine - INFO - Iter(train) [ 45850/160000] lr: 7.4056e-03 eta: 17:29:51 time: 0.5615 data_time: 0.0066 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0091 decode.acc_seg: 99.5231 aux.loss_ce: 0.0086 aux.acc_seg: 99.2666 +04/18 09:25:14 - mmengine - INFO - Iter(train) [ 45900/160000] lr: 7.4028e-03 eta: 17:29:23 time: 0.5515 data_time: 0.0064 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0084 decode.acc_seg: 99.6744 aux.loss_ce: 0.0079 aux.acc_seg: 99.1490 +04/18 09:25:42 - mmengine - INFO - Iter(train) [ 45950/160000] lr: 7.3999e-03 eta: 17:28:56 time: 0.5509 data_time: 0.0065 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0072 decode.acc_seg: 99.7808 aux.loss_ce: 0.0070 aux.acc_seg: 99.5138 +04/18 09:26:09 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 09:26:09 - mmengine - INFO - Iter(train) [ 46000/160000] lr: 7.3970e-03 eta: 17:28:28 time: 0.5508 data_time: 0.0061 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0076 decode.acc_seg: 99.7713 aux.loss_ce: 0.0076 aux.acc_seg: 99.3100 +04/18 09:26:37 - mmengine - INFO - Iter(train) [ 46050/160000] lr: 7.3941e-03 eta: 17:28:01 time: 0.5511 data_time: 0.0064 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0086 decode.acc_seg: 99.6272 aux.loss_ce: 0.0085 aux.acc_seg: 99.0365 +04/18 09:27:05 - mmengine - INFO - Iter(train) [ 46100/160000] lr: 7.3912e-03 eta: 17:27:33 time: 0.5517 data_time: 0.0063 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0081 decode.acc_seg: 99.7292 aux.loss_ce: 0.0076 aux.acc_seg: 99.2253 +04/18 09:27:32 - mmengine - INFO - Iter(train) [ 46150/160000] lr: 7.3884e-03 eta: 17:27:06 time: 0.5525 data_time: 0.0062 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0082 decode.acc_seg: 99.7085 aux.loss_ce: 0.0071 aux.acc_seg: 99.2450 +04/18 09:28:00 - mmengine - INFO - Iter(train) [ 46200/160000] lr: 7.3855e-03 eta: 17:26:38 time: 0.5533 data_time: 0.0063 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0090 decode.acc_seg: 99.7254 aux.loss_ce: 0.0084 aux.acc_seg: 99.3106 +04/18 09:28:27 - mmengine - INFO - Iter(train) [ 46250/160000] lr: 7.3826e-03 eta: 17:26:11 time: 0.5510 data_time: 0.0062 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0084 decode.acc_seg: 99.6780 aux.loss_ce: 0.0080 aux.acc_seg: 98.9218 +04/18 09:28:55 - mmengine - INFO - Iter(train) [ 46300/160000] lr: 7.3797e-03 eta: 17:25:43 time: 0.5525 data_time: 0.0059 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0095 decode.acc_seg: 99.6510 aux.loss_ce: 0.0083 aux.acc_seg: 99.0337 +04/18 09:29:23 - mmengine - INFO - Iter(train) [ 46350/160000] lr: 7.3768e-03 eta: 17:25:16 time: 0.5534 data_time: 0.0070 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0101 decode.acc_seg: 99.5838 aux.loss_ce: 0.0086 aux.acc_seg: 99.1583 +04/18 09:29:50 - mmengine - INFO - Iter(train) [ 46400/160000] lr: 7.3739e-03 eta: 17:24:48 time: 0.5523 data_time: 0.0058 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0086 decode.acc_seg: 99.6799 aux.loss_ce: 0.0078 aux.acc_seg: 99.2291 +04/18 09:30:18 - mmengine - INFO - Iter(train) [ 46450/160000] lr: 7.3711e-03 eta: 17:24:21 time: 0.5515 data_time: 0.0061 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0082 decode.acc_seg: 99.6817 aux.loss_ce: 0.0080 aux.acc_seg: 99.2268 +04/18 09:30:46 - mmengine - INFO - Iter(train) [ 46500/160000] lr: 7.3682e-03 eta: 17:23:53 time: 0.5517 data_time: 0.0066 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0094 decode.acc_seg: 99.3092 aux.loss_ce: 0.0085 aux.acc_seg: 98.7722 +04/18 09:31:13 - mmengine - INFO - Iter(train) [ 46550/160000] lr: 7.3653e-03 eta: 17:23:26 time: 0.5520 data_time: 0.0062 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0087 decode.acc_seg: 99.6584 aux.loss_ce: 0.0081 aux.acc_seg: 99.2076 +04/18 09:31:41 - mmengine - INFO - Iter(train) [ 46600/160000] lr: 7.3624e-03 eta: 17:22:58 time: 0.5515 data_time: 0.0065 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0090 decode.acc_seg: 99.6082 aux.loss_ce: 0.0083 aux.acc_seg: 98.9571 +04/18 09:32:08 - mmengine - INFO - Iter(train) [ 46650/160000] lr: 7.3595e-03 eta: 17:22:31 time: 0.5520 data_time: 0.0071 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0096 decode.acc_seg: 99.7278 aux.loss_ce: 0.0086 aux.acc_seg: 99.2349 +04/18 09:32:36 - mmengine - INFO - Iter(train) [ 46700/160000] lr: 7.3567e-03 eta: 17:22:03 time: 0.5520 data_time: 0.0067 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0087 decode.acc_seg: 99.6774 aux.loss_ce: 0.0080 aux.acc_seg: 99.1701 +04/18 09:33:04 - mmengine - INFO - Iter(train) [ 46750/160000] lr: 7.3538e-03 eta: 17:21:36 time: 0.5519 data_time: 0.0063 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0086 decode.acc_seg: 99.6366 aux.loss_ce: 0.0080 aux.acc_seg: 99.0657 +04/18 09:33:31 - mmengine - INFO - Iter(train) [ 46800/160000] lr: 7.3509e-03 eta: 17:21:08 time: 0.5513 data_time: 0.0063 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0079 decode.acc_seg: 99.6181 aux.loss_ce: 0.0075 aux.acc_seg: 99.1819 +04/18 09:33:59 - mmengine - INFO - Iter(train) [ 46850/160000] lr: 7.3480e-03 eta: 17:20:41 time: 0.5534 data_time: 0.0065 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0080 decode.acc_seg: 99.6229 aux.loss_ce: 0.0079 aux.acc_seg: 98.9841 +04/18 09:34:27 - mmengine - INFO - Iter(train) [ 46900/160000] lr: 7.3451e-03 eta: 17:20:13 time: 0.5522 data_time: 0.0069 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0074 decode.acc_seg: 99.7402 aux.loss_ce: 0.0069 aux.acc_seg: 99.2221 +04/18 09:34:54 - mmengine - INFO - Iter(train) [ 46950/160000] lr: 7.3422e-03 eta: 17:19:46 time: 0.5533 data_time: 0.0061 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0103 decode.acc_seg: 99.5117 aux.loss_ce: 0.0085 aux.acc_seg: 98.9362 +04/18 09:35:22 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 09:35:22 - mmengine - INFO - Iter(train) [ 47000/160000] lr: 7.3394e-03 eta: 17:19:18 time: 0.5526 data_time: 0.0064 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0094 decode.acc_seg: 99.6680 aux.loss_ce: 0.0085 aux.acc_seg: 99.1042 +04/18 09:35:50 - mmengine - INFO - Iter(train) [ 47050/160000] lr: 7.3365e-03 eta: 17:18:51 time: 0.5524 data_time: 0.0063 memory: 7635 loss: 0.0205 decode.loss_ce: 0.0115 decode.acc_seg: 99.3422 aux.loss_ce: 0.0090 aux.acc_seg: 98.6979 +04/18 09:36:17 - mmengine - INFO - Iter(train) [ 47100/160000] lr: 7.3336e-03 eta: 17:18:23 time: 0.5527 data_time: 0.0071 memory: 7635 loss: 0.0190 decode.loss_ce: 0.0103 decode.acc_seg: 99.6514 aux.loss_ce: 0.0087 aux.acc_seg: 99.1013 +04/18 09:36:45 - mmengine - INFO - Iter(train) [ 47150/160000] lr: 7.3307e-03 eta: 17:17:56 time: 0.5517 data_time: 0.0063 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0084 decode.acc_seg: 99.7099 aux.loss_ce: 0.0078 aux.acc_seg: 99.3170 +04/18 09:37:12 - mmengine - INFO - Iter(train) [ 47200/160000] lr: 7.3278e-03 eta: 17:17:28 time: 0.5493 data_time: 0.0062 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0094 decode.acc_seg: 99.6957 aux.loss_ce: 0.0086 aux.acc_seg: 99.2879 +04/18 09:37:40 - mmengine - INFO - Iter(train) [ 47250/160000] lr: 7.3249e-03 eta: 17:17:00 time: 0.5519 data_time: 0.0063 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0095 decode.acc_seg: 99.5607 aux.loss_ce: 0.0082 aux.acc_seg: 99.0628 +04/18 09:38:08 - mmengine - INFO - Iter(train) [ 47300/160000] lr: 7.3221e-03 eta: 17:16:33 time: 0.5520 data_time: 0.0059 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0091 decode.acc_seg: 99.6764 aux.loss_ce: 0.0082 aux.acc_seg: 99.3887 +04/18 09:38:35 - mmengine - INFO - Iter(train) [ 47350/160000] lr: 7.3192e-03 eta: 17:16:05 time: 0.5526 data_time: 0.0065 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0088 decode.acc_seg: 99.6683 aux.loss_ce: 0.0081 aux.acc_seg: 99.2015 +04/18 09:39:03 - mmengine - INFO - Iter(train) [ 47400/160000] lr: 7.3163e-03 eta: 17:15:38 time: 0.5517 data_time: 0.0067 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0078 decode.acc_seg: 99.6044 aux.loss_ce: 0.0075 aux.acc_seg: 99.1012 +04/18 09:39:30 - mmengine - INFO - Iter(train) [ 47450/160000] lr: 7.3134e-03 eta: 17:15:10 time: 0.5514 data_time: 0.0067 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0091 decode.acc_seg: 99.6548 aux.loss_ce: 0.0084 aux.acc_seg: 99.2237 +04/18 09:39:58 - mmengine - INFO - Iter(train) [ 47500/160000] lr: 7.3105e-03 eta: 17:14:42 time: 0.5513 data_time: 0.0066 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0096 decode.acc_seg: 99.6684 aux.loss_ce: 0.0085 aux.acc_seg: 99.2130 +04/18 09:40:26 - mmengine - INFO - Iter(train) [ 47550/160000] lr: 7.3076e-03 eta: 17:14:15 time: 0.5509 data_time: 0.0064 memory: 7635 loss: 0.0226 decode.loss_ce: 0.0126 decode.acc_seg: 99.5839 aux.loss_ce: 0.0100 aux.acc_seg: 98.9649 +04/18 09:40:53 - mmengine - INFO - Iter(train) [ 47600/160000] lr: 7.3048e-03 eta: 17:13:47 time: 0.5520 data_time: 0.0068 memory: 7635 loss: 0.0236 decode.loss_ce: 0.0134 decode.acc_seg: 99.2194 aux.loss_ce: 0.0102 aux.acc_seg: 98.6823 +04/18 09:41:21 - mmengine - INFO - Iter(train) [ 47650/160000] lr: 7.3019e-03 eta: 17:13:20 time: 0.5522 data_time: 0.0064 memory: 7635 loss: 0.0197 decode.loss_ce: 0.0108 decode.acc_seg: 99.6519 aux.loss_ce: 0.0089 aux.acc_seg: 99.2672 +04/18 09:41:48 - mmengine - INFO - Iter(train) [ 47700/160000] lr: 7.2990e-03 eta: 17:12:52 time: 0.5513 data_time: 0.0060 memory: 7635 loss: 0.0190 decode.loss_ce: 0.0104 decode.acc_seg: 99.4692 aux.loss_ce: 0.0086 aux.acc_seg: 99.1116 +04/18 09:42:16 - mmengine - INFO - Iter(train) [ 47750/160000] lr: 7.2961e-03 eta: 17:12:25 time: 0.5505 data_time: 0.0066 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0089 decode.acc_seg: 99.6764 aux.loss_ce: 0.0079 aux.acc_seg: 99.1522 +04/18 09:42:44 - mmengine - INFO - Iter(train) [ 47800/160000] lr: 7.2932e-03 eta: 17:11:57 time: 0.5531 data_time: 0.0067 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0089 decode.acc_seg: 99.6787 aux.loss_ce: 0.0080 aux.acc_seg: 99.2792 +04/18 09:43:11 - mmengine - INFO - Iter(train) [ 47850/160000] lr: 7.2903e-03 eta: 17:11:30 time: 0.5513 data_time: 0.0064 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0084 decode.acc_seg: 99.7006 aux.loss_ce: 0.0079 aux.acc_seg: 99.2769 +04/18 09:43:39 - mmengine - INFO - Iter(train) [ 47900/160000] lr: 7.2874e-03 eta: 17:11:02 time: 0.5525 data_time: 0.0063 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0089 decode.acc_seg: 99.6428 aux.loss_ce: 0.0083 aux.acc_seg: 98.9472 +04/18 09:44:07 - mmengine - INFO - Iter(train) [ 47950/160000] lr: 7.2846e-03 eta: 17:10:35 time: 0.5516 data_time: 0.0063 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0089 decode.acc_seg: 99.6430 aux.loss_ce: 0.0082 aux.acc_seg: 99.1258 +04/18 09:44:34 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 09:44:34 - mmengine - INFO - Iter(train) [ 48000/160000] lr: 7.2817e-03 eta: 17:10:07 time: 0.5605 data_time: 0.0063 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0092 decode.acc_seg: 99.5721 aux.loss_ce: 0.0083 aux.acc_seg: 99.0805 +04/18 09:45:02 - mmengine - INFO - Iter(train) [ 48050/160000] lr: 7.2788e-03 eta: 17:09:40 time: 0.5507 data_time: 0.0067 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0080 decode.acc_seg: 99.7004 aux.loss_ce: 0.0074 aux.acc_seg: 99.3549 +04/18 09:45:29 - mmengine - INFO - Iter(train) [ 48100/160000] lr: 7.2759e-03 eta: 17:09:12 time: 0.5531 data_time: 0.0066 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0096 decode.acc_seg: 99.6485 aux.loss_ce: 0.0086 aux.acc_seg: 99.1604 +04/18 09:45:57 - mmengine - INFO - Iter(train) [ 48150/160000] lr: 7.2730e-03 eta: 17:08:44 time: 0.5517 data_time: 0.0062 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0083 decode.acc_seg: 99.6278 aux.loss_ce: 0.0080 aux.acc_seg: 99.2423 +04/18 09:46:25 - mmengine - INFO - Iter(train) [ 48200/160000] lr: 7.2701e-03 eta: 17:08:17 time: 0.5525 data_time: 0.0063 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0088 decode.acc_seg: 99.5866 aux.loss_ce: 0.0079 aux.acc_seg: 99.2974 +04/18 09:46:52 - mmengine - INFO - Iter(train) [ 48250/160000] lr: 7.2672e-03 eta: 17:07:49 time: 0.5526 data_time: 0.0067 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0094 decode.acc_seg: 99.5598 aux.loss_ce: 0.0084 aux.acc_seg: 98.9615 +04/18 09:47:20 - mmengine - INFO - Iter(train) [ 48300/160000] lr: 7.2644e-03 eta: 17:07:22 time: 0.5520 data_time: 0.0065 memory: 7635 loss: 0.0219 decode.loss_ce: 0.0123 decode.acc_seg: 99.5892 aux.loss_ce: 0.0096 aux.acc_seg: 98.7660 +04/18 09:47:48 - mmengine - INFO - Iter(train) [ 48350/160000] lr: 7.2615e-03 eta: 17:06:54 time: 0.5517 data_time: 0.0065 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0085 decode.acc_seg: 99.6997 aux.loss_ce: 0.0076 aux.acc_seg: 99.2211 +04/18 09:48:15 - mmengine - INFO - Iter(train) [ 48400/160000] lr: 7.2586e-03 eta: 17:06:27 time: 0.5529 data_time: 0.0065 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0094 decode.acc_seg: 99.6227 aux.loss_ce: 0.0078 aux.acc_seg: 99.1485 +04/18 09:48:43 - mmengine - INFO - Iter(train) [ 48450/160000] lr: 7.2557e-03 eta: 17:05:59 time: 0.5516 data_time: 0.0062 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0088 decode.acc_seg: 99.6257 aux.loss_ce: 0.0076 aux.acc_seg: 99.0685 +04/18 09:49:10 - mmengine - INFO - Iter(train) [ 48500/160000] lr: 7.2528e-03 eta: 17:05:31 time: 0.5507 data_time: 0.0060 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0093 decode.acc_seg: 99.7208 aux.loss_ce: 0.0079 aux.acc_seg: 99.2226 +04/18 09:49:38 - mmengine - INFO - Iter(train) [ 48550/160000] lr: 7.2499e-03 eta: 17:05:04 time: 0.5540 data_time: 0.0076 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0088 decode.acc_seg: 99.3439 aux.loss_ce: 0.0077 aux.acc_seg: 98.9624 +04/18 09:50:06 - mmengine - INFO - Iter(train) [ 48600/160000] lr: 7.2470e-03 eta: 17:04:36 time: 0.5529 data_time: 0.0067 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0083 decode.acc_seg: 99.7175 aux.loss_ce: 0.0078 aux.acc_seg: 99.3053 +04/18 09:50:33 - mmengine - INFO - Iter(train) [ 48650/160000] lr: 7.2442e-03 eta: 17:04:09 time: 0.5520 data_time: 0.0060 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0097 decode.acc_seg: 99.3075 aux.loss_ce: 0.0088 aux.acc_seg: 98.7808 +04/18 09:51:01 - mmengine - INFO - Iter(train) [ 48700/160000] lr: 7.2413e-03 eta: 17:03:41 time: 0.5515 data_time: 0.0069 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0082 decode.acc_seg: 99.7275 aux.loss_ce: 0.0077 aux.acc_seg: 99.4738 +04/18 09:51:28 - mmengine - INFO - Iter(train) [ 48750/160000] lr: 7.2384e-03 eta: 17:03:14 time: 0.5525 data_time: 0.0068 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0082 decode.acc_seg: 99.5996 aux.loss_ce: 0.0078 aux.acc_seg: 99.1944 +04/18 09:51:56 - mmengine - INFO - Iter(train) [ 48800/160000] lr: 7.2355e-03 eta: 17:02:46 time: 0.5519 data_time: 0.0060 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0087 decode.acc_seg: 99.7039 aux.loss_ce: 0.0080 aux.acc_seg: 99.2879 +04/18 09:52:24 - mmengine - INFO - Iter(train) [ 48850/160000] lr: 7.2326e-03 eta: 17:02:19 time: 0.5526 data_time: 0.0064 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0090 decode.acc_seg: 99.6964 aux.loss_ce: 0.0085 aux.acc_seg: 99.1631 +04/18 09:52:51 - mmengine - INFO - Iter(train) [ 48900/160000] lr: 7.2297e-03 eta: 17:01:51 time: 0.5520 data_time: 0.0063 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0089 decode.acc_seg: 99.6101 aux.loss_ce: 0.0082 aux.acc_seg: 99.1882 +04/18 09:53:19 - mmengine - INFO - Iter(train) [ 48950/160000] lr: 7.2268e-03 eta: 17:01:24 time: 0.5513 data_time: 0.0065 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0084 decode.acc_seg: 99.7277 aux.loss_ce: 0.0077 aux.acc_seg: 99.3268 +04/18 09:53:47 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 09:53:47 - mmengine - INFO - Iter(train) [ 49000/160000] lr: 7.2239e-03 eta: 17:00:56 time: 0.5522 data_time: 0.0072 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0099 decode.acc_seg: 99.6124 aux.loss_ce: 0.0084 aux.acc_seg: 99.0455 +04/18 09:54:14 - mmengine - INFO - Iter(train) [ 49050/160000] lr: 7.2211e-03 eta: 17:00:29 time: 0.5514 data_time: 0.0068 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0088 decode.acc_seg: 99.6682 aux.loss_ce: 0.0085 aux.acc_seg: 99.1941 +04/18 09:54:42 - mmengine - INFO - Iter(train) [ 49100/160000] lr: 7.2182e-03 eta: 17:00:01 time: 0.5519 data_time: 0.0071 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0079 decode.acc_seg: 99.6275 aux.loss_ce: 0.0077 aux.acc_seg: 99.1561 +04/18 09:55:10 - mmengine - INFO - Iter(train) [ 49150/160000] lr: 7.2153e-03 eta: 16:59:34 time: 0.5519 data_time: 0.0065 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0089 decode.acc_seg: 99.5423 aux.loss_ce: 0.0080 aux.acc_seg: 99.0732 +04/18 09:55:37 - mmengine - INFO - Iter(train) [ 49200/160000] lr: 7.2124e-03 eta: 16:59:06 time: 0.5534 data_time: 0.0067 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0080 decode.acc_seg: 99.7293 aux.loss_ce: 0.0073 aux.acc_seg: 99.3897 +04/18 09:56:05 - mmengine - INFO - Iter(train) [ 49250/160000] lr: 7.2095e-03 eta: 16:58:39 time: 0.5522 data_time: 0.0060 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0083 decode.acc_seg: 99.5699 aux.loss_ce: 0.0076 aux.acc_seg: 99.1997 +04/18 09:56:32 - mmengine - INFO - Iter(train) [ 49300/160000] lr: 7.2066e-03 eta: 16:58:11 time: 0.5515 data_time: 0.0065 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0078 decode.acc_seg: 99.5998 aux.loss_ce: 0.0077 aux.acc_seg: 99.1120 +04/18 09:57:00 - mmengine - INFO - Iter(train) [ 49350/160000] lr: 7.2037e-03 eta: 16:57:43 time: 0.5515 data_time: 0.0065 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0078 decode.acc_seg: 99.6954 aux.loss_ce: 0.0076 aux.acc_seg: 99.2864 +04/18 09:57:27 - mmengine - INFO - Iter(train) [ 49400/160000] lr: 7.2008e-03 eta: 16:57:16 time: 0.5518 data_time: 0.0063 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0076 decode.acc_seg: 99.6588 aux.loss_ce: 0.0076 aux.acc_seg: 99.0170 +04/18 09:57:55 - mmengine - INFO - Iter(train) [ 49450/160000] lr: 7.1979e-03 eta: 16:56:48 time: 0.5498 data_time: 0.0057 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0079 decode.acc_seg: 99.7878 aux.loss_ce: 0.0080 aux.acc_seg: 99.3705 +04/18 09:58:23 - mmengine - INFO - Iter(train) [ 49500/160000] lr: 7.1951e-03 eta: 16:56:20 time: 0.5513 data_time: 0.0066 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0091 decode.acc_seg: 99.6921 aux.loss_ce: 0.0077 aux.acc_seg: 99.1593 +04/18 09:58:50 - mmengine - INFO - Iter(train) [ 49550/160000] lr: 7.1922e-03 eta: 16:55:53 time: 0.5501 data_time: 0.0060 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0090 decode.acc_seg: 99.6333 aux.loss_ce: 0.0084 aux.acc_seg: 99.1590 +04/18 09:59:18 - mmengine - INFO - Iter(train) [ 49600/160000] lr: 7.1893e-03 eta: 16:55:25 time: 0.5501 data_time: 0.0060 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0090 decode.acc_seg: 99.5110 aux.loss_ce: 0.0079 aux.acc_seg: 98.7459 +04/18 09:59:45 - mmengine - INFO - Iter(train) [ 49650/160000] lr: 7.1864e-03 eta: 16:54:58 time: 0.5508 data_time: 0.0065 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0091 decode.acc_seg: 99.6520 aux.loss_ce: 0.0082 aux.acc_seg: 99.2351 +04/18 10:00:13 - mmengine - INFO - Iter(train) [ 49700/160000] lr: 7.1835e-03 eta: 16:54:30 time: 0.5505 data_time: 0.0065 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0075 decode.acc_seg: 99.6666 aux.loss_ce: 0.0073 aux.acc_seg: 99.3447 +04/18 10:00:41 - mmengine - INFO - Iter(train) [ 49750/160000] lr: 7.1806e-03 eta: 16:54:02 time: 0.5512 data_time: 0.0069 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0073 decode.acc_seg: 99.6932 aux.loss_ce: 0.0072 aux.acc_seg: 99.3723 +04/18 10:01:08 - mmengine - INFO - Iter(train) [ 49800/160000] lr: 7.1777e-03 eta: 16:53:35 time: 0.5522 data_time: 0.0077 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0092 decode.acc_seg: 99.7171 aux.loss_ce: 0.0083 aux.acc_seg: 99.3166 +04/18 10:01:36 - mmengine - INFO - Iter(train) [ 49850/160000] lr: 7.1748e-03 eta: 16:53:07 time: 0.5522 data_time: 0.0061 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0089 decode.acc_seg: 99.7813 aux.loss_ce: 0.0082 aux.acc_seg: 99.4865 +04/18 10:02:04 - mmengine - INFO - Iter(train) [ 49900/160000] lr: 7.1719e-03 eta: 16:52:40 time: 0.5525 data_time: 0.0061 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0073 decode.acc_seg: 99.7205 aux.loss_ce: 0.0077 aux.acc_seg: 99.1055 +04/18 10:02:31 - mmengine - INFO - Iter(train) [ 49950/160000] lr: 7.1690e-03 eta: 16:52:12 time: 0.5514 data_time: 0.0073 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0081 decode.acc_seg: 99.7235 aux.loss_ce: 0.0074 aux.acc_seg: 99.3393 +04/18 10:02:59 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 10:02:59 - mmengine - INFO - Iter(train) [ 50000/160000] lr: 7.1662e-03 eta: 16:51:45 time: 0.5518 data_time: 0.0062 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0086 decode.acc_seg: 99.6929 aux.loss_ce: 0.0083 aux.acc_seg: 99.2558 +04/18 10:02:59 - mmengine - INFO - Saving checkpoint at 50000 iterations +04/18 10:03:03 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:07 time: 0.0466 data_time: 0.0015 memory: 1657 +04/18 10:03:05 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:04 time: 0.0468 data_time: 0.0013 memory: 1657 +04/18 10:03:07 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:02 time: 0.0470 data_time: 0.0016 memory: 1657 +04/18 10:03:10 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.0460 data_time: 0.0013 memory: 1657 +04/18 10:03:10 - mmengine - INFO - per class results: +04/18 10:03:10 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 99.06 | 99.46 | 99.53 | 99.6 | 99.46 | +| contrast | 79.91 | 90.32 | 88.83 | 87.39 | 90.32 | ++------------+-------+-------+--------+-----------+--------+ +04/18 10:03:10 - mmengine - INFO - Iter(val) [200/200] aAcc: 99.0900 mIoU: 89.4800 mAcc: 94.8900 mFscore: 94.1800 mPrecision: 93.4900 mRecall: 94.8900 data_time: 0.0015 time: 0.0467 +04/18 10:03:37 - mmengine - INFO - Iter(train) [ 50050/160000] lr: 7.1633e-03 eta: 16:51:17 time: 0.5494 data_time: 0.0061 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0084 decode.acc_seg: 99.6303 aux.loss_ce: 0.0078 aux.acc_seg: 99.1614 +04/18 10:04:05 - mmengine - INFO - Iter(train) [ 50100/160000] lr: 7.1604e-03 eta: 16:50:50 time: 0.5520 data_time: 0.0065 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0081 decode.acc_seg: 99.7437 aux.loss_ce: 0.0078 aux.acc_seg: 99.2997 +04/18 10:04:33 - mmengine - INFO - Iter(train) [ 50150/160000] lr: 7.1575e-03 eta: 16:50:22 time: 0.5524 data_time: 0.0063 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0077 decode.acc_seg: 99.7126 aux.loss_ce: 0.0075 aux.acc_seg: 99.1648 +04/18 10:05:00 - mmengine - INFO - Iter(train) [ 50200/160000] lr: 7.1546e-03 eta: 16:49:55 time: 0.5508 data_time: 0.0068 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0087 decode.acc_seg: 99.7084 aux.loss_ce: 0.0076 aux.acc_seg: 99.4479 +04/18 10:05:28 - mmengine - INFO - Iter(train) [ 50250/160000] lr: 7.1517e-03 eta: 16:49:27 time: 0.5500 data_time: 0.0060 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0089 decode.acc_seg: 99.6487 aux.loss_ce: 0.0082 aux.acc_seg: 99.1045 +04/18 10:05:55 - mmengine - INFO - Iter(train) [ 50300/160000] lr: 7.1488e-03 eta: 16:48:59 time: 0.5500 data_time: 0.0063 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0088 decode.acc_seg: 99.6172 aux.loss_ce: 0.0081 aux.acc_seg: 98.9217 +04/18 10:06:23 - mmengine - INFO - Iter(train) [ 50350/160000] lr: 7.1459e-03 eta: 16:48:32 time: 0.5518 data_time: 0.0066 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0084 decode.acc_seg: 99.4938 aux.loss_ce: 0.0076 aux.acc_seg: 99.0082 +04/18 10:06:51 - mmengine - INFO - Iter(train) [ 50400/160000] lr: 7.1430e-03 eta: 16:48:04 time: 0.5517 data_time: 0.0069 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0091 decode.acc_seg: 99.6501 aux.loss_ce: 0.0088 aux.acc_seg: 99.2349 +04/18 10:07:18 - mmengine - INFO - Iter(train) [ 50450/160000] lr: 7.1401e-03 eta: 16:47:37 time: 0.5510 data_time: 0.0062 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0093 decode.acc_seg: 99.5913 aux.loss_ce: 0.0083 aux.acc_seg: 99.2395 +04/18 10:07:46 - mmengine - INFO - Iter(train) [ 50500/160000] lr: 7.1372e-03 eta: 16:47:09 time: 0.5531 data_time: 0.0066 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0091 decode.acc_seg: 99.7045 aux.loss_ce: 0.0085 aux.acc_seg: 99.1563 +04/18 10:08:14 - mmengine - INFO - Iter(train) [ 50550/160000] lr: 7.1343e-03 eta: 16:46:42 time: 0.5514 data_time: 0.0063 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0103 decode.acc_seg: 99.6216 aux.loss_ce: 0.0083 aux.acc_seg: 99.1870 +04/18 10:08:41 - mmengine - INFO - Iter(train) [ 50600/160000] lr: 7.1315e-03 eta: 16:46:14 time: 0.5515 data_time: 0.0061 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0085 decode.acc_seg: 99.6038 aux.loss_ce: 0.0079 aux.acc_seg: 99.2425 +04/18 10:09:09 - mmengine - INFO - Iter(train) [ 50650/160000] lr: 7.1286e-03 eta: 16:45:46 time: 0.5527 data_time: 0.0073 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0091 decode.acc_seg: 99.6921 aux.loss_ce: 0.0085 aux.acc_seg: 99.2167 +04/18 10:09:36 - mmengine - INFO - Iter(train) [ 50700/160000] lr: 7.1257e-03 eta: 16:45:19 time: 0.5527 data_time: 0.0067 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0098 decode.acc_seg: 99.6984 aux.loss_ce: 0.0082 aux.acc_seg: 99.3204 +04/18 10:10:04 - mmengine - INFO - Iter(train) [ 50750/160000] lr: 7.1228e-03 eta: 16:44:51 time: 0.5508 data_time: 0.0060 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0074 decode.acc_seg: 99.7176 aux.loss_ce: 0.0072 aux.acc_seg: 99.1421 +04/18 10:10:31 - mmengine - INFO - Iter(train) [ 50800/160000] lr: 7.1199e-03 eta: 16:44:23 time: 0.5517 data_time: 0.0061 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0094 decode.acc_seg: 99.6450 aux.loss_ce: 0.0081 aux.acc_seg: 99.1481 +04/18 10:10:59 - mmengine - INFO - Iter(train) [ 50850/160000] lr: 7.1170e-03 eta: 16:43:56 time: 0.5520 data_time: 0.0070 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0092 decode.acc_seg: 99.6152 aux.loss_ce: 0.0084 aux.acc_seg: 99.1262 +04/18 10:11:27 - mmengine - INFO - Iter(train) [ 50900/160000] lr: 7.1141e-03 eta: 16:43:28 time: 0.5511 data_time: 0.0063 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0086 decode.acc_seg: 99.7204 aux.loss_ce: 0.0077 aux.acc_seg: 99.4347 +04/18 10:11:54 - mmengine - INFO - Iter(train) [ 50950/160000] lr: 7.1112e-03 eta: 16:43:01 time: 0.5504 data_time: 0.0060 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0082 decode.acc_seg: 99.7308 aux.loss_ce: 0.0077 aux.acc_seg: 99.2590 +04/18 10:12:22 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 10:12:22 - mmengine - INFO - Iter(train) [ 51000/160000] lr: 7.1083e-03 eta: 16:42:33 time: 0.5508 data_time: 0.0061 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0084 decode.acc_seg: 99.7273 aux.loss_ce: 0.0080 aux.acc_seg: 99.3155 +04/18 10:12:49 - mmengine - INFO - Iter(train) [ 51050/160000] lr: 7.1054e-03 eta: 16:42:06 time: 0.5505 data_time: 0.0065 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0088 decode.acc_seg: 99.6755 aux.loss_ce: 0.0080 aux.acc_seg: 99.0463 +04/18 10:13:17 - mmengine - INFO - Iter(train) [ 51100/160000] lr: 7.1025e-03 eta: 16:41:38 time: 0.5522 data_time: 0.0068 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0080 decode.acc_seg: 99.6352 aux.loss_ce: 0.0081 aux.acc_seg: 99.0726 +04/18 10:13:45 - mmengine - INFO - Iter(train) [ 51150/160000] lr: 7.0996e-03 eta: 16:41:10 time: 0.5523 data_time: 0.0060 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0074 decode.acc_seg: 99.6531 aux.loss_ce: 0.0072 aux.acc_seg: 99.2215 +04/18 10:14:12 - mmengine - INFO - Iter(train) [ 51200/160000] lr: 7.0967e-03 eta: 16:40:43 time: 0.5516 data_time: 0.0063 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0091 decode.acc_seg: 99.5941 aux.loss_ce: 0.0082 aux.acc_seg: 98.9841 +04/18 10:14:40 - mmengine - INFO - Iter(train) [ 51250/160000] lr: 7.0938e-03 eta: 16:40:15 time: 0.5519 data_time: 0.0072 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0083 decode.acc_seg: 99.7270 aux.loss_ce: 0.0075 aux.acc_seg: 99.2428 +04/18 10:15:07 - mmengine - INFO - Iter(train) [ 51300/160000] lr: 7.0910e-03 eta: 16:39:48 time: 0.5517 data_time: 0.0064 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0089 decode.acc_seg: 99.4575 aux.loss_ce: 0.0082 aux.acc_seg: 99.0960 +04/18 10:15:35 - mmengine - INFO - Iter(train) [ 51350/160000] lr: 7.0881e-03 eta: 16:39:20 time: 0.5506 data_time: 0.0065 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0076 decode.acc_seg: 99.7530 aux.loss_ce: 0.0074 aux.acc_seg: 99.3126 +04/18 10:16:03 - mmengine - INFO - Iter(train) [ 51400/160000] lr: 7.0852e-03 eta: 16:38:52 time: 0.5527 data_time: 0.0061 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0084 decode.acc_seg: 99.7535 aux.loss_ce: 0.0078 aux.acc_seg: 99.3334 +04/18 10:16:30 - mmengine - INFO - Iter(train) [ 51450/160000] lr: 7.0823e-03 eta: 16:38:25 time: 0.5519 data_time: 0.0064 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0098 decode.acc_seg: 99.6828 aux.loss_ce: 0.0083 aux.acc_seg: 99.2489 +04/18 10:16:58 - mmengine - INFO - Iter(train) [ 51500/160000] lr: 7.0794e-03 eta: 16:37:57 time: 0.5517 data_time: 0.0065 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0087 decode.acc_seg: 99.6384 aux.loss_ce: 0.0081 aux.acc_seg: 99.1660 +04/18 10:17:25 - mmengine - INFO - Iter(train) [ 51550/160000] lr: 7.0765e-03 eta: 16:37:30 time: 0.5522 data_time: 0.0066 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0081 decode.acc_seg: 99.7622 aux.loss_ce: 0.0075 aux.acc_seg: 99.3407 +04/18 10:17:53 - mmengine - INFO - Iter(train) [ 51600/160000] lr: 7.0736e-03 eta: 16:37:02 time: 0.5505 data_time: 0.0065 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0079 decode.acc_seg: 99.7211 aux.loss_ce: 0.0079 aux.acc_seg: 99.2453 +04/18 10:18:21 - mmengine - INFO - Iter(train) [ 51650/160000] lr: 7.0707e-03 eta: 16:36:34 time: 0.5527 data_time: 0.0068 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0086 decode.acc_seg: 99.4807 aux.loss_ce: 0.0081 aux.acc_seg: 99.0744 +04/18 10:18:48 - mmengine - INFO - Iter(train) [ 51700/160000] lr: 7.0678e-03 eta: 16:36:07 time: 0.5525 data_time: 0.0075 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0083 decode.acc_seg: 99.7381 aux.loss_ce: 0.0081 aux.acc_seg: 99.2805 +04/18 10:19:16 - mmengine - INFO - Iter(train) [ 51750/160000] lr: 7.0649e-03 eta: 16:35:39 time: 0.5515 data_time: 0.0064 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0087 decode.acc_seg: 99.6535 aux.loss_ce: 0.0082 aux.acc_seg: 99.1571 +04/18 10:19:43 - mmengine - INFO - Iter(train) [ 51800/160000] lr: 7.0620e-03 eta: 16:35:12 time: 0.5509 data_time: 0.0066 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0075 decode.acc_seg: 99.7637 aux.loss_ce: 0.0073 aux.acc_seg: 99.2836 +04/18 10:20:11 - mmengine - INFO - Iter(train) [ 51850/160000] lr: 7.0591e-03 eta: 16:34:44 time: 0.5509 data_time: 0.0064 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0098 decode.acc_seg: 99.6417 aux.loss_ce: 0.0085 aux.acc_seg: 99.1693 +04/18 10:20:39 - mmengine - INFO - Iter(train) [ 51900/160000] lr: 7.0562e-03 eta: 16:34:16 time: 0.5524 data_time: 0.0062 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0091 decode.acc_seg: 99.6486 aux.loss_ce: 0.0082 aux.acc_seg: 99.2789 +04/18 10:21:06 - mmengine - INFO - Iter(train) [ 51950/160000] lr: 7.0533e-03 eta: 16:33:49 time: 0.5602 data_time: 0.0062 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0092 decode.acc_seg: 99.6376 aux.loss_ce: 0.0082 aux.acc_seg: 99.2391 +04/18 10:21:34 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 10:21:34 - mmengine - INFO - Iter(train) [ 52000/160000] lr: 7.0504e-03 eta: 16:33:21 time: 0.5521 data_time: 0.0064 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0085 decode.acc_seg: 99.6520 aux.loss_ce: 0.0078 aux.acc_seg: 99.3771 +04/18 10:22:02 - mmengine - INFO - Iter(train) [ 52050/160000] lr: 7.0475e-03 eta: 16:32:54 time: 0.5515 data_time: 0.0070 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0076 decode.acc_seg: 99.6364 aux.loss_ce: 0.0075 aux.acc_seg: 99.0461 +04/18 10:22:29 - mmengine - INFO - Iter(train) [ 52100/160000] lr: 7.0446e-03 eta: 16:32:26 time: 0.5516 data_time: 0.0070 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0095 decode.acc_seg: 99.6502 aux.loss_ce: 0.0087 aux.acc_seg: 99.2046 +04/18 10:22:57 - mmengine - INFO - Iter(train) [ 52150/160000] lr: 7.0417e-03 eta: 16:31:59 time: 0.5526 data_time: 0.0069 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0087 decode.acc_seg: 99.6889 aux.loss_ce: 0.0079 aux.acc_seg: 99.3339 +04/18 10:23:24 - mmengine - INFO - Iter(train) [ 52200/160000] lr: 7.0388e-03 eta: 16:31:31 time: 0.5527 data_time: 0.0064 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0075 decode.acc_seg: 99.6093 aux.loss_ce: 0.0070 aux.acc_seg: 99.0957 +04/18 10:23:52 - mmengine - INFO - Iter(train) [ 52250/160000] lr: 7.0359e-03 eta: 16:31:03 time: 0.5514 data_time: 0.0064 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0092 decode.acc_seg: 99.6466 aux.loss_ce: 0.0088 aux.acc_seg: 98.8050 +04/18 10:24:19 - mmengine - INFO - Iter(train) [ 52300/160000] lr: 7.0330e-03 eta: 16:30:36 time: 0.5519 data_time: 0.0065 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0086 decode.acc_seg: 99.6104 aux.loss_ce: 0.0081 aux.acc_seg: 99.2305 +04/18 10:24:47 - mmengine - INFO - Iter(train) [ 52350/160000] lr: 7.0301e-03 eta: 16:30:08 time: 0.5518 data_time: 0.0066 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0097 decode.acc_seg: 99.7419 aux.loss_ce: 0.0085 aux.acc_seg: 99.3154 +04/18 10:25:15 - mmengine - INFO - Iter(train) [ 52400/160000] lr: 7.0272e-03 eta: 16:29:41 time: 0.5508 data_time: 0.0066 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0088 decode.acc_seg: 99.7513 aux.loss_ce: 0.0086 aux.acc_seg: 99.3443 +04/18 10:25:42 - mmengine - INFO - Iter(train) [ 52450/160000] lr: 7.0244e-03 eta: 16:29:13 time: 0.5508 data_time: 0.0060 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0075 decode.acc_seg: 99.6203 aux.loss_ce: 0.0074 aux.acc_seg: 99.1570 +04/18 10:26:10 - mmengine - INFO - Iter(train) [ 52500/160000] lr: 7.0215e-03 eta: 16:28:45 time: 0.5515 data_time: 0.0059 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0080 decode.acc_seg: 99.7168 aux.loss_ce: 0.0074 aux.acc_seg: 99.2850 +04/18 10:26:37 - mmengine - INFO - Iter(train) [ 52550/160000] lr: 7.0186e-03 eta: 16:28:18 time: 0.5516 data_time: 0.0064 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0075 decode.acc_seg: 99.6927 aux.loss_ce: 0.0074 aux.acc_seg: 99.2032 +04/18 10:27:05 - mmengine - INFO - Iter(train) [ 52600/160000] lr: 7.0157e-03 eta: 16:27:50 time: 0.5519 data_time: 0.0065 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0089 decode.acc_seg: 99.6718 aux.loss_ce: 0.0081 aux.acc_seg: 99.2196 +04/18 10:27:33 - mmengine - INFO - Iter(train) [ 52650/160000] lr: 7.0128e-03 eta: 16:27:23 time: 0.5524 data_time: 0.0063 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0076 decode.acc_seg: 99.7131 aux.loss_ce: 0.0072 aux.acc_seg: 99.2583 +04/18 10:28:00 - mmengine - INFO - Iter(train) [ 52700/160000] lr: 7.0099e-03 eta: 16:26:55 time: 0.5527 data_time: 0.0070 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0102 decode.acc_seg: 99.6510 aux.loss_ce: 0.0081 aux.acc_seg: 99.2446 +04/18 10:28:28 - mmengine - INFO - Iter(train) [ 52750/160000] lr: 7.0070e-03 eta: 16:26:27 time: 0.5522 data_time: 0.0068 memory: 7635 loss: 0.0210 decode.loss_ce: 0.0114 decode.acc_seg: 99.5508 aux.loss_ce: 0.0095 aux.acc_seg: 99.0664 +04/18 10:28:55 - mmengine - INFO - Iter(train) [ 52800/160000] lr: 7.0041e-03 eta: 16:26:00 time: 0.5517 data_time: 0.0071 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0091 decode.acc_seg: 99.5591 aux.loss_ce: 0.0077 aux.acc_seg: 98.9380 +04/18 10:29:23 - mmengine - INFO - Iter(train) [ 52850/160000] lr: 7.0012e-03 eta: 16:25:32 time: 0.5509 data_time: 0.0064 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0092 decode.acc_seg: 99.6599 aux.loss_ce: 0.0083 aux.acc_seg: 99.0653 +04/18 10:29:51 - mmengine - INFO - Iter(train) [ 52900/160000] lr: 6.9983e-03 eta: 16:25:05 time: 0.5520 data_time: 0.0062 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0097 decode.acc_seg: 99.7003 aux.loss_ce: 0.0088 aux.acc_seg: 99.2614 +04/18 10:30:18 - mmengine - INFO - Iter(train) [ 52950/160000] lr: 6.9954e-03 eta: 16:24:37 time: 0.5528 data_time: 0.0063 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0096 decode.acc_seg: 99.5527 aux.loss_ce: 0.0083 aux.acc_seg: 99.1763 +04/18 10:30:46 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 10:30:46 - mmengine - INFO - Iter(train) [ 53000/160000] lr: 6.9925e-03 eta: 16:24:09 time: 0.5511 data_time: 0.0055 memory: 7635 loss: 0.0190 decode.loss_ce: 0.0100 decode.acc_seg: 99.7108 aux.loss_ce: 0.0090 aux.acc_seg: 99.2771 +04/18 10:31:14 - mmengine - INFO - Iter(train) [ 53050/160000] lr: 6.9896e-03 eta: 16:23:42 time: 0.5529 data_time: 0.0059 memory: 7635 loss: 0.0186 decode.loss_ce: 0.0101 decode.acc_seg: 99.6829 aux.loss_ce: 0.0085 aux.acc_seg: 99.2652 +04/18 10:31:41 - mmengine - INFO - Iter(train) [ 53100/160000] lr: 6.9867e-03 eta: 16:23:15 time: 0.5501 data_time: 0.0058 memory: 7635 loss: 0.0186 decode.loss_ce: 0.0098 decode.acc_seg: 99.5369 aux.loss_ce: 0.0088 aux.acc_seg: 98.8965 +04/18 10:32:09 - mmengine - INFO - Iter(train) [ 53150/160000] lr: 6.9838e-03 eta: 16:22:47 time: 0.5526 data_time: 0.0064 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0093 decode.acc_seg: 99.6242 aux.loss_ce: 0.0081 aux.acc_seg: 99.2612 +04/18 10:32:36 - mmengine - INFO - Iter(train) [ 53200/160000] lr: 6.9809e-03 eta: 16:22:19 time: 0.5519 data_time: 0.0068 memory: 7635 loss: 0.0189 decode.loss_ce: 0.0103 decode.acc_seg: 99.5462 aux.loss_ce: 0.0086 aux.acc_seg: 98.9160 +04/18 10:33:04 - mmengine - INFO - Iter(train) [ 53250/160000] lr: 6.9780e-03 eta: 16:21:52 time: 0.5522 data_time: 0.0064 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0099 decode.acc_seg: 99.6872 aux.loss_ce: 0.0086 aux.acc_seg: 99.2883 +04/18 10:33:32 - mmengine - INFO - Iter(train) [ 53300/160000] lr: 6.9751e-03 eta: 16:21:24 time: 0.5513 data_time: 0.0061 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0095 decode.acc_seg: 99.6056 aux.loss_ce: 0.0082 aux.acc_seg: 99.2646 +04/18 10:33:59 - mmengine - INFO - Iter(train) [ 53350/160000] lr: 6.9722e-03 eta: 16:20:57 time: 0.5496 data_time: 0.0068 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0080 decode.acc_seg: 99.7437 aux.loss_ce: 0.0078 aux.acc_seg: 99.2231 +04/18 10:34:27 - mmengine - INFO - Iter(train) [ 53400/160000] lr: 6.9693e-03 eta: 16:20:29 time: 0.5525 data_time: 0.0071 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0100 decode.acc_seg: 99.6415 aux.loss_ce: 0.0082 aux.acc_seg: 99.1047 +04/18 10:34:54 - mmengine - INFO - Iter(train) [ 53450/160000] lr: 6.9664e-03 eta: 16:20:01 time: 0.5512 data_time: 0.0071 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0083 decode.acc_seg: 99.6616 aux.loss_ce: 0.0076 aux.acc_seg: 99.1767 +04/18 10:35:22 - mmengine - INFO - Iter(train) [ 53500/160000] lr: 6.9635e-03 eta: 16:19:34 time: 0.5510 data_time: 0.0063 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0079 decode.acc_seg: 99.6323 aux.loss_ce: 0.0079 aux.acc_seg: 99.0246 +04/18 10:35:50 - mmengine - INFO - Iter(train) [ 53550/160000] lr: 6.9606e-03 eta: 16:19:06 time: 0.5522 data_time: 0.0071 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0094 decode.acc_seg: 99.6636 aux.loss_ce: 0.0084 aux.acc_seg: 99.0689 +04/18 10:36:17 - mmengine - INFO - Iter(train) [ 53600/160000] lr: 6.9577e-03 eta: 16:18:39 time: 0.5532 data_time: 0.0071 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0084 decode.acc_seg: 99.6490 aux.loss_ce: 0.0080 aux.acc_seg: 99.1546 +04/18 10:36:45 - mmengine - INFO - Iter(train) [ 53650/160000] lr: 6.9548e-03 eta: 16:18:11 time: 0.5524 data_time: 0.0064 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0089 decode.acc_seg: 99.7125 aux.loss_ce: 0.0083 aux.acc_seg: 99.2896 +04/18 10:37:12 - mmengine - INFO - Iter(train) [ 53700/160000] lr: 6.9519e-03 eta: 16:17:44 time: 0.5513 data_time: 0.0063 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0093 decode.acc_seg: 99.7014 aux.loss_ce: 0.0083 aux.acc_seg: 99.2769 +04/18 10:37:40 - mmengine - INFO - Iter(train) [ 53750/160000] lr: 6.9490e-03 eta: 16:17:16 time: 0.5508 data_time: 0.0071 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0083 decode.acc_seg: 99.6697 aux.loss_ce: 0.0080 aux.acc_seg: 99.1458 +04/18 10:38:08 - mmengine - INFO - Iter(train) [ 53800/160000] lr: 6.9461e-03 eta: 16:16:48 time: 0.5534 data_time: 0.0066 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0078 decode.acc_seg: 99.7752 aux.loss_ce: 0.0074 aux.acc_seg: 99.4241 +04/18 10:38:35 - mmengine - INFO - Iter(train) [ 53850/160000] lr: 6.9432e-03 eta: 16:16:21 time: 0.5525 data_time: 0.0059 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0083 decode.acc_seg: 99.6603 aux.loss_ce: 0.0076 aux.acc_seg: 99.3434 +04/18 10:39:03 - mmengine - INFO - Iter(train) [ 53900/160000] lr: 6.9403e-03 eta: 16:15:53 time: 0.5512 data_time: 0.0063 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0086 decode.acc_seg: 99.6707 aux.loss_ce: 0.0079 aux.acc_seg: 99.0522 +04/18 10:39:30 - mmengine - INFO - Iter(train) [ 53950/160000] lr: 6.9374e-03 eta: 16:15:26 time: 0.5522 data_time: 0.0065 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0082 decode.acc_seg: 99.6368 aux.loss_ce: 0.0078 aux.acc_seg: 99.3043 +04/18 10:39:58 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 10:39:58 - mmengine - INFO - Iter(train) [ 54000/160000] lr: 6.9345e-03 eta: 16:14:58 time: 0.5502 data_time: 0.0063 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0090 decode.acc_seg: 99.7044 aux.loss_ce: 0.0079 aux.acc_seg: 99.3671 +04/18 10:40:26 - mmengine - INFO - Iter(train) [ 54050/160000] lr: 6.9316e-03 eta: 16:14:30 time: 0.5515 data_time: 0.0063 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0084 decode.acc_seg: 99.6016 aux.loss_ce: 0.0081 aux.acc_seg: 99.1320 +04/18 10:40:53 - mmengine - INFO - Iter(train) [ 54100/160000] lr: 6.9287e-03 eta: 16:14:03 time: 0.5505 data_time: 0.0068 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0091 decode.acc_seg: 99.6835 aux.loss_ce: 0.0078 aux.acc_seg: 99.2524 +04/18 10:41:21 - mmengine - INFO - Iter(train) [ 54150/160000] lr: 6.9258e-03 eta: 16:13:35 time: 0.5519 data_time: 0.0065 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0086 decode.acc_seg: 99.6569 aux.loss_ce: 0.0078 aux.acc_seg: 99.2126 +04/18 10:41:49 - mmengine - INFO - Iter(train) [ 54200/160000] lr: 6.9229e-03 eta: 16:13:08 time: 0.5525 data_time: 0.0062 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0077 decode.acc_seg: 99.7202 aux.loss_ce: 0.0074 aux.acc_seg: 99.4156 +04/18 10:42:16 - mmengine - INFO - Iter(train) [ 54250/160000] lr: 6.9200e-03 eta: 16:12:40 time: 0.5512 data_time: 0.0063 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0084 decode.acc_seg: 99.5613 aux.loss_ce: 0.0075 aux.acc_seg: 99.0467 +04/18 10:42:44 - mmengine - INFO - Iter(train) [ 54300/160000] lr: 6.9171e-03 eta: 16:12:13 time: 0.5514 data_time: 0.0062 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0079 decode.acc_seg: 99.7068 aux.loss_ce: 0.0077 aux.acc_seg: 99.3682 +04/18 10:43:11 - mmengine - INFO - Iter(train) [ 54350/160000] lr: 6.9142e-03 eta: 16:11:45 time: 0.5512 data_time: 0.0071 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0084 decode.acc_seg: 99.6916 aux.loss_ce: 0.0076 aux.acc_seg: 99.2680 +04/18 10:43:39 - mmengine - INFO - Iter(train) [ 54400/160000] lr: 6.9113e-03 eta: 16:11:17 time: 0.5515 data_time: 0.0066 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0089 decode.acc_seg: 99.6791 aux.loss_ce: 0.0084 aux.acc_seg: 99.2279 +04/18 10:44:07 - mmengine - INFO - Iter(train) [ 54450/160000] lr: 6.9084e-03 eta: 16:10:50 time: 0.5521 data_time: 0.0067 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0096 decode.acc_seg: 99.6384 aux.loss_ce: 0.0084 aux.acc_seg: 99.0820 +04/18 10:44:34 - mmengine - INFO - Iter(train) [ 54500/160000] lr: 6.9055e-03 eta: 16:10:22 time: 0.5512 data_time: 0.0068 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0093 decode.acc_seg: 99.5789 aux.loss_ce: 0.0085 aux.acc_seg: 99.2597 +04/18 10:45:02 - mmengine - INFO - Iter(train) [ 54550/160000] lr: 6.9025e-03 eta: 16:09:55 time: 0.5517 data_time: 0.0070 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0080 decode.acc_seg: 99.6681 aux.loss_ce: 0.0079 aux.acc_seg: 99.1036 +04/18 10:45:29 - mmengine - INFO - Iter(train) [ 54600/160000] lr: 6.8996e-03 eta: 16:09:27 time: 0.5519 data_time: 0.0058 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0079 decode.acc_seg: 99.7135 aux.loss_ce: 0.0080 aux.acc_seg: 99.3480 +04/18 10:45:57 - mmengine - INFO - Iter(train) [ 54650/160000] lr: 6.8967e-03 eta: 16:08:59 time: 0.5526 data_time: 0.0073 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0085 decode.acc_seg: 99.6518 aux.loss_ce: 0.0078 aux.acc_seg: 99.1810 +04/18 10:46:25 - mmengine - INFO - Iter(train) [ 54700/160000] lr: 6.8938e-03 eta: 16:08:32 time: 0.5516 data_time: 0.0067 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0093 decode.acc_seg: 99.6024 aux.loss_ce: 0.0091 aux.acc_seg: 98.9346 +04/18 10:46:52 - mmengine - INFO - Iter(train) [ 54750/160000] lr: 6.8909e-03 eta: 16:08:04 time: 0.5523 data_time: 0.0064 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0080 decode.acc_seg: 99.7321 aux.loss_ce: 0.0079 aux.acc_seg: 99.2522 +04/18 10:47:20 - mmengine - INFO - Iter(train) [ 54800/160000] lr: 6.8880e-03 eta: 16:07:37 time: 0.5510 data_time: 0.0070 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0092 decode.acc_seg: 99.6336 aux.loss_ce: 0.0080 aux.acc_seg: 99.1231 +04/18 10:47:47 - mmengine - INFO - Iter(train) [ 54850/160000] lr: 6.8851e-03 eta: 16:07:09 time: 0.5519 data_time: 0.0064 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0088 decode.acc_seg: 99.6531 aux.loss_ce: 0.0082 aux.acc_seg: 99.1193 +04/18 10:48:15 - mmengine - INFO - Iter(train) [ 54900/160000] lr: 6.8822e-03 eta: 16:06:42 time: 0.5518 data_time: 0.0063 memory: 7635 loss: 0.0206 decode.loss_ce: 0.0119 decode.acc_seg: 99.6003 aux.loss_ce: 0.0087 aux.acc_seg: 99.1993 +04/18 10:48:43 - mmengine - INFO - Iter(train) [ 54950/160000] lr: 6.8793e-03 eta: 16:06:14 time: 0.5521 data_time: 0.0066 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0104 decode.acc_seg: 99.6529 aux.loss_ce: 0.0087 aux.acc_seg: 99.1139 +04/18 10:49:10 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 10:49:10 - mmengine - INFO - Iter(train) [ 55000/160000] lr: 6.8764e-03 eta: 16:05:46 time: 0.5526 data_time: 0.0065 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0100 decode.acc_seg: 99.3012 aux.loss_ce: 0.0084 aux.acc_seg: 98.9002 +04/18 10:49:38 - mmengine - INFO - Iter(train) [ 55050/160000] lr: 6.8735e-03 eta: 16:05:19 time: 0.5533 data_time: 0.0063 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0095 decode.acc_seg: 99.6247 aux.loss_ce: 0.0085 aux.acc_seg: 99.1119 +04/18 10:50:05 - mmengine - INFO - Iter(train) [ 55100/160000] lr: 6.8706e-03 eta: 16:04:51 time: 0.5527 data_time: 0.0067 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0094 decode.acc_seg: 99.6411 aux.loss_ce: 0.0089 aux.acc_seg: 99.0154 +04/18 10:50:33 - mmengine - INFO - Iter(train) [ 55150/160000] lr: 6.8677e-03 eta: 16:04:24 time: 0.5511 data_time: 0.0063 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0083 decode.acc_seg: 99.6661 aux.loss_ce: 0.0077 aux.acc_seg: 99.2380 +04/18 10:51:01 - mmengine - INFO - Iter(train) [ 55200/160000] lr: 6.8648e-03 eta: 16:03:56 time: 0.5512 data_time: 0.0065 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0087 decode.acc_seg: 99.6775 aux.loss_ce: 0.0080 aux.acc_seg: 99.2329 +04/18 10:51:28 - mmengine - INFO - Iter(train) [ 55250/160000] lr: 6.8619e-03 eta: 16:03:29 time: 0.5520 data_time: 0.0069 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0085 decode.acc_seg: 99.5274 aux.loss_ce: 0.0079 aux.acc_seg: 98.9709 +04/18 10:51:56 - mmengine - INFO - Iter(train) [ 55300/160000] lr: 6.8590e-03 eta: 16:03:01 time: 0.5508 data_time: 0.0067 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0093 decode.acc_seg: 99.6504 aux.loss_ce: 0.0082 aux.acc_seg: 99.0874 +04/18 10:52:24 - mmengine - INFO - Iter(train) [ 55350/160000] lr: 6.8561e-03 eta: 16:02:34 time: 0.5521 data_time: 0.0063 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0083 decode.acc_seg: 99.6454 aux.loss_ce: 0.0077 aux.acc_seg: 99.1481 +04/18 10:52:51 - mmengine - INFO - Iter(train) [ 55400/160000] lr: 6.8532e-03 eta: 16:02:06 time: 0.5520 data_time: 0.0076 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0104 decode.acc_seg: 99.7279 aux.loss_ce: 0.0088 aux.acc_seg: 99.1906 +04/18 10:53:19 - mmengine - INFO - Iter(train) [ 55450/160000] lr: 6.8503e-03 eta: 16:01:38 time: 0.5512 data_time: 0.0067 memory: 7635 loss: 0.0217 decode.loss_ce: 0.0121 decode.acc_seg: 99.5101 aux.loss_ce: 0.0096 aux.acc_seg: 98.7386 +04/18 10:53:46 - mmengine - INFO - Iter(train) [ 55500/160000] lr: 6.8474e-03 eta: 16:01:11 time: 0.5519 data_time: 0.0064 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0084 decode.acc_seg: 99.6796 aux.loss_ce: 0.0077 aux.acc_seg: 99.2816 +04/18 10:54:14 - mmengine - INFO - Iter(train) [ 55550/160000] lr: 6.8445e-03 eta: 16:00:43 time: 0.5518 data_time: 0.0063 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0096 decode.acc_seg: 99.6401 aux.loss_ce: 0.0089 aux.acc_seg: 99.1049 +04/18 10:54:42 - mmengine - INFO - Iter(train) [ 55600/160000] lr: 6.8416e-03 eta: 16:00:16 time: 0.5524 data_time: 0.0070 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0092 decode.acc_seg: 99.4155 aux.loss_ce: 0.0081 aux.acc_seg: 99.0135 +04/18 10:55:09 - mmengine - INFO - Iter(train) [ 55650/160000] lr: 6.8386e-03 eta: 15:59:48 time: 0.5518 data_time: 0.0062 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0084 decode.acc_seg: 99.6309 aux.loss_ce: 0.0079 aux.acc_seg: 99.1927 +04/18 10:55:37 - mmengine - INFO - Iter(train) [ 55700/160000] lr: 6.8357e-03 eta: 15:59:20 time: 0.5519 data_time: 0.0062 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0079 decode.acc_seg: 99.7025 aux.loss_ce: 0.0079 aux.acc_seg: 99.2126 +04/18 10:56:04 - mmengine - INFO - Iter(train) [ 55750/160000] lr: 6.8328e-03 eta: 15:58:53 time: 0.5510 data_time: 0.0062 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0083 decode.acc_seg: 99.5903 aux.loss_ce: 0.0079 aux.acc_seg: 99.2051 +04/18 10:56:32 - mmengine - INFO - Iter(train) [ 55800/160000] lr: 6.8299e-03 eta: 15:58:25 time: 0.5524 data_time: 0.0065 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0082 decode.acc_seg: 99.7313 aux.loss_ce: 0.0076 aux.acc_seg: 99.3532 +04/18 10:57:00 - mmengine - INFO - Iter(train) [ 55850/160000] lr: 6.8270e-03 eta: 15:57:58 time: 0.5504 data_time: 0.0061 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0079 decode.acc_seg: 99.6006 aux.loss_ce: 0.0074 aux.acc_seg: 99.1119 +04/18 10:57:27 - mmengine - INFO - Iter(train) [ 55900/160000] lr: 6.8241e-03 eta: 15:57:30 time: 0.5519 data_time: 0.0066 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0075 decode.acc_seg: 99.6524 aux.loss_ce: 0.0075 aux.acc_seg: 99.1038 +04/18 10:57:55 - mmengine - INFO - Iter(train) [ 55950/160000] lr: 6.8212e-03 eta: 15:57:03 time: 0.5494 data_time: 0.0055 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0090 decode.acc_seg: 99.6177 aux.loss_ce: 0.0081 aux.acc_seg: 99.1803 +04/18 10:58:22 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 10:58:22 - mmengine - INFO - Iter(train) [ 56000/160000] lr: 6.8183e-03 eta: 15:56:35 time: 0.5524 data_time: 0.0063 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0077 decode.acc_seg: 99.6518 aux.loss_ce: 0.0076 aux.acc_seg: 99.2232 +04/18 10:58:50 - mmengine - INFO - Iter(train) [ 56050/160000] lr: 6.8154e-03 eta: 15:56:07 time: 0.5505 data_time: 0.0064 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0084 decode.acc_seg: 99.6863 aux.loss_ce: 0.0075 aux.acc_seg: 99.2343 +04/18 10:59:18 - mmengine - INFO - Iter(train) [ 56100/160000] lr: 6.8125e-03 eta: 15:55:40 time: 0.5526 data_time: 0.0062 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0092 decode.acc_seg: 99.7315 aux.loss_ce: 0.0081 aux.acc_seg: 99.4270 +04/18 10:59:45 - mmengine - INFO - Iter(train) [ 56150/160000] lr: 6.8096e-03 eta: 15:55:12 time: 0.5525 data_time: 0.0072 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0078 decode.acc_seg: 99.7353 aux.loss_ce: 0.0077 aux.acc_seg: 99.2964 +04/18 11:00:13 - mmengine - INFO - Iter(train) [ 56200/160000] lr: 6.8067e-03 eta: 15:54:44 time: 0.5529 data_time: 0.0066 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0090 decode.acc_seg: 99.7451 aux.loss_ce: 0.0085 aux.acc_seg: 99.1852 +04/18 11:00:40 - mmengine - INFO - Iter(train) [ 56250/160000] lr: 6.8038e-03 eta: 15:54:17 time: 0.5504 data_time: 0.0064 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0092 decode.acc_seg: 99.5820 aux.loss_ce: 0.0084 aux.acc_seg: 99.1267 +04/18 11:01:08 - mmengine - INFO - Iter(train) [ 56300/160000] lr: 6.8009e-03 eta: 15:53:50 time: 0.5510 data_time: 0.0064 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0084 decode.acc_seg: 99.5940 aux.loss_ce: 0.0077 aux.acc_seg: 99.0654 +04/18 11:01:36 - mmengine - INFO - Iter(train) [ 56350/160000] lr: 6.7980e-03 eta: 15:53:22 time: 0.5511 data_time: 0.0066 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0090 decode.acc_seg: 99.6688 aux.loss_ce: 0.0083 aux.acc_seg: 99.1933 +04/18 11:02:03 - mmengine - INFO - Iter(train) [ 56400/160000] lr: 6.7950e-03 eta: 15:52:54 time: 0.5526 data_time: 0.0062 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0094 decode.acc_seg: 99.5991 aux.loss_ce: 0.0084 aux.acc_seg: 99.0885 +04/18 11:02:31 - mmengine - INFO - Iter(train) [ 56450/160000] lr: 6.7921e-03 eta: 15:52:27 time: 0.5520 data_time: 0.0065 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0086 decode.acc_seg: 99.6034 aux.loss_ce: 0.0082 aux.acc_seg: 99.0778 +04/18 11:02:58 - mmengine - INFO - Iter(train) [ 56500/160000] lr: 6.7892e-03 eta: 15:51:59 time: 0.5517 data_time: 0.0071 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0091 decode.acc_seg: 99.5968 aux.loss_ce: 0.0084 aux.acc_seg: 99.0685 +04/18 11:03:26 - mmengine - INFO - Iter(train) [ 56550/160000] lr: 6.7863e-03 eta: 15:51:31 time: 0.5508 data_time: 0.0065 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0087 decode.acc_seg: 99.6239 aux.loss_ce: 0.0085 aux.acc_seg: 99.1067 +04/18 11:03:54 - mmengine - INFO - Iter(train) [ 56600/160000] lr: 6.7834e-03 eta: 15:51:04 time: 0.5506 data_time: 0.0066 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0076 decode.acc_seg: 99.7949 aux.loss_ce: 0.0077 aux.acc_seg: 99.3800 +04/18 11:04:21 - mmengine - INFO - Iter(train) [ 56650/160000] lr: 6.7805e-03 eta: 15:50:36 time: 0.5518 data_time: 0.0073 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0081 decode.acc_seg: 99.6842 aux.loss_ce: 0.0079 aux.acc_seg: 98.9669 +04/18 11:04:49 - mmengine - INFO - Iter(train) [ 56700/160000] lr: 6.7776e-03 eta: 15:50:08 time: 0.5525 data_time: 0.0064 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0076 decode.acc_seg: 99.7064 aux.loss_ce: 0.0075 aux.acc_seg: 99.3327 +04/18 11:05:16 - mmengine - INFO - Iter(train) [ 56750/160000] lr: 6.7747e-03 eta: 15:49:41 time: 0.5531 data_time: 0.0064 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0088 decode.acc_seg: 99.5516 aux.loss_ce: 0.0080 aux.acc_seg: 99.0128 +04/18 11:05:44 - mmengine - INFO - Iter(train) [ 56800/160000] lr: 6.7718e-03 eta: 15:49:13 time: 0.5524 data_time: 0.0064 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0084 decode.acc_seg: 99.6427 aux.loss_ce: 0.0076 aux.acc_seg: 98.9308 +04/18 11:06:11 - mmengine - INFO - Iter(train) [ 56850/160000] lr: 6.7689e-03 eta: 15:48:46 time: 0.5513 data_time: 0.0060 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0076 decode.acc_seg: 99.7216 aux.loss_ce: 0.0076 aux.acc_seg: 99.3689 +04/18 11:06:39 - mmengine - INFO - Iter(train) [ 56900/160000] lr: 6.7660e-03 eta: 15:48:18 time: 0.5510 data_time: 0.0059 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0091 decode.acc_seg: 99.6036 aux.loss_ce: 0.0088 aux.acc_seg: 98.8121 +04/18 11:07:07 - mmengine - INFO - Iter(train) [ 56950/160000] lr: 6.7630e-03 eta: 15:47:50 time: 0.5505 data_time: 0.0067 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0088 decode.acc_seg: 99.6259 aux.loss_ce: 0.0082 aux.acc_seg: 99.0293 +04/18 11:07:34 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 11:07:34 - mmengine - INFO - Iter(train) [ 57000/160000] lr: 6.7601e-03 eta: 15:47:23 time: 0.5509 data_time: 0.0060 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0092 decode.acc_seg: 99.6693 aux.loss_ce: 0.0081 aux.acc_seg: 99.0869 +04/18 11:08:02 - mmengine - INFO - Iter(train) [ 57050/160000] lr: 6.7572e-03 eta: 15:46:55 time: 0.5509 data_time: 0.0060 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0080 decode.acc_seg: 99.6544 aux.loss_ce: 0.0081 aux.acc_seg: 99.0747 +04/18 11:08:29 - mmengine - INFO - Iter(train) [ 57100/160000] lr: 6.7543e-03 eta: 15:46:27 time: 0.5516 data_time: 0.0067 memory: 7635 loss: 0.0192 decode.loss_ce: 0.0102 decode.acc_seg: 99.7098 aux.loss_ce: 0.0091 aux.acc_seg: 99.2833 +04/18 11:08:57 - mmengine - INFO - Iter(train) [ 57150/160000] lr: 6.7514e-03 eta: 15:46:00 time: 0.5514 data_time: 0.0069 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0104 decode.acc_seg: 99.5404 aux.loss_ce: 0.0084 aux.acc_seg: 99.0448 +04/18 11:09:25 - mmengine - INFO - Iter(train) [ 57200/160000] lr: 6.7485e-03 eta: 15:45:32 time: 0.5514 data_time: 0.0060 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0092 decode.acc_seg: 99.7238 aux.loss_ce: 0.0083 aux.acc_seg: 99.3373 +04/18 11:09:52 - mmengine - INFO - Iter(train) [ 57250/160000] lr: 6.7456e-03 eta: 15:45:05 time: 0.5520 data_time: 0.0066 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0083 decode.acc_seg: 99.6117 aux.loss_ce: 0.0077 aux.acc_seg: 99.0710 +04/18 11:10:20 - mmengine - INFO - Iter(train) [ 57300/160000] lr: 6.7427e-03 eta: 15:44:37 time: 0.5508 data_time: 0.0064 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0083 decode.acc_seg: 99.5892 aux.loss_ce: 0.0080 aux.acc_seg: 99.1467 +04/18 11:10:47 - mmengine - INFO - Iter(train) [ 57350/160000] lr: 6.7398e-03 eta: 15:44:09 time: 0.5499 data_time: 0.0062 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0092 decode.acc_seg: 99.6417 aux.loss_ce: 0.0086 aux.acc_seg: 99.1454 +04/18 11:11:15 - mmengine - INFO - Iter(train) [ 57400/160000] lr: 6.7369e-03 eta: 15:43:42 time: 0.5508 data_time: 0.0066 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0086 decode.acc_seg: 99.8440 aux.loss_ce: 0.0083 aux.acc_seg: 99.5678 +04/18 11:11:43 - mmengine - INFO - Iter(train) [ 57450/160000] lr: 6.7339e-03 eta: 15:43:14 time: 0.5520 data_time: 0.0062 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0085 decode.acc_seg: 99.7033 aux.loss_ce: 0.0080 aux.acc_seg: 99.3217 +04/18 11:12:10 - mmengine - INFO - Iter(train) [ 57500/160000] lr: 6.7310e-03 eta: 15:42:47 time: 0.5516 data_time: 0.0073 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0094 decode.acc_seg: 99.6173 aux.loss_ce: 0.0085 aux.acc_seg: 99.0421 +04/18 11:12:38 - mmengine - INFO - Iter(train) [ 57550/160000] lr: 6.7281e-03 eta: 15:42:19 time: 0.5508 data_time: 0.0067 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0076 decode.acc_seg: 99.7353 aux.loss_ce: 0.0077 aux.acc_seg: 99.3336 +04/18 11:13:05 - mmengine - INFO - Iter(train) [ 57600/160000] lr: 6.7252e-03 eta: 15:41:51 time: 0.5518 data_time: 0.0068 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0093 decode.acc_seg: 99.4840 aux.loss_ce: 0.0082 aux.acc_seg: 99.0304 +04/18 11:13:33 - mmengine - INFO - Iter(train) [ 57650/160000] lr: 6.7223e-03 eta: 15:41:24 time: 0.5531 data_time: 0.0063 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0087 decode.acc_seg: 99.6527 aux.loss_ce: 0.0078 aux.acc_seg: 99.0683 +04/18 11:14:00 - mmengine - INFO - Iter(train) [ 57700/160000] lr: 6.7194e-03 eta: 15:40:56 time: 0.5517 data_time: 0.0061 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0082 decode.acc_seg: 99.6759 aux.loss_ce: 0.0078 aux.acc_seg: 99.2591 +04/18 11:14:28 - mmengine - INFO - Iter(train) [ 57750/160000] lr: 6.7165e-03 eta: 15:40:28 time: 0.5514 data_time: 0.0064 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0090 decode.acc_seg: 99.7263 aux.loss_ce: 0.0084 aux.acc_seg: 99.3308 +04/18 11:14:56 - mmengine - INFO - Iter(train) [ 57800/160000] lr: 6.7136e-03 eta: 15:40:01 time: 0.5523 data_time: 0.0061 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0076 decode.acc_seg: 99.7467 aux.loss_ce: 0.0074 aux.acc_seg: 99.4081 +04/18 11:15:23 - mmengine - INFO - Iter(train) [ 57850/160000] lr: 6.7107e-03 eta: 15:39:33 time: 0.5522 data_time: 0.0064 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0086 decode.acc_seg: 99.6671 aux.loss_ce: 0.0079 aux.acc_seg: 99.3294 +04/18 11:15:51 - mmengine - INFO - Iter(train) [ 57900/160000] lr: 6.7077e-03 eta: 15:39:05 time: 0.5506 data_time: 0.0065 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0082 decode.acc_seg: 99.6827 aux.loss_ce: 0.0085 aux.acc_seg: 99.1842 +04/18 11:16:18 - mmengine - INFO - Iter(train) [ 57950/160000] lr: 6.7048e-03 eta: 15:38:38 time: 0.5503 data_time: 0.0060 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0075 decode.acc_seg: 99.6693 aux.loss_ce: 0.0072 aux.acc_seg: 99.3571 +04/18 11:16:46 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 11:16:46 - mmengine - INFO - Iter(train) [ 58000/160000] lr: 6.7019e-03 eta: 15:38:10 time: 0.5511 data_time: 0.0064 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0079 decode.acc_seg: 99.7209 aux.loss_ce: 0.0077 aux.acc_seg: 99.3996 +04/18 11:17:13 - mmengine - INFO - Iter(train) [ 58050/160000] lr: 6.6990e-03 eta: 15:37:42 time: 0.5505 data_time: 0.0068 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0085 decode.acc_seg: 99.7552 aux.loss_ce: 0.0080 aux.acc_seg: 99.3280 +04/18 11:17:41 - mmengine - INFO - Iter(train) [ 58100/160000] lr: 6.6961e-03 eta: 15:37:15 time: 0.5514 data_time: 0.0067 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0089 decode.acc_seg: 99.6637 aux.loss_ce: 0.0080 aux.acc_seg: 99.1197 +04/18 11:18:08 - mmengine - INFO - Iter(train) [ 58150/160000] lr: 6.6932e-03 eta: 15:36:47 time: 0.5507 data_time: 0.0069 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0107 decode.acc_seg: 99.4823 aux.loss_ce: 0.0088 aux.acc_seg: 98.7581 +04/18 11:18:36 - mmengine - INFO - Iter(train) [ 58200/160000] lr: 6.6903e-03 eta: 15:36:19 time: 0.5510 data_time: 0.0069 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0093 decode.acc_seg: 99.5644 aux.loss_ce: 0.0087 aux.acc_seg: 99.1653 +04/18 11:19:04 - mmengine - INFO - Iter(train) [ 58250/160000] lr: 6.6873e-03 eta: 15:35:52 time: 0.5503 data_time: 0.0065 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0095 decode.acc_seg: 99.5499 aux.loss_ce: 0.0084 aux.acc_seg: 99.2001 +04/18 11:19:31 - mmengine - INFO - Iter(train) [ 58300/160000] lr: 6.6844e-03 eta: 15:35:24 time: 0.5503 data_time: 0.0063 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0081 decode.acc_seg: 99.6187 aux.loss_ce: 0.0076 aux.acc_seg: 99.1308 +04/18 11:19:59 - mmengine - INFO - Iter(train) [ 58350/160000] lr: 6.6815e-03 eta: 15:34:56 time: 0.5507 data_time: 0.0063 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0085 decode.acc_seg: 99.6239 aux.loss_ce: 0.0082 aux.acc_seg: 99.1610 +04/18 11:20:26 - mmengine - INFO - Iter(train) [ 58400/160000] lr: 6.6786e-03 eta: 15:34:29 time: 0.5600 data_time: 0.0066 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0072 decode.acc_seg: 99.6520 aux.loss_ce: 0.0067 aux.acc_seg: 99.0705 +04/18 11:20:54 - mmengine - INFO - Iter(train) [ 58450/160000] lr: 6.6757e-03 eta: 15:34:01 time: 0.5526 data_time: 0.0068 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0076 decode.acc_seg: 99.6987 aux.loss_ce: 0.0075 aux.acc_seg: 99.2962 +04/18 11:21:22 - mmengine - INFO - Iter(train) [ 58500/160000] lr: 6.6728e-03 eta: 15:33:34 time: 0.5530 data_time: 0.0070 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0080 decode.acc_seg: 99.7561 aux.loss_ce: 0.0079 aux.acc_seg: 99.3450 +04/18 11:21:49 - mmengine - INFO - Iter(train) [ 58550/160000] lr: 6.6699e-03 eta: 15:33:06 time: 0.5514 data_time: 0.0062 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0076 decode.acc_seg: 99.6677 aux.loss_ce: 0.0077 aux.acc_seg: 99.0487 +04/18 11:22:17 - mmengine - INFO - Iter(train) [ 58600/160000] lr: 6.6670e-03 eta: 15:32:39 time: 0.5516 data_time: 0.0070 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0076 decode.acc_seg: 99.6812 aux.loss_ce: 0.0077 aux.acc_seg: 99.1712 +04/18 11:22:44 - mmengine - INFO - Iter(train) [ 58650/160000] lr: 6.6640e-03 eta: 15:32:11 time: 0.5511 data_time: 0.0063 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0085 decode.acc_seg: 99.5533 aux.loss_ce: 0.0079 aux.acc_seg: 98.9416 +04/18 11:23:12 - mmengine - INFO - Iter(train) [ 58700/160000] lr: 6.6611e-03 eta: 15:31:43 time: 0.5511 data_time: 0.0062 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0075 decode.acc_seg: 99.7086 aux.loss_ce: 0.0072 aux.acc_seg: 99.3958 +04/18 11:23:40 - mmengine - INFO - Iter(train) [ 58750/160000] lr: 6.6582e-03 eta: 15:31:16 time: 0.5524 data_time: 0.0065 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0088 decode.acc_seg: 99.5885 aux.loss_ce: 0.0086 aux.acc_seg: 98.8952 +04/18 11:24:07 - mmengine - INFO - Iter(train) [ 58800/160000] lr: 6.6553e-03 eta: 15:30:48 time: 0.5517 data_time: 0.0065 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0074 decode.acc_seg: 99.6488 aux.loss_ce: 0.0078 aux.acc_seg: 98.9652 +04/18 11:24:35 - mmengine - INFO - Iter(train) [ 58850/160000] lr: 6.6524e-03 eta: 15:30:20 time: 0.5511 data_time: 0.0060 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0077 decode.acc_seg: 99.5907 aux.loss_ce: 0.0076 aux.acc_seg: 98.9631 +04/18 11:25:02 - mmengine - INFO - Iter(train) [ 58900/160000] lr: 6.6495e-03 eta: 15:29:53 time: 0.5522 data_time: 0.0067 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0083 decode.acc_seg: 99.6446 aux.loss_ce: 0.0081 aux.acc_seg: 99.4057 +04/18 11:25:30 - mmengine - INFO - Iter(train) [ 58950/160000] lr: 6.6465e-03 eta: 15:29:25 time: 0.5521 data_time: 0.0066 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0081 decode.acc_seg: 99.6372 aux.loss_ce: 0.0079 aux.acc_seg: 99.0109 +04/18 11:25:57 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 11:25:57 - mmengine - INFO - Iter(train) [ 59000/160000] lr: 6.6436e-03 eta: 15:28:58 time: 0.5523 data_time: 0.0066 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0082 decode.acc_seg: 99.5790 aux.loss_ce: 0.0079 aux.acc_seg: 99.1814 +04/18 11:26:25 - mmengine - INFO - Iter(train) [ 59050/160000] lr: 6.6407e-03 eta: 15:28:30 time: 0.5517 data_time: 0.0067 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0079 decode.acc_seg: 99.6516 aux.loss_ce: 0.0078 aux.acc_seg: 99.1557 +04/18 11:26:53 - mmengine - INFO - Iter(train) [ 59100/160000] lr: 6.6378e-03 eta: 15:28:02 time: 0.5521 data_time: 0.0071 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0077 decode.acc_seg: 99.6309 aux.loss_ce: 0.0079 aux.acc_seg: 99.1999 +04/18 11:27:20 - mmengine - INFO - Iter(train) [ 59150/160000] lr: 6.6349e-03 eta: 15:27:35 time: 0.5497 data_time: 0.0068 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0070 decode.acc_seg: 99.6934 aux.loss_ce: 0.0073 aux.acc_seg: 99.2998 +04/18 11:27:48 - mmengine - INFO - Iter(train) [ 59200/160000] lr: 6.6320e-03 eta: 15:27:07 time: 0.5516 data_time: 0.0062 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0073 decode.acc_seg: 99.7056 aux.loss_ce: 0.0074 aux.acc_seg: 99.2040 +04/18 11:28:15 - mmengine - INFO - Iter(train) [ 59250/160000] lr: 6.6291e-03 eta: 15:26:39 time: 0.5504 data_time: 0.0064 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0073 decode.acc_seg: 99.5614 aux.loss_ce: 0.0074 aux.acc_seg: 98.7409 +04/18 11:28:43 - mmengine - INFO - Iter(train) [ 59300/160000] lr: 6.6261e-03 eta: 15:26:12 time: 0.5523 data_time: 0.0071 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0079 decode.acc_seg: 99.6745 aux.loss_ce: 0.0078 aux.acc_seg: 99.0722 +04/18 11:29:10 - mmengine - INFO - Iter(train) [ 59350/160000] lr: 6.6232e-03 eta: 15:25:44 time: 0.5518 data_time: 0.0069 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0071 decode.acc_seg: 99.7550 aux.loss_ce: 0.0073 aux.acc_seg: 99.2683 +04/18 11:29:38 - mmengine - INFO - Iter(train) [ 59400/160000] lr: 6.6203e-03 eta: 15:25:17 time: 0.5527 data_time: 0.0062 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0075 decode.acc_seg: 99.7341 aux.loss_ce: 0.0074 aux.acc_seg: 99.1190 +04/18 11:30:06 - mmengine - INFO - Iter(train) [ 59450/160000] lr: 6.6174e-03 eta: 15:24:49 time: 0.5511 data_time: 0.0068 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0084 decode.acc_seg: 99.6515 aux.loss_ce: 0.0078 aux.acc_seg: 99.1786 +04/18 11:30:33 - mmengine - INFO - Iter(train) [ 59500/160000] lr: 6.6145e-03 eta: 15:24:21 time: 0.5499 data_time: 0.0061 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0079 decode.acc_seg: 99.7619 aux.loss_ce: 0.0075 aux.acc_seg: 99.3440 +04/18 11:31:01 - mmengine - INFO - Iter(train) [ 59550/160000] lr: 6.6116e-03 eta: 15:23:54 time: 0.5514 data_time: 0.0063 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0089 decode.acc_seg: 99.5975 aux.loss_ce: 0.0085 aux.acc_seg: 98.8824 +04/18 11:31:28 - mmengine - INFO - Iter(train) [ 59600/160000] lr: 6.6086e-03 eta: 15:23:26 time: 0.5508 data_time: 0.0063 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0088 decode.acc_seg: 99.6656 aux.loss_ce: 0.0081 aux.acc_seg: 99.1330 +04/18 11:31:56 - mmengine - INFO - Iter(train) [ 59650/160000] lr: 6.6057e-03 eta: 15:22:58 time: 0.5508 data_time: 0.0063 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0082 decode.acc_seg: 99.6439 aux.loss_ce: 0.0081 aux.acc_seg: 99.0078 +04/18 11:32:24 - mmengine - INFO - Iter(train) [ 59700/160000] lr: 6.6028e-03 eta: 15:22:31 time: 0.5511 data_time: 0.0068 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0085 decode.acc_seg: 99.6826 aux.loss_ce: 0.0083 aux.acc_seg: 99.2955 +04/18 11:32:51 - mmengine - INFO - Iter(train) [ 59750/160000] lr: 6.5999e-03 eta: 15:22:03 time: 0.5512 data_time: 0.0072 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0091 decode.acc_seg: 99.6603 aux.loss_ce: 0.0080 aux.acc_seg: 99.3613 +04/18 11:33:19 - mmengine - INFO - Iter(train) [ 59800/160000] lr: 6.5970e-03 eta: 15:21:35 time: 0.5516 data_time: 0.0069 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0092 decode.acc_seg: 99.7003 aux.loss_ce: 0.0079 aux.acc_seg: 99.1389 +04/18 11:33:46 - mmengine - INFO - Iter(train) [ 59850/160000] lr: 6.5940e-03 eta: 15:21:08 time: 0.5512 data_time: 0.0073 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0086 decode.acc_seg: 99.6959 aux.loss_ce: 0.0081 aux.acc_seg: 99.3964 +04/18 11:34:14 - mmengine - INFO - Iter(train) [ 59900/160000] lr: 6.5911e-03 eta: 15:20:40 time: 0.5511 data_time: 0.0063 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0096 decode.acc_seg: 99.4831 aux.loss_ce: 0.0087 aux.acc_seg: 98.9328 +04/18 11:34:41 - mmengine - INFO - Iter(train) [ 59950/160000] lr: 6.5882e-03 eta: 15:20:13 time: 0.5516 data_time: 0.0068 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0076 decode.acc_seg: 99.7497 aux.loss_ce: 0.0073 aux.acc_seg: 99.2151 +04/18 11:35:09 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 11:35:09 - mmengine - INFO - Iter(train) [ 60000/160000] lr: 6.5853e-03 eta: 15:19:45 time: 0.5519 data_time: 0.0064 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0087 decode.acc_seg: 99.6785 aux.loss_ce: 0.0081 aux.acc_seg: 99.2693 +04/18 11:35:09 - mmengine - INFO - Saving checkpoint at 60000 iterations +04/18 11:35:13 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:07 time: 0.0462 data_time: 0.0013 memory: 1657 +04/18 11:35:15 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:04 time: 0.0462 data_time: 0.0014 memory: 1657 +04/18 11:35:18 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:02 time: 0.0462 data_time: 0.0013 memory: 1657 +04/18 11:35:20 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.0454 data_time: 0.0013 memory: 1657 +04/18 11:35:20 - mmengine - INFO - per class results: +04/18 11:35:20 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 99.06 | 99.58 | 99.53 | 99.48 | 99.58 | +| contrast | 79.36 | 87.46 | 88.49 | 89.55 | 87.46 | ++------------+-------+-------+--------+-----------+--------+ +04/18 11:35:20 - mmengine - INFO - Iter(val) [200/200] aAcc: 99.0900 mIoU: 89.2100 mAcc: 93.5200 mFscore: 94.0100 mPrecision: 94.5200 mRecall: 93.5200 data_time: 0.0015 time: 0.0465 +04/18 11:35:48 - mmengine - INFO - Iter(train) [ 60050/160000] lr: 6.5824e-03 eta: 15:19:18 time: 0.5519 data_time: 0.0062 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0083 decode.acc_seg: 99.7064 aux.loss_ce: 0.0078 aux.acc_seg: 99.3352 +04/18 11:36:15 - mmengine - INFO - Iter(train) [ 60100/160000] lr: 6.5795e-03 eta: 15:18:50 time: 0.5499 data_time: 0.0063 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0089 decode.acc_seg: 99.4396 aux.loss_ce: 0.0081 aux.acc_seg: 98.9150 +04/18 11:36:43 - mmengine - INFO - Iter(train) [ 60150/160000] lr: 6.5765e-03 eta: 15:18:22 time: 0.5512 data_time: 0.0060 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0109 decode.acc_seg: 99.6948 aux.loss_ce: 0.0086 aux.acc_seg: 99.2352 +04/18 11:37:10 - mmengine - INFO - Iter(train) [ 60200/160000] lr: 6.5736e-03 eta: 15:17:54 time: 0.5514 data_time: 0.0066 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0085 decode.acc_seg: 99.6398 aux.loss_ce: 0.0081 aux.acc_seg: 99.0908 +04/18 11:37:38 - mmengine - INFO - Iter(train) [ 60250/160000] lr: 6.5707e-03 eta: 15:17:27 time: 0.5507 data_time: 0.0063 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0089 decode.acc_seg: 99.7389 aux.loss_ce: 0.0083 aux.acc_seg: 99.3309 +04/18 11:38:05 - mmengine - INFO - Iter(train) [ 60300/160000] lr: 6.5678e-03 eta: 15:16:59 time: 0.5512 data_time: 0.0060 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0088 decode.acc_seg: 99.5622 aux.loss_ce: 0.0078 aux.acc_seg: 99.2293 +04/18 11:38:33 - mmengine - INFO - Iter(train) [ 60350/160000] lr: 6.5649e-03 eta: 15:16:32 time: 0.5517 data_time: 0.0067 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0072 decode.acc_seg: 99.7232 aux.loss_ce: 0.0070 aux.acc_seg: 99.2214 +04/18 11:39:01 - mmengine - INFO - Iter(train) [ 60400/160000] lr: 6.5619e-03 eta: 15:16:04 time: 0.5524 data_time: 0.0070 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0078 decode.acc_seg: 99.7540 aux.loss_ce: 0.0081 aux.acc_seg: 99.2719 +04/18 11:39:28 - mmengine - INFO - Iter(train) [ 60450/160000] lr: 6.5590e-03 eta: 15:15:36 time: 0.5500 data_time: 0.0063 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0083 decode.acc_seg: 99.5952 aux.loss_ce: 0.0077 aux.acc_seg: 99.0477 +04/18 11:39:56 - mmengine - INFO - Iter(train) [ 60500/160000] lr: 6.5561e-03 eta: 15:15:09 time: 0.5512 data_time: 0.0061 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0083 decode.acc_seg: 99.7226 aux.loss_ce: 0.0082 aux.acc_seg: 99.1763 +04/18 11:40:23 - mmengine - INFO - Iter(train) [ 60550/160000] lr: 6.5532e-03 eta: 15:14:41 time: 0.5591 data_time: 0.0070 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0076 decode.acc_seg: 99.7407 aux.loss_ce: 0.0075 aux.acc_seg: 99.0932 +04/18 11:40:51 - mmengine - INFO - Iter(train) [ 60600/160000] lr: 6.5503e-03 eta: 15:14:14 time: 0.5582 data_time: 0.0068 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0075 decode.acc_seg: 99.7237 aux.loss_ce: 0.0074 aux.acc_seg: 99.2720 +04/18 11:41:19 - mmengine - INFO - Iter(train) [ 60650/160000] lr: 6.5473e-03 eta: 15:13:46 time: 0.5513 data_time: 0.0058 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0077 decode.acc_seg: 99.7244 aux.loss_ce: 0.0080 aux.acc_seg: 99.3115 +04/18 11:41:46 - mmengine - INFO - Iter(train) [ 60700/160000] lr: 6.5444e-03 eta: 15:13:18 time: 0.5509 data_time: 0.0062 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0080 decode.acc_seg: 99.6064 aux.loss_ce: 0.0075 aux.acc_seg: 99.2377 +04/18 11:42:14 - mmengine - INFO - Iter(train) [ 60750/160000] lr: 6.5415e-03 eta: 15:12:51 time: 0.5510 data_time: 0.0062 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0072 decode.acc_seg: 99.6486 aux.loss_ce: 0.0072 aux.acc_seg: 99.1142 +04/18 11:42:41 - mmengine - INFO - Iter(train) [ 60800/160000] lr: 6.5386e-03 eta: 15:12:23 time: 0.5524 data_time: 0.0068 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0078 decode.acc_seg: 99.6813 aux.loss_ce: 0.0077 aux.acc_seg: 99.2957 +04/18 11:43:09 - mmengine - INFO - Iter(train) [ 60850/160000] lr: 6.5357e-03 eta: 15:11:55 time: 0.5508 data_time: 0.0063 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0071 decode.acc_seg: 99.7236 aux.loss_ce: 0.0072 aux.acc_seg: 99.3913 +04/18 11:43:36 - mmengine - INFO - Iter(train) [ 60900/160000] lr: 6.5327e-03 eta: 15:11:28 time: 0.5526 data_time: 0.0075 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0078 decode.acc_seg: 99.6771 aux.loss_ce: 0.0076 aux.acc_seg: 99.3098 +04/18 11:44:04 - mmengine - INFO - Iter(train) [ 60950/160000] lr: 6.5298e-03 eta: 15:11:00 time: 0.5517 data_time: 0.0059 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0075 decode.acc_seg: 99.6767 aux.loss_ce: 0.0075 aux.acc_seg: 99.2650 +04/18 11:44:32 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 11:44:32 - mmengine - INFO - Iter(train) [ 61000/160000] lr: 6.5269e-03 eta: 15:10:32 time: 0.5503 data_time: 0.0064 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0074 decode.acc_seg: 99.6779 aux.loss_ce: 0.0075 aux.acc_seg: 99.0499 +04/18 11:44:59 - mmengine - INFO - Iter(train) [ 61050/160000] lr: 6.5240e-03 eta: 15:10:05 time: 0.5510 data_time: 0.0064 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0079 decode.acc_seg: 99.7244 aux.loss_ce: 0.0076 aux.acc_seg: 99.3732 +04/18 11:45:27 - mmengine - INFO - Iter(train) [ 61100/160000] lr: 6.5211e-03 eta: 15:09:37 time: 0.5503 data_time: 0.0061 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0072 decode.acc_seg: 99.7668 aux.loss_ce: 0.0067 aux.acc_seg: 99.5061 +04/18 11:45:54 - mmengine - INFO - Iter(train) [ 61150/160000] lr: 6.5181e-03 eta: 15:09:10 time: 0.5511 data_time: 0.0064 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0074 decode.acc_seg: 99.7355 aux.loss_ce: 0.0079 aux.acc_seg: 99.3558 +04/18 11:46:22 - mmengine - INFO - Iter(train) [ 61200/160000] lr: 6.5152e-03 eta: 15:08:42 time: 0.5528 data_time: 0.0064 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0076 decode.acc_seg: 99.7243 aux.loss_ce: 0.0079 aux.acc_seg: 99.3180 +04/18 11:46:49 - mmengine - INFO - Iter(train) [ 61250/160000] lr: 6.5123e-03 eta: 15:08:14 time: 0.5514 data_time: 0.0063 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0087 decode.acc_seg: 99.6181 aux.loss_ce: 0.0081 aux.acc_seg: 98.9738 +04/18 11:47:17 - mmengine - INFO - Iter(train) [ 61300/160000] lr: 6.5094e-03 eta: 15:07:47 time: 0.5513 data_time: 0.0072 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0073 decode.acc_seg: 99.6649 aux.loss_ce: 0.0074 aux.acc_seg: 99.2748 +04/18 11:47:45 - mmengine - INFO - Iter(train) [ 61350/160000] lr: 6.5064e-03 eta: 15:07:19 time: 0.5510 data_time: 0.0065 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0071 decode.acc_seg: 99.6904 aux.loss_ce: 0.0070 aux.acc_seg: 99.2872 +04/18 11:48:12 - mmengine - INFO - Iter(train) [ 61400/160000] lr: 6.5035e-03 eta: 15:06:51 time: 0.5499 data_time: 0.0060 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0073 decode.acc_seg: 99.7398 aux.loss_ce: 0.0079 aux.acc_seg: 99.1993 +04/18 11:48:40 - mmengine - INFO - Iter(train) [ 61450/160000] lr: 6.5006e-03 eta: 15:06:24 time: 0.5509 data_time: 0.0065 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0073 decode.acc_seg: 99.6499 aux.loss_ce: 0.0073 aux.acc_seg: 99.1858 +04/18 11:49:07 - mmengine - INFO - Iter(train) [ 61500/160000] lr: 6.4977e-03 eta: 15:05:56 time: 0.5512 data_time: 0.0066 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0079 decode.acc_seg: 99.6922 aux.loss_ce: 0.0080 aux.acc_seg: 99.2148 +04/18 11:49:35 - mmengine - INFO - Iter(train) [ 61550/160000] lr: 6.4948e-03 eta: 15:05:29 time: 0.5604 data_time: 0.0062 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0077 decode.acc_seg: 99.6339 aux.loss_ce: 0.0078 aux.acc_seg: 99.1366 +04/18 11:50:02 - mmengine - INFO - Iter(train) [ 61600/160000] lr: 6.4918e-03 eta: 15:05:01 time: 0.5510 data_time: 0.0071 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0078 decode.acc_seg: 99.7356 aux.loss_ce: 0.0076 aux.acc_seg: 99.3462 +04/18 11:50:30 - mmengine - INFO - Iter(train) [ 61650/160000] lr: 6.4889e-03 eta: 15:04:34 time: 0.5510 data_time: 0.0064 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0071 decode.acc_seg: 99.7243 aux.loss_ce: 0.0072 aux.acc_seg: 99.2740 +04/18 11:50:58 - mmengine - INFO - Iter(train) [ 61700/160000] lr: 6.4860e-03 eta: 15:04:06 time: 0.5504 data_time: 0.0065 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0073 decode.acc_seg: 99.7153 aux.loss_ce: 0.0074 aux.acc_seg: 99.3555 +04/18 11:51:25 - mmengine - INFO - Iter(train) [ 61750/160000] lr: 6.4831e-03 eta: 15:03:38 time: 0.5530 data_time: 0.0072 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0079 decode.acc_seg: 99.7630 aux.loss_ce: 0.0078 aux.acc_seg: 99.3668 +04/18 11:51:53 - mmengine - INFO - Iter(train) [ 61800/160000] lr: 6.4801e-03 eta: 15:03:11 time: 0.5507 data_time: 0.0064 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0076 decode.acc_seg: 99.6725 aux.loss_ce: 0.0074 aux.acc_seg: 99.3766 +04/18 11:52:20 - mmengine - INFO - Iter(train) [ 61850/160000] lr: 6.4772e-03 eta: 15:02:43 time: 0.5516 data_time: 0.0063 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0080 decode.acc_seg: 99.6899 aux.loss_ce: 0.0077 aux.acc_seg: 99.1867 +04/18 11:52:48 - mmengine - INFO - Iter(train) [ 61900/160000] lr: 6.4743e-03 eta: 15:02:15 time: 0.5504 data_time: 0.0065 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0074 decode.acc_seg: 99.7381 aux.loss_ce: 0.0077 aux.acc_seg: 99.1766 +04/18 11:53:16 - mmengine - INFO - Iter(train) [ 61950/160000] lr: 6.4714e-03 eta: 15:01:48 time: 0.5504 data_time: 0.0066 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0075 decode.acc_seg: 99.7315 aux.loss_ce: 0.0076 aux.acc_seg: 99.3332 +04/18 11:53:43 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 11:53:43 - mmengine - INFO - Iter(train) [ 62000/160000] lr: 6.4684e-03 eta: 15:01:20 time: 0.5514 data_time: 0.0063 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0076 decode.acc_seg: 99.7339 aux.loss_ce: 0.0075 aux.acc_seg: 99.2385 +04/18 11:54:11 - mmengine - INFO - Iter(train) [ 62050/160000] lr: 6.4655e-03 eta: 15:00:52 time: 0.5513 data_time: 0.0068 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0072 decode.acc_seg: 99.7033 aux.loss_ce: 0.0076 aux.acc_seg: 98.9886 +04/18 11:54:38 - mmengine - INFO - Iter(train) [ 62100/160000] lr: 6.4626e-03 eta: 15:00:25 time: 0.5516 data_time: 0.0063 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0076 decode.acc_seg: 99.6926 aux.loss_ce: 0.0080 aux.acc_seg: 99.2916 +04/18 11:55:06 - mmengine - INFO - Iter(train) [ 62150/160000] lr: 6.4597e-03 eta: 14:59:57 time: 0.5506 data_time: 0.0072 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0069 decode.acc_seg: 99.7195 aux.loss_ce: 0.0069 aux.acc_seg: 99.2236 +04/18 11:55:33 - mmengine - INFO - Iter(train) [ 62200/160000] lr: 6.4567e-03 eta: 14:59:30 time: 0.5500 data_time: 0.0061 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0077 decode.acc_seg: 99.5876 aux.loss_ce: 0.0076 aux.acc_seg: 99.0355 +04/18 11:56:01 - mmengine - INFO - Iter(train) [ 62250/160000] lr: 6.4538e-03 eta: 14:59:02 time: 0.5513 data_time: 0.0061 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0088 decode.acc_seg: 99.6231 aux.loss_ce: 0.0077 aux.acc_seg: 99.2679 +04/18 11:56:29 - mmengine - INFO - Iter(train) [ 62300/160000] lr: 6.4509e-03 eta: 14:58:34 time: 0.5528 data_time: 0.0058 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0092 decode.acc_seg: 99.6235 aux.loss_ce: 0.0084 aux.acc_seg: 99.0924 +04/18 11:56:56 - mmengine - INFO - Iter(train) [ 62350/160000] lr: 6.4480e-03 eta: 14:58:07 time: 0.5515 data_time: 0.0059 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0086 decode.acc_seg: 99.6250 aux.loss_ce: 0.0080 aux.acc_seg: 99.3446 +04/18 11:57:24 - mmengine - INFO - Iter(train) [ 62400/160000] lr: 6.4450e-03 eta: 14:57:39 time: 0.5521 data_time: 0.0061 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0075 decode.acc_seg: 99.7289 aux.loss_ce: 0.0074 aux.acc_seg: 99.4224 +04/18 11:57:51 - mmengine - INFO - Iter(train) [ 62450/160000] lr: 6.4421e-03 eta: 14:57:11 time: 0.5504 data_time: 0.0065 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0078 decode.acc_seg: 99.7499 aux.loss_ce: 0.0076 aux.acc_seg: 99.4639 +04/18 11:58:19 - mmengine - INFO - Iter(train) [ 62500/160000] lr: 6.4392e-03 eta: 14:56:44 time: 0.5516 data_time: 0.0067 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0085 decode.acc_seg: 99.6426 aux.loss_ce: 0.0079 aux.acc_seg: 99.1708 +04/18 11:58:46 - mmengine - INFO - Iter(train) [ 62550/160000] lr: 6.4363e-03 eta: 14:56:16 time: 0.5524 data_time: 0.0063 memory: 7635 loss: 0.0197 decode.loss_ce: 0.0105 decode.acc_seg: 99.4515 aux.loss_ce: 0.0091 aux.acc_seg: 98.9024 +04/18 11:59:14 - mmengine - INFO - Iter(train) [ 62600/160000] lr: 6.4333e-03 eta: 14:55:48 time: 0.5528 data_time: 0.0059 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0091 decode.acc_seg: 99.5965 aux.loss_ce: 0.0083 aux.acc_seg: 99.0153 +04/18 11:59:42 - mmengine - INFO - Iter(train) [ 62650/160000] lr: 6.4304e-03 eta: 14:55:21 time: 0.5504 data_time: 0.0068 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0095 decode.acc_seg: 99.6124 aux.loss_ce: 0.0084 aux.acc_seg: 99.0512 +04/18 12:00:09 - mmengine - INFO - Iter(train) [ 62700/160000] lr: 6.4275e-03 eta: 14:54:53 time: 0.5606 data_time: 0.0062 memory: 7635 loss: 0.0266 decode.loss_ce: 0.0156 decode.acc_seg: 99.2701 aux.loss_ce: 0.0110 aux.acc_seg: 98.7589 +04/18 12:00:37 - mmengine - INFO - Iter(train) [ 62750/160000] lr: 6.4246e-03 eta: 14:54:26 time: 0.5517 data_time: 0.0062 memory: 7635 loss: 0.0219 decode.loss_ce: 0.0124 decode.acc_seg: 99.6608 aux.loss_ce: 0.0095 aux.acc_seg: 99.0810 +04/18 12:01:05 - mmengine - INFO - Iter(train) [ 62800/160000] lr: 6.4216e-03 eta: 14:53:58 time: 0.5515 data_time: 0.0065 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0095 decode.acc_seg: 99.4710 aux.loss_ce: 0.0087 aux.acc_seg: 98.8008 +04/18 12:01:32 - mmengine - INFO - Iter(train) [ 62850/160000] lr: 6.4187e-03 eta: 14:53:31 time: 0.5511 data_time: 0.0070 memory: 7635 loss: 0.0202 decode.loss_ce: 0.0111 decode.acc_seg: 99.6174 aux.loss_ce: 0.0090 aux.acc_seg: 99.0654 +04/18 12:02:00 - mmengine - INFO - Iter(train) [ 62900/160000] lr: 6.4158e-03 eta: 14:53:03 time: 0.5524 data_time: 0.0069 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0085 decode.acc_seg: 99.6057 aux.loss_ce: 0.0076 aux.acc_seg: 99.3256 +04/18 12:02:27 - mmengine - INFO - Iter(train) [ 62950/160000] lr: 6.4129e-03 eta: 14:52:35 time: 0.5521 data_time: 0.0063 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0080 decode.acc_seg: 99.6496 aux.loss_ce: 0.0073 aux.acc_seg: 99.2011 +04/18 12:02:55 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 12:02:55 - mmengine - INFO - Iter(train) [ 63000/160000] lr: 6.4099e-03 eta: 14:52:08 time: 0.5515 data_time: 0.0060 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0096 decode.acc_seg: 99.3012 aux.loss_ce: 0.0087 aux.acc_seg: 98.7721 +04/18 12:03:22 - mmengine - INFO - Iter(train) [ 63050/160000] lr: 6.4070e-03 eta: 14:51:40 time: 0.5515 data_time: 0.0058 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0083 decode.acc_seg: 99.7809 aux.loss_ce: 0.0079 aux.acc_seg: 99.3269 +04/18 12:03:50 - mmengine - INFO - Iter(train) [ 63100/160000] lr: 6.4041e-03 eta: 14:51:13 time: 0.5504 data_time: 0.0068 memory: 7635 loss: 0.0194 decode.loss_ce: 0.0105 decode.acc_seg: 99.6111 aux.loss_ce: 0.0090 aux.acc_seg: 99.1572 +04/18 12:04:18 - mmengine - INFO - Iter(train) [ 63150/160000] lr: 6.4011e-03 eta: 14:50:45 time: 0.5515 data_time: 0.0062 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0090 decode.acc_seg: 99.7617 aux.loss_ce: 0.0079 aux.acc_seg: 99.3432 +04/18 12:04:45 - mmengine - INFO - Iter(train) [ 63200/160000] lr: 6.3982e-03 eta: 14:50:17 time: 0.5522 data_time: 0.0072 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0086 decode.acc_seg: 99.6676 aux.loss_ce: 0.0078 aux.acc_seg: 99.3055 +04/18 12:05:13 - mmengine - INFO - Iter(train) [ 63250/160000] lr: 6.3953e-03 eta: 14:49:50 time: 0.5516 data_time: 0.0068 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0089 decode.acc_seg: 99.5938 aux.loss_ce: 0.0084 aux.acc_seg: 99.3444 +04/18 12:05:40 - mmengine - INFO - Iter(train) [ 63300/160000] lr: 6.3924e-03 eta: 14:49:22 time: 0.5525 data_time: 0.0065 memory: 7635 loss: 0.0233 decode.loss_ce: 0.0133 decode.acc_seg: 99.6986 aux.loss_ce: 0.0101 aux.acc_seg: 99.2738 +04/18 12:06:08 - mmengine - INFO - Iter(train) [ 63350/160000] lr: 6.3894e-03 eta: 14:48:55 time: 0.5510 data_time: 0.0069 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0102 decode.acc_seg: 99.7083 aux.loss_ce: 0.0086 aux.acc_seg: 99.1220 +04/18 12:06:36 - mmengine - INFO - Iter(train) [ 63400/160000] lr: 6.3865e-03 eta: 14:48:27 time: 0.5513 data_time: 0.0067 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0087 decode.acc_seg: 99.7265 aux.loss_ce: 0.0080 aux.acc_seg: 99.3199 +04/18 12:07:03 - mmengine - INFO - Iter(train) [ 63450/160000] lr: 6.3836e-03 eta: 14:47:59 time: 0.5518 data_time: 0.0062 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0085 decode.acc_seg: 99.7110 aux.loss_ce: 0.0073 aux.acc_seg: 99.2965 +04/18 12:07:31 - mmengine - INFO - Iter(train) [ 63500/160000] lr: 6.3806e-03 eta: 14:47:32 time: 0.5521 data_time: 0.0063 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0083 decode.acc_seg: 99.6314 aux.loss_ce: 0.0081 aux.acc_seg: 99.1665 +04/18 12:07:58 - mmengine - INFO - Iter(train) [ 63550/160000] lr: 6.3777e-03 eta: 14:47:04 time: 0.5510 data_time: 0.0071 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0080 decode.acc_seg: 99.6786 aux.loss_ce: 0.0080 aux.acc_seg: 99.1355 +04/18 12:08:26 - mmengine - INFO - Iter(train) [ 63600/160000] lr: 6.3748e-03 eta: 14:46:36 time: 0.5505 data_time: 0.0071 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0088 decode.acc_seg: 99.6197 aux.loss_ce: 0.0082 aux.acc_seg: 99.1724 +04/18 12:08:53 - mmengine - INFO - Iter(train) [ 63650/160000] lr: 6.3719e-03 eta: 14:46:09 time: 0.5515 data_time: 0.0068 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0091 decode.acc_seg: 99.6713 aux.loss_ce: 0.0087 aux.acc_seg: 98.9584 +04/18 12:09:21 - mmengine - INFO - Iter(train) [ 63700/160000] lr: 6.3689e-03 eta: 14:45:41 time: 0.5517 data_time: 0.0060 memory: 7635 loss: 0.0217 decode.loss_ce: 0.0119 decode.acc_seg: 99.4674 aux.loss_ce: 0.0098 aux.acc_seg: 98.8377 +04/18 12:09:49 - mmengine - INFO - Iter(train) [ 63750/160000] lr: 6.3660e-03 eta: 14:45:14 time: 0.5514 data_time: 0.0064 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0092 decode.acc_seg: 99.5516 aux.loss_ce: 0.0085 aux.acc_seg: 99.1558 +04/18 12:10:16 - mmengine - INFO - Iter(train) [ 63800/160000] lr: 6.3631e-03 eta: 14:44:46 time: 0.5507 data_time: 0.0062 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0090 decode.acc_seg: 99.6948 aux.loss_ce: 0.0082 aux.acc_seg: 99.1970 +04/18 12:10:44 - mmengine - INFO - Iter(train) [ 63850/160000] lr: 6.3601e-03 eta: 14:44:19 time: 0.5519 data_time: 0.0061 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0087 decode.acc_seg: 99.7017 aux.loss_ce: 0.0080 aux.acc_seg: 99.1553 +04/18 12:11:12 - mmengine - INFO - Iter(train) [ 63900/160000] lr: 6.3572e-03 eta: 14:43:51 time: 0.5513 data_time: 0.0065 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0076 decode.acc_seg: 99.7107 aux.loss_ce: 0.0073 aux.acc_seg: 99.2823 +04/18 12:11:39 - mmengine - INFO - Iter(train) [ 63950/160000] lr: 6.3543e-03 eta: 14:43:23 time: 0.5515 data_time: 0.0067 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0087 decode.acc_seg: 99.6617 aux.loss_ce: 0.0082 aux.acc_seg: 99.1934 +04/18 12:12:07 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 12:12:07 - mmengine - INFO - Iter(train) [ 64000/160000] lr: 6.3513e-03 eta: 14:42:56 time: 0.5514 data_time: 0.0070 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0083 decode.acc_seg: 99.6565 aux.loss_ce: 0.0076 aux.acc_seg: 99.2448 +04/18 12:12:34 - mmengine - INFO - Iter(train) [ 64050/160000] lr: 6.3484e-03 eta: 14:42:28 time: 0.5506 data_time: 0.0063 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0069 decode.acc_seg: 99.6697 aux.loss_ce: 0.0071 aux.acc_seg: 99.0233 +04/18 12:13:02 - mmengine - INFO - Iter(train) [ 64100/160000] lr: 6.3455e-03 eta: 14:42:01 time: 0.5509 data_time: 0.0069 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0088 decode.acc_seg: 99.6822 aux.loss_ce: 0.0085 aux.acc_seg: 99.2085 +04/18 12:13:29 - mmengine - INFO - Iter(train) [ 64150/160000] lr: 6.3426e-03 eta: 14:41:33 time: 0.5498 data_time: 0.0063 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0089 decode.acc_seg: 99.5502 aux.loss_ce: 0.0081 aux.acc_seg: 98.8747 +04/18 12:13:57 - mmengine - INFO - Iter(train) [ 64200/160000] lr: 6.3396e-03 eta: 14:41:05 time: 0.5510 data_time: 0.0062 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0083 decode.acc_seg: 99.6749 aux.loss_ce: 0.0078 aux.acc_seg: 99.2393 +04/18 12:14:25 - mmengine - INFO - Iter(train) [ 64250/160000] lr: 6.3367e-03 eta: 14:40:38 time: 0.5511 data_time: 0.0065 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0082 decode.acc_seg: 99.6074 aux.loss_ce: 0.0074 aux.acc_seg: 99.1533 +04/18 12:14:52 - mmengine - INFO - Iter(train) [ 64300/160000] lr: 6.3338e-03 eta: 14:40:10 time: 0.5516 data_time: 0.0059 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0085 decode.acc_seg: 99.5456 aux.loss_ce: 0.0084 aux.acc_seg: 98.9036 +04/18 12:15:20 - mmengine - INFO - Iter(train) [ 64350/160000] lr: 6.3308e-03 eta: 14:39:42 time: 0.5512 data_time: 0.0066 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0098 decode.acc_seg: 99.5765 aux.loss_ce: 0.0088 aux.acc_seg: 99.2641 +04/18 12:15:47 - mmengine - INFO - Iter(train) [ 64400/160000] lr: 6.3279e-03 eta: 14:39:15 time: 0.5505 data_time: 0.0061 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0083 decode.acc_seg: 99.6955 aux.loss_ce: 0.0080 aux.acc_seg: 99.3121 +04/18 12:16:15 - mmengine - INFO - Iter(train) [ 64450/160000] lr: 6.3250e-03 eta: 14:38:47 time: 0.5507 data_time: 0.0063 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0093 decode.acc_seg: 99.6198 aux.loss_ce: 0.0082 aux.acc_seg: 99.0871 +04/18 12:16:42 - mmengine - INFO - Iter(train) [ 64500/160000] lr: 6.3220e-03 eta: 14:38:19 time: 0.5518 data_time: 0.0064 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0089 decode.acc_seg: 99.5845 aux.loss_ce: 0.0082 aux.acc_seg: 99.1028 +04/18 12:17:10 - mmengine - INFO - Iter(train) [ 64550/160000] lr: 6.3191e-03 eta: 14:37:52 time: 0.5518 data_time: 0.0070 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0079 decode.acc_seg: 99.6962 aux.loss_ce: 0.0076 aux.acc_seg: 99.2999 +04/18 12:17:38 - mmengine - INFO - Iter(train) [ 64600/160000] lr: 6.3162e-03 eta: 14:37:24 time: 0.5527 data_time: 0.0064 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0080 decode.acc_seg: 99.6766 aux.loss_ce: 0.0077 aux.acc_seg: 99.3655 +04/18 12:18:05 - mmengine - INFO - Iter(train) [ 64650/160000] lr: 6.3132e-03 eta: 14:36:57 time: 0.5502 data_time: 0.0060 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0097 decode.acc_seg: 99.5746 aux.loss_ce: 0.0085 aux.acc_seg: 99.2217 +04/18 12:18:33 - mmengine - INFO - Iter(train) [ 64700/160000] lr: 6.3103e-03 eta: 14:36:29 time: 0.5505 data_time: 0.0072 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0092 decode.acc_seg: 99.6564 aux.loss_ce: 0.0079 aux.acc_seg: 99.1640 +04/18 12:19:00 - mmengine - INFO - Iter(train) [ 64750/160000] lr: 6.3074e-03 eta: 14:36:01 time: 0.5514 data_time: 0.0070 memory: 7635 loss: 0.0200 decode.loss_ce: 0.0112 decode.acc_seg: 99.6309 aux.loss_ce: 0.0088 aux.acc_seg: 99.1497 +04/18 12:19:28 - mmengine - INFO - Iter(train) [ 64800/160000] lr: 6.3044e-03 eta: 14:35:34 time: 0.5519 data_time: 0.0070 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0095 decode.acc_seg: 99.6595 aux.loss_ce: 0.0085 aux.acc_seg: 99.1340 +04/18 12:19:56 - mmengine - INFO - Iter(train) [ 64850/160000] lr: 6.3015e-03 eta: 14:35:06 time: 0.5502 data_time: 0.0066 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0092 decode.acc_seg: 99.6979 aux.loss_ce: 0.0080 aux.acc_seg: 99.3130 +04/18 12:20:23 - mmengine - INFO - Iter(train) [ 64900/160000] lr: 6.2986e-03 eta: 14:34:39 time: 0.5522 data_time: 0.0070 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0076 decode.acc_seg: 99.6875 aux.loss_ce: 0.0076 aux.acc_seg: 99.2385 +04/18 12:20:51 - mmengine - INFO - Iter(train) [ 64950/160000] lr: 6.2956e-03 eta: 14:34:11 time: 0.5515 data_time: 0.0067 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0079 decode.acc_seg: 99.7477 aux.loss_ce: 0.0079 aux.acc_seg: 99.1247 +04/18 12:21:18 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 12:21:18 - mmengine - INFO - Iter(train) [ 65000/160000] lr: 6.2927e-03 eta: 14:33:44 time: 0.5523 data_time: 0.0066 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0072 decode.acc_seg: 99.7310 aux.loss_ce: 0.0074 aux.acc_seg: 99.1592 +04/18 12:21:46 - mmengine - INFO - Iter(train) [ 65050/160000] lr: 6.2898e-03 eta: 14:33:16 time: 0.5518 data_time: 0.0065 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0079 decode.acc_seg: 99.7298 aux.loss_ce: 0.0075 aux.acc_seg: 99.3653 +04/18 12:22:14 - mmengine - INFO - Iter(train) [ 65100/160000] lr: 6.2868e-03 eta: 14:32:48 time: 0.5513 data_time: 0.0059 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0080 decode.acc_seg: 99.7375 aux.loss_ce: 0.0079 aux.acc_seg: 99.2416 +04/18 12:22:41 - mmengine - INFO - Iter(train) [ 65150/160000] lr: 6.2839e-03 eta: 14:32:21 time: 0.5517 data_time: 0.0061 memory: 7635 loss: 0.0190 decode.loss_ce: 0.0100 decode.acc_seg: 99.6408 aux.loss_ce: 0.0090 aux.acc_seg: 99.1256 +04/18 12:23:09 - mmengine - INFO - Iter(train) [ 65200/160000] lr: 6.2810e-03 eta: 14:31:53 time: 0.5528 data_time: 0.0072 memory: 7635 loss: 0.0186 decode.loss_ce: 0.0099 decode.acc_seg: 99.6358 aux.loss_ce: 0.0087 aux.acc_seg: 99.0165 +04/18 12:23:36 - mmengine - INFO - Iter(train) [ 65250/160000] lr: 6.2780e-03 eta: 14:31:25 time: 0.5512 data_time: 0.0059 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0079 decode.acc_seg: 99.6083 aux.loss_ce: 0.0078 aux.acc_seg: 99.0692 +04/18 12:24:04 - mmengine - INFO - Iter(train) [ 65300/160000] lr: 6.2751e-03 eta: 14:30:58 time: 0.5505 data_time: 0.0066 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0078 decode.acc_seg: 99.7094 aux.loss_ce: 0.0080 aux.acc_seg: 99.0721 +04/18 12:24:31 - mmengine - INFO - Iter(train) [ 65350/160000] lr: 6.2722e-03 eta: 14:30:30 time: 0.5510 data_time: 0.0068 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0083 decode.acc_seg: 99.6728 aux.loss_ce: 0.0081 aux.acc_seg: 99.2450 +04/18 12:24:59 - mmengine - INFO - Iter(train) [ 65400/160000] lr: 6.2692e-03 eta: 14:30:02 time: 0.5513 data_time: 0.0064 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0086 decode.acc_seg: 99.6277 aux.loss_ce: 0.0084 aux.acc_seg: 98.9611 +04/18 12:25:27 - mmengine - INFO - Iter(train) [ 65450/160000] lr: 6.2663e-03 eta: 14:29:35 time: 0.5513 data_time: 0.0062 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0085 decode.acc_seg: 99.6113 aux.loss_ce: 0.0079 aux.acc_seg: 99.1420 +04/18 12:25:54 - mmengine - INFO - Iter(train) [ 65500/160000] lr: 6.2634e-03 eta: 14:29:07 time: 0.5505 data_time: 0.0062 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0077 decode.acc_seg: 99.6889 aux.loss_ce: 0.0072 aux.acc_seg: 99.1988 +04/18 12:26:22 - mmengine - INFO - Iter(train) [ 65550/160000] lr: 6.2604e-03 eta: 14:28:40 time: 0.5523 data_time: 0.0074 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0092 decode.acc_seg: 99.7055 aux.loss_ce: 0.0087 aux.acc_seg: 99.0686 +04/18 12:26:49 - mmengine - INFO - Iter(train) [ 65600/160000] lr: 6.2575e-03 eta: 14:28:12 time: 0.5511 data_time: 0.0068 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0077 decode.acc_seg: 99.6802 aux.loss_ce: 0.0076 aux.acc_seg: 99.2186 +04/18 12:27:17 - mmengine - INFO - Iter(train) [ 65650/160000] lr: 6.2546e-03 eta: 14:27:44 time: 0.5513 data_time: 0.0063 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0075 decode.acc_seg: 99.6564 aux.loss_ce: 0.0070 aux.acc_seg: 99.3765 +04/18 12:27:44 - mmengine - INFO - Iter(train) [ 65700/160000] lr: 6.2516e-03 eta: 14:27:17 time: 0.5525 data_time: 0.0064 memory: 7635 loss: 0.0202 decode.loss_ce: 0.0115 decode.acc_seg: 99.6966 aux.loss_ce: 0.0087 aux.acc_seg: 99.4294 +04/18 12:28:12 - mmengine - INFO - Iter(train) [ 65750/160000] lr: 6.2487e-03 eta: 14:26:49 time: 0.5521 data_time: 0.0067 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0083 decode.acc_seg: 99.5690 aux.loss_ce: 0.0076 aux.acc_seg: 99.0780 +04/18 12:28:40 - mmengine - INFO - Iter(train) [ 65800/160000] lr: 6.2458e-03 eta: 14:26:21 time: 0.5508 data_time: 0.0065 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0088 decode.acc_seg: 99.5194 aux.loss_ce: 0.0080 aux.acc_seg: 98.9784 +04/18 12:29:07 - mmengine - INFO - Iter(train) [ 65850/160000] lr: 6.2428e-03 eta: 14:25:54 time: 0.5507 data_time: 0.0065 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0080 decode.acc_seg: 99.7318 aux.loss_ce: 0.0073 aux.acc_seg: 99.1998 +04/18 12:29:35 - mmengine - INFO - Iter(train) [ 65900/160000] lr: 6.2399e-03 eta: 14:25:26 time: 0.5510 data_time: 0.0058 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0091 decode.acc_seg: 99.7058 aux.loss_ce: 0.0079 aux.acc_seg: 99.3279 +04/18 12:30:03 - mmengine - INFO - Iter(train) [ 65950/160000] lr: 6.2369e-03 eta: 14:24:59 time: 0.5516 data_time: 0.0063 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0092 decode.acc_seg: 99.6469 aux.loss_ce: 0.0085 aux.acc_seg: 98.9905 +04/18 12:30:30 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 12:30:30 - mmengine - INFO - Iter(train) [ 66000/160000] lr: 6.2340e-03 eta: 14:24:32 time: 0.5514 data_time: 0.0080 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0083 decode.acc_seg: 99.7491 aux.loss_ce: 0.0081 aux.acc_seg: 99.3939 +04/18 12:30:58 - mmengine - INFO - Iter(train) [ 66050/160000] lr: 6.2311e-03 eta: 14:24:04 time: 0.5508 data_time: 0.0069 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0074 decode.acc_seg: 99.7186 aux.loss_ce: 0.0075 aux.acc_seg: 99.1418 +04/18 12:31:25 - mmengine - INFO - Iter(train) [ 66100/160000] lr: 6.2281e-03 eta: 14:23:36 time: 0.5509 data_time: 0.0061 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0079 decode.acc_seg: 99.6423 aux.loss_ce: 0.0074 aux.acc_seg: 99.1426 +04/18 12:31:53 - mmengine - INFO - Iter(train) [ 66150/160000] lr: 6.2252e-03 eta: 14:23:09 time: 0.5523 data_time: 0.0065 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0079 decode.acc_seg: 99.7011 aux.loss_ce: 0.0078 aux.acc_seg: 99.2793 +04/18 12:32:21 - mmengine - INFO - Iter(train) [ 66200/160000] lr: 6.2223e-03 eta: 14:22:41 time: 0.5513 data_time: 0.0070 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0076 decode.acc_seg: 99.7555 aux.loss_ce: 0.0079 aux.acc_seg: 99.2439 +04/18 12:32:48 - mmengine - INFO - Iter(train) [ 66250/160000] lr: 6.2193e-03 eta: 14:22:13 time: 0.5502 data_time: 0.0066 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0069 decode.acc_seg: 99.6030 aux.loss_ce: 0.0068 aux.acc_seg: 99.2440 +04/18 12:33:16 - mmengine - INFO - Iter(train) [ 66300/160000] lr: 6.2164e-03 eta: 14:21:46 time: 0.5508 data_time: 0.0065 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0076 decode.acc_seg: 99.7398 aux.loss_ce: 0.0076 aux.acc_seg: 99.4539 +04/18 12:33:43 - mmengine - INFO - Iter(train) [ 66350/160000] lr: 6.2135e-03 eta: 14:21:18 time: 0.5521 data_time: 0.0068 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0074 decode.acc_seg: 99.6774 aux.loss_ce: 0.0072 aux.acc_seg: 99.1939 +04/18 12:34:11 - mmengine - INFO - Iter(train) [ 66400/160000] lr: 6.2105e-03 eta: 14:20:50 time: 0.5514 data_time: 0.0061 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0078 decode.acc_seg: 99.6902 aux.loss_ce: 0.0073 aux.acc_seg: 99.3798 +04/18 12:34:38 - mmengine - INFO - Iter(train) [ 66450/160000] lr: 6.2076e-03 eta: 14:20:23 time: 0.5526 data_time: 0.0068 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0076 decode.acc_seg: 99.7005 aux.loss_ce: 0.0078 aux.acc_seg: 99.1853 +04/18 12:35:06 - mmengine - INFO - Iter(train) [ 66500/160000] lr: 6.2046e-03 eta: 14:19:55 time: 0.5516 data_time: 0.0062 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0069 decode.acc_seg: 99.8232 aux.loss_ce: 0.0075 aux.acc_seg: 99.3772 +04/18 12:35:33 - mmengine - INFO - Iter(train) [ 66550/160000] lr: 6.2017e-03 eta: 14:19:28 time: 0.5500 data_time: 0.0082 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0082 decode.acc_seg: 99.6459 aux.loss_ce: 0.0077 aux.acc_seg: 99.1036 +04/18 12:36:01 - mmengine - INFO - Iter(train) [ 66600/160000] lr: 6.1988e-03 eta: 14:19:00 time: 0.5517 data_time: 0.0067 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0073 decode.acc_seg: 99.6597 aux.loss_ce: 0.0074 aux.acc_seg: 99.0806 +04/18 12:36:29 - mmengine - INFO - Iter(train) [ 66650/160000] lr: 6.1958e-03 eta: 14:18:32 time: 0.5512 data_time: 0.0065 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0075 decode.acc_seg: 99.6656 aux.loss_ce: 0.0074 aux.acc_seg: 99.1944 +04/18 12:36:56 - mmengine - INFO - Iter(train) [ 66700/160000] lr: 6.1929e-03 eta: 14:18:05 time: 0.5526 data_time: 0.0072 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0074 decode.acc_seg: 99.6866 aux.loss_ce: 0.0073 aux.acc_seg: 99.2916 +04/18 12:37:24 - mmengine - INFO - Iter(train) [ 66750/160000] lr: 6.1899e-03 eta: 14:17:37 time: 0.5517 data_time: 0.0062 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0076 decode.acc_seg: 99.7363 aux.loss_ce: 0.0079 aux.acc_seg: 99.0746 +04/18 12:37:51 - mmengine - INFO - Iter(train) [ 66800/160000] lr: 6.1870e-03 eta: 14:17:09 time: 0.5501 data_time: 0.0062 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0076 decode.acc_seg: 99.5464 aux.loss_ce: 0.0076 aux.acc_seg: 98.8960 +04/18 12:38:19 - mmengine - INFO - Iter(train) [ 66850/160000] lr: 6.1841e-03 eta: 14:16:42 time: 0.5516 data_time: 0.0064 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0077 decode.acc_seg: 99.6633 aux.loss_ce: 0.0078 aux.acc_seg: 99.1499 +04/18 12:38:46 - mmengine - INFO - Iter(train) [ 66900/160000] lr: 6.1811e-03 eta: 14:16:14 time: 0.5518 data_time: 0.0062 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0073 decode.acc_seg: 99.6593 aux.loss_ce: 0.0074 aux.acc_seg: 99.3145 +04/18 12:39:14 - mmengine - INFO - Iter(train) [ 66950/160000] lr: 6.1782e-03 eta: 14:15:47 time: 0.5515 data_time: 0.0066 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0078 decode.acc_seg: 99.6510 aux.loss_ce: 0.0080 aux.acc_seg: 99.1521 +04/18 12:39:42 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 12:39:42 - mmengine - INFO - Iter(train) [ 67000/160000] lr: 6.1753e-03 eta: 14:15:19 time: 0.5611 data_time: 0.0062 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0076 decode.acc_seg: 99.6480 aux.loss_ce: 0.0075 aux.acc_seg: 99.1254 +04/18 12:40:09 - mmengine - INFO - Iter(train) [ 67050/160000] lr: 6.1723e-03 eta: 14:14:51 time: 0.5510 data_time: 0.0059 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0077 decode.acc_seg: 99.6650 aux.loss_ce: 0.0078 aux.acc_seg: 99.1677 +04/18 12:40:37 - mmengine - INFO - Iter(train) [ 67100/160000] lr: 6.1694e-03 eta: 14:14:24 time: 0.5521 data_time: 0.0060 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0072 decode.acc_seg: 99.6387 aux.loss_ce: 0.0073 aux.acc_seg: 99.0540 +04/18 12:41:05 - mmengine - INFO - Iter(train) [ 67150/160000] lr: 6.1664e-03 eta: 14:13:56 time: 0.5517 data_time: 0.0068 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0073 decode.acc_seg: 99.7343 aux.loss_ce: 0.0073 aux.acc_seg: 99.3545 +04/18 12:41:32 - mmengine - INFO - Iter(train) [ 67200/160000] lr: 6.1635e-03 eta: 14:13:29 time: 0.5520 data_time: 0.0076 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0074 decode.acc_seg: 99.6905 aux.loss_ce: 0.0075 aux.acc_seg: 99.4012 +04/18 12:42:00 - mmengine - INFO - Iter(train) [ 67250/160000] lr: 6.1606e-03 eta: 14:13:01 time: 0.5515 data_time: 0.0071 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0070 decode.acc_seg: 99.7496 aux.loss_ce: 0.0073 aux.acc_seg: 99.2722 +04/18 12:42:27 - mmengine - INFO - Iter(train) [ 67300/160000] lr: 6.1576e-03 eta: 14:12:33 time: 0.5520 data_time: 0.0066 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0068 decode.acc_seg: 99.7684 aux.loss_ce: 0.0069 aux.acc_seg: 99.2039 +04/18 12:42:55 - mmengine - INFO - Iter(train) [ 67350/160000] lr: 6.1547e-03 eta: 14:12:06 time: 0.5507 data_time: 0.0065 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0079 decode.acc_seg: 99.7545 aux.loss_ce: 0.0080 aux.acc_seg: 99.4624 +04/18 12:43:22 - mmengine - INFO - Iter(train) [ 67400/160000] lr: 6.1517e-03 eta: 14:11:38 time: 0.5511 data_time: 0.0068 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0090 decode.acc_seg: 99.6899 aux.loss_ce: 0.0077 aux.acc_seg: 99.2165 +04/18 12:43:50 - mmengine - INFO - Iter(train) [ 67450/160000] lr: 6.1488e-03 eta: 14:11:10 time: 0.5508 data_time: 0.0064 memory: 7635 loss: 0.0186 decode.loss_ce: 0.0099 decode.acc_seg: 99.5389 aux.loss_ce: 0.0087 aux.acc_seg: 98.9109 +04/18 12:44:17 - mmengine - INFO - Iter(train) [ 67500/160000] lr: 6.1458e-03 eta: 14:10:43 time: 0.5513 data_time: 0.0064 memory: 7635 loss: 0.0196 decode.loss_ce: 0.0107 decode.acc_seg: 99.4827 aux.loss_ce: 0.0089 aux.acc_seg: 99.0738 +04/18 12:44:45 - mmengine - INFO - Iter(train) [ 67550/160000] lr: 6.1429e-03 eta: 14:10:15 time: 0.5509 data_time: 0.0062 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0087 decode.acc_seg: 99.7127 aux.loss_ce: 0.0083 aux.acc_seg: 99.0096 +04/18 12:45:13 - mmengine - INFO - Iter(train) [ 67600/160000] lr: 6.1400e-03 eta: 14:09:47 time: 0.5525 data_time: 0.0065 memory: 7635 loss: 0.0190 decode.loss_ce: 0.0104 decode.acc_seg: 99.5878 aux.loss_ce: 0.0087 aux.acc_seg: 99.0252 +04/18 12:45:40 - mmengine - INFO - Iter(train) [ 67650/160000] lr: 6.1370e-03 eta: 14:09:20 time: 0.5486 data_time: 0.0066 memory: 7635 loss: 0.0181 decode.loss_ce: 0.0095 decode.acc_seg: 99.6152 aux.loss_ce: 0.0086 aux.acc_seg: 99.1368 +04/18 12:46:08 - mmengine - INFO - Iter(train) [ 67700/160000] lr: 6.1341e-03 eta: 14:08:52 time: 0.5500 data_time: 0.0062 memory: 7635 loss: 0.0204 decode.loss_ce: 0.0111 decode.acc_seg: 99.6368 aux.loss_ce: 0.0093 aux.acc_seg: 99.0959 +04/18 12:46:35 - mmengine - INFO - Iter(train) [ 67750/160000] lr: 6.1311e-03 eta: 14:08:24 time: 0.5519 data_time: 0.0063 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0088 decode.acc_seg: 99.6267 aux.loss_ce: 0.0080 aux.acc_seg: 99.2095 +04/18 12:47:03 - mmengine - INFO - Iter(train) [ 67800/160000] lr: 6.1282e-03 eta: 14:07:57 time: 0.5505 data_time: 0.0064 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0085 decode.acc_seg: 99.7315 aux.loss_ce: 0.0077 aux.acc_seg: 99.3533 +04/18 12:47:30 - mmengine - INFO - Iter(train) [ 67850/160000] lr: 6.1253e-03 eta: 14:07:29 time: 0.5505 data_time: 0.0065 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0083 decode.acc_seg: 99.6808 aux.loss_ce: 0.0076 aux.acc_seg: 99.2458 +04/18 12:47:58 - mmengine - INFO - Iter(train) [ 67900/160000] lr: 6.1223e-03 eta: 14:07:01 time: 0.5512 data_time: 0.0072 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0078 decode.acc_seg: 99.7409 aux.loss_ce: 0.0075 aux.acc_seg: 99.3134 +04/18 12:48:25 - mmengine - INFO - Iter(train) [ 67950/160000] lr: 6.1194e-03 eta: 14:06:34 time: 0.5510 data_time: 0.0067 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0079 decode.acc_seg: 99.6959 aux.loss_ce: 0.0076 aux.acc_seg: 99.3150 +04/18 12:48:53 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 12:48:53 - mmengine - INFO - Iter(train) [ 68000/160000] lr: 6.1164e-03 eta: 14:06:06 time: 0.5515 data_time: 0.0063 memory: 7635 loss: 0.0186 decode.loss_ce: 0.0099 decode.acc_seg: 99.6530 aux.loss_ce: 0.0086 aux.acc_seg: 99.1934 +04/18 12:49:21 - mmengine - INFO - Iter(train) [ 68050/160000] lr: 6.1135e-03 eta: 14:05:39 time: 0.5508 data_time: 0.0067 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0091 decode.acc_seg: 99.6723 aux.loss_ce: 0.0085 aux.acc_seg: 99.1469 +04/18 12:49:48 - mmengine - INFO - Iter(train) [ 68100/160000] lr: 6.1105e-03 eta: 14:05:11 time: 0.5508 data_time: 0.0060 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0084 decode.acc_seg: 99.6007 aux.loss_ce: 0.0081 aux.acc_seg: 98.8534 +04/18 12:50:16 - mmengine - INFO - Iter(train) [ 68150/160000] lr: 6.1076e-03 eta: 14:04:44 time: 0.5515 data_time: 0.0067 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0078 decode.acc_seg: 99.7796 aux.loss_ce: 0.0079 aux.acc_seg: 99.3534 +04/18 12:50:44 - mmengine - INFO - Iter(train) [ 68200/160000] lr: 6.1047e-03 eta: 14:04:16 time: 0.5523 data_time: 0.0074 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0079 decode.acc_seg: 99.7065 aux.loss_ce: 0.0078 aux.acc_seg: 99.1721 +04/18 12:51:11 - mmengine - INFO - Iter(train) [ 68250/160000] lr: 6.1017e-03 eta: 14:03:49 time: 0.5510 data_time: 0.0066 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0092 decode.acc_seg: 99.4453 aux.loss_ce: 0.0081 aux.acc_seg: 99.0548 +04/18 12:51:39 - mmengine - INFO - Iter(train) [ 68300/160000] lr: 6.0988e-03 eta: 14:03:21 time: 0.5502 data_time: 0.0070 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0082 decode.acc_seg: 99.5903 aux.loss_ce: 0.0077 aux.acc_seg: 98.9627 +04/18 12:52:06 - mmengine - INFO - Iter(train) [ 68350/160000] lr: 6.0958e-03 eta: 14:02:53 time: 0.5509 data_time: 0.0068 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0080 decode.acc_seg: 99.6009 aux.loss_ce: 0.0075 aux.acc_seg: 99.1240 +04/18 12:52:34 - mmengine - INFO - Iter(train) [ 68400/160000] lr: 6.0929e-03 eta: 14:02:26 time: 0.5503 data_time: 0.0070 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0078 decode.acc_seg: 99.6278 aux.loss_ce: 0.0075 aux.acc_seg: 99.0184 +04/18 12:53:01 - mmengine - INFO - Iter(train) [ 68450/160000] lr: 6.0899e-03 eta: 14:01:58 time: 0.5499 data_time: 0.0060 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0079 decode.acc_seg: 99.7849 aux.loss_ce: 0.0076 aux.acc_seg: 99.3752 +04/18 12:53:29 - mmengine - INFO - Iter(train) [ 68500/160000] lr: 6.0870e-03 eta: 14:01:30 time: 0.5518 data_time: 0.0061 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0074 decode.acc_seg: 99.6494 aux.loss_ce: 0.0074 aux.acc_seg: 99.0561 +04/18 12:53:57 - mmengine - INFO - Iter(train) [ 68550/160000] lr: 6.0840e-03 eta: 14:01:03 time: 0.5506 data_time: 0.0069 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0080 decode.acc_seg: 99.7048 aux.loss_ce: 0.0077 aux.acc_seg: 99.2647 +04/18 12:54:24 - mmengine - INFO - Iter(train) [ 68600/160000] lr: 6.0811e-03 eta: 14:00:35 time: 0.5503 data_time: 0.0062 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0075 decode.acc_seg: 99.6004 aux.loss_ce: 0.0073 aux.acc_seg: 99.1581 +04/18 12:54:52 - mmengine - INFO - Iter(train) [ 68650/160000] lr: 6.0782e-03 eta: 14:00:07 time: 0.5499 data_time: 0.0058 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0079 decode.acc_seg: 99.6217 aux.loss_ce: 0.0080 aux.acc_seg: 99.0794 +04/18 12:55:19 - mmengine - INFO - Iter(train) [ 68700/160000] lr: 6.0752e-03 eta: 13:59:40 time: 0.5519 data_time: 0.0070 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0076 decode.acc_seg: 99.7984 aux.loss_ce: 0.0076 aux.acc_seg: 99.4544 +04/18 12:55:47 - mmengine - INFO - Iter(train) [ 68750/160000] lr: 6.0723e-03 eta: 13:59:12 time: 0.5500 data_time: 0.0058 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0075 decode.acc_seg: 99.6758 aux.loss_ce: 0.0077 aux.acc_seg: 99.0948 +04/18 12:56:14 - mmengine - INFO - Iter(train) [ 68800/160000] lr: 6.0693e-03 eta: 13:58:44 time: 0.5506 data_time: 0.0061 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0075 decode.acc_seg: 99.6945 aux.loss_ce: 0.0074 aux.acc_seg: 99.2737 +04/18 12:56:42 - mmengine - INFO - Iter(train) [ 68850/160000] lr: 6.0664e-03 eta: 13:58:17 time: 0.5505 data_time: 0.0061 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0079 decode.acc_seg: 99.7359 aux.loss_ce: 0.0078 aux.acc_seg: 99.2151 +04/18 12:57:09 - mmengine - INFO - Iter(train) [ 68900/160000] lr: 6.0634e-03 eta: 13:57:49 time: 0.5498 data_time: 0.0059 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0069 decode.acc_seg: 99.7167 aux.loss_ce: 0.0072 aux.acc_seg: 99.2746 +04/18 12:57:37 - mmengine - INFO - Iter(train) [ 68950/160000] lr: 6.0605e-03 eta: 13:57:21 time: 0.5515 data_time: 0.0060 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0077 decode.acc_seg: 99.6468 aux.loss_ce: 0.0074 aux.acc_seg: 99.1611 +04/18 12:58:05 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 12:58:05 - mmengine - INFO - Iter(train) [ 69000/160000] lr: 6.0575e-03 eta: 13:56:54 time: 0.5516 data_time: 0.0062 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0075 decode.acc_seg: 99.6240 aux.loss_ce: 0.0073 aux.acc_seg: 99.1300 +04/18 12:58:32 - mmengine - INFO - Iter(train) [ 69050/160000] lr: 6.0546e-03 eta: 13:56:26 time: 0.5504 data_time: 0.0061 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0068 decode.acc_seg: 99.6925 aux.loss_ce: 0.0071 aux.acc_seg: 99.2249 +04/18 12:59:00 - mmengine - INFO - Iter(train) [ 69100/160000] lr: 6.0516e-03 eta: 13:55:59 time: 0.5507 data_time: 0.0063 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0093 decode.acc_seg: 99.6224 aux.loss_ce: 0.0081 aux.acc_seg: 99.2496 +04/18 12:59:27 - mmengine - INFO - Iter(train) [ 69150/160000] lr: 6.0487e-03 eta: 13:55:31 time: 0.5623 data_time: 0.0063 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0076 decode.acc_seg: 99.5975 aux.loss_ce: 0.0076 aux.acc_seg: 99.0157 +04/18 12:59:55 - mmengine - INFO - Iter(train) [ 69200/160000] lr: 6.0458e-03 eta: 13:55:04 time: 0.5602 data_time: 0.0068 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0082 decode.acc_seg: 99.7068 aux.loss_ce: 0.0080 aux.acc_seg: 99.0401 +04/18 13:00:23 - mmengine - INFO - Iter(train) [ 69250/160000] lr: 6.0428e-03 eta: 13:54:36 time: 0.5513 data_time: 0.0064 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0079 decode.acc_seg: 99.7402 aux.loss_ce: 0.0078 aux.acc_seg: 99.2205 +04/18 13:00:50 - mmengine - INFO - Iter(train) [ 69300/160000] lr: 6.0399e-03 eta: 13:54:09 time: 0.5508 data_time: 0.0065 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0078 decode.acc_seg: 99.7506 aux.loss_ce: 0.0077 aux.acc_seg: 99.2965 +04/18 13:01:18 - mmengine - INFO - Iter(train) [ 69350/160000] lr: 6.0369e-03 eta: 13:53:41 time: 0.5513 data_time: 0.0069 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0079 decode.acc_seg: 99.7013 aux.loss_ce: 0.0074 aux.acc_seg: 99.3340 +04/18 13:01:45 - mmengine - INFO - Iter(train) [ 69400/160000] lr: 6.0340e-03 eta: 13:53:13 time: 0.5512 data_time: 0.0065 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0077 decode.acc_seg: 99.6690 aux.loss_ce: 0.0073 aux.acc_seg: 99.2468 +04/18 13:02:13 - mmengine - INFO - Iter(train) [ 69450/160000] lr: 6.0310e-03 eta: 13:52:46 time: 0.5509 data_time: 0.0064 memory: 7635 loss: 0.0193 decode.loss_ce: 0.0105 decode.acc_seg: 99.6007 aux.loss_ce: 0.0089 aux.acc_seg: 99.1673 +04/18 13:02:41 - mmengine - INFO - Iter(train) [ 69500/160000] lr: 6.0281e-03 eta: 13:52:18 time: 0.5520 data_time: 0.0061 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0078 decode.acc_seg: 99.7236 aux.loss_ce: 0.0073 aux.acc_seg: 99.3158 +04/18 13:03:08 - mmengine - INFO - Iter(train) [ 69550/160000] lr: 6.0251e-03 eta: 13:51:50 time: 0.5511 data_time: 0.0061 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0079 decode.acc_seg: 99.6690 aux.loss_ce: 0.0074 aux.acc_seg: 99.1807 +04/18 13:03:36 - mmengine - INFO - Iter(train) [ 69600/160000] lr: 6.0222e-03 eta: 13:51:23 time: 0.5529 data_time: 0.0069 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0096 decode.acc_seg: 99.6646 aux.loss_ce: 0.0086 aux.acc_seg: 99.1276 +04/18 13:04:03 - mmengine - INFO - Iter(train) [ 69650/160000] lr: 6.0192e-03 eta: 13:50:55 time: 0.5506 data_time: 0.0065 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0081 decode.acc_seg: 99.6944 aux.loss_ce: 0.0079 aux.acc_seg: 99.4189 +04/18 13:04:31 - mmengine - INFO - Iter(train) [ 69700/160000] lr: 6.0163e-03 eta: 13:50:28 time: 0.5500 data_time: 0.0070 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0093 decode.acc_seg: 99.6805 aux.loss_ce: 0.0084 aux.acc_seg: 99.1603 +04/18 13:04:58 - mmengine - INFO - Iter(train) [ 69750/160000] lr: 6.0133e-03 eta: 13:50:00 time: 0.5520 data_time: 0.0065 memory: 7635 loss: 0.0203 decode.loss_ce: 0.0111 decode.acc_seg: 99.6716 aux.loss_ce: 0.0092 aux.acc_seg: 99.1631 +04/18 13:05:26 - mmengine - INFO - Iter(train) [ 69800/160000] lr: 6.0104e-03 eta: 13:49:32 time: 0.5505 data_time: 0.0066 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0096 decode.acc_seg: 99.5812 aux.loss_ce: 0.0087 aux.acc_seg: 99.0966 +04/18 13:05:54 - mmengine - INFO - Iter(train) [ 69850/160000] lr: 6.0074e-03 eta: 13:49:05 time: 0.5523 data_time: 0.0062 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0094 decode.acc_seg: 99.7193 aux.loss_ce: 0.0083 aux.acc_seg: 99.1334 +04/18 13:06:21 - mmengine - INFO - Iter(train) [ 69900/160000] lr: 6.0045e-03 eta: 13:48:37 time: 0.5517 data_time: 0.0065 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0082 decode.acc_seg: 99.6904 aux.loss_ce: 0.0080 aux.acc_seg: 99.2169 +04/18 13:06:49 - mmengine - INFO - Iter(train) [ 69950/160000] lr: 6.0015e-03 eta: 13:48:09 time: 0.5511 data_time: 0.0066 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0078 decode.acc_seg: 99.6339 aux.loss_ce: 0.0076 aux.acc_seg: 99.2690 +04/18 13:07:16 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 13:07:16 - mmengine - INFO - Iter(train) [ 70000/160000] lr: 5.9986e-03 eta: 13:47:42 time: 0.5509 data_time: 0.0063 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0068 decode.acc_seg: 99.7218 aux.loss_ce: 0.0071 aux.acc_seg: 99.2356 +04/18 13:07:16 - mmengine - INFO - Saving checkpoint at 70000 iterations +04/18 13:07:20 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:07 time: 0.0468 data_time: 0.0015 memory: 1657 +04/18 13:07:23 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:04 time: 0.0466 data_time: 0.0014 memory: 1657 +04/18 13:07:25 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:02 time: 0.0466 data_time: 0.0013 memory: 1657 +04/18 13:07:27 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.0458 data_time: 0.0012 memory: 1657 +04/18 13:07:28 - mmengine - INFO - per class results: +04/18 13:07:28 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 99.06 | 99.44 | 99.53 | 99.62 | 99.44 | +| contrast | 80.05 | 90.89 | 88.92 | 87.04 | 90.89 | ++------------+-------+-------+--------+-----------+--------+ +04/18 13:07:28 - mmengine - INFO - Iter(val) [200/200] aAcc: 99.1000 mIoU: 89.5600 mAcc: 95.1600 mFscore: 94.2200 mPrecision: 93.3300 mRecall: 95.1600 data_time: 0.0015 time: 0.0467 +04/18 13:07:55 - mmengine - INFO - Iter(train) [ 70050/160000] lr: 5.9956e-03 eta: 13:47:14 time: 0.5514 data_time: 0.0064 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0075 decode.acc_seg: 99.7017 aux.loss_ce: 0.0077 aux.acc_seg: 99.0495 +04/18 13:08:23 - mmengine - INFO - Iter(train) [ 70100/160000] lr: 5.9927e-03 eta: 13:46:47 time: 0.5523 data_time: 0.0077 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0080 decode.acc_seg: 99.7711 aux.loss_ce: 0.0075 aux.acc_seg: 99.3266 +04/18 13:08:50 - mmengine - INFO - Iter(train) [ 70150/160000] lr: 5.9897e-03 eta: 13:46:19 time: 0.5521 data_time: 0.0064 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0084 decode.acc_seg: 99.6188 aux.loss_ce: 0.0082 aux.acc_seg: 98.9286 +04/18 13:09:18 - mmengine - INFO - Iter(train) [ 70200/160000] lr: 5.9868e-03 eta: 13:45:52 time: 0.5494 data_time: 0.0061 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0080 decode.acc_seg: 99.6087 aux.loss_ce: 0.0078 aux.acc_seg: 99.0937 +04/18 13:09:45 - mmengine - INFO - Iter(train) [ 70250/160000] lr: 5.9838e-03 eta: 13:45:24 time: 0.5502 data_time: 0.0067 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0085 decode.acc_seg: 99.7354 aux.loss_ce: 0.0076 aux.acc_seg: 99.2040 +04/18 13:10:13 - mmengine - INFO - Iter(train) [ 70300/160000] lr: 5.9809e-03 eta: 13:44:57 time: 0.5513 data_time: 0.0067 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0083 decode.acc_seg: 99.6964 aux.loss_ce: 0.0078 aux.acc_seg: 99.2430 +04/18 13:10:41 - mmengine - INFO - Iter(train) [ 70350/160000] lr: 5.9779e-03 eta: 13:44:29 time: 0.5516 data_time: 0.0066 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0077 decode.acc_seg: 99.6499 aux.loss_ce: 0.0075 aux.acc_seg: 99.1297 +04/18 13:11:08 - mmengine - INFO - Iter(train) [ 70400/160000] lr: 5.9750e-03 eta: 13:44:01 time: 0.5504 data_time: 0.0066 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0077 decode.acc_seg: 99.6212 aux.loss_ce: 0.0075 aux.acc_seg: 99.3167 +04/18 13:11:36 - mmengine - INFO - Iter(train) [ 70450/160000] lr: 5.9720e-03 eta: 13:43:34 time: 0.5527 data_time: 0.0064 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0078 decode.acc_seg: 99.6600 aux.loss_ce: 0.0080 aux.acc_seg: 99.1551 +04/18 13:12:04 - mmengine - INFO - Iter(train) [ 70500/160000] lr: 5.9691e-03 eta: 13:43:06 time: 0.5520 data_time: 0.0066 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0079 decode.acc_seg: 99.7507 aux.loss_ce: 0.0075 aux.acc_seg: 99.5346 +04/18 13:12:31 - mmengine - INFO - Iter(train) [ 70550/160000] lr: 5.9661e-03 eta: 13:42:39 time: 0.5515 data_time: 0.0065 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0077 decode.acc_seg: 99.7468 aux.loss_ce: 0.0080 aux.acc_seg: 99.0608 +04/18 13:12:59 - mmengine - INFO - Iter(train) [ 70600/160000] lr: 5.9632e-03 eta: 13:42:11 time: 0.5515 data_time: 0.0069 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0076 decode.acc_seg: 99.6384 aux.loss_ce: 0.0078 aux.acc_seg: 98.9050 +04/18 13:13:26 - mmengine - INFO - Iter(train) [ 70650/160000] lr: 5.9602e-03 eta: 13:41:44 time: 0.5515 data_time: 0.0068 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0087 decode.acc_seg: 99.7053 aux.loss_ce: 0.0082 aux.acc_seg: 99.2291 +04/18 13:13:54 - mmengine - INFO - Iter(train) [ 70700/160000] lr: 5.9573e-03 eta: 13:41:16 time: 0.5520 data_time: 0.0071 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0079 decode.acc_seg: 99.7040 aux.loss_ce: 0.0077 aux.acc_seg: 99.1316 +04/18 13:14:22 - mmengine - INFO - Iter(train) [ 70750/160000] lr: 5.9543e-03 eta: 13:40:48 time: 0.5527 data_time: 0.0065 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0082 decode.acc_seg: 99.5953 aux.loss_ce: 0.0078 aux.acc_seg: 98.9807 +04/18 13:14:49 - mmengine - INFO - Iter(train) [ 70800/160000] lr: 5.9514e-03 eta: 13:40:21 time: 0.5515 data_time: 0.0060 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0077 decode.acc_seg: 99.6740 aux.loss_ce: 0.0073 aux.acc_seg: 99.1641 +04/18 13:15:17 - mmengine - INFO - Iter(train) [ 70850/160000] lr: 5.9484e-03 eta: 13:39:53 time: 0.5507 data_time: 0.0065 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0081 decode.acc_seg: 99.6813 aux.loss_ce: 0.0082 aux.acc_seg: 99.0903 +04/18 13:15:44 - mmengine - INFO - Iter(train) [ 70900/160000] lr: 5.9455e-03 eta: 13:39:26 time: 0.5528 data_time: 0.0061 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0090 decode.acc_seg: 99.5258 aux.loss_ce: 0.0084 aux.acc_seg: 98.8321 +04/18 13:16:12 - mmengine - INFO - Iter(train) [ 70950/160000] lr: 5.9425e-03 eta: 13:38:58 time: 0.5521 data_time: 0.0064 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0082 decode.acc_seg: 99.6739 aux.loss_ce: 0.0082 aux.acc_seg: 98.9123 +04/18 13:16:40 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 13:16:40 - mmengine - INFO - Iter(train) [ 71000/160000] lr: 5.9396e-03 eta: 13:38:30 time: 0.5512 data_time: 0.0060 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0083 decode.acc_seg: 99.5271 aux.loss_ce: 0.0075 aux.acc_seg: 99.0655 +04/18 13:17:07 - mmengine - INFO - Iter(train) [ 71050/160000] lr: 5.9366e-03 eta: 13:38:03 time: 0.5532 data_time: 0.0065 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0072 decode.acc_seg: 99.7773 aux.loss_ce: 0.0075 aux.acc_seg: 99.3194 +04/18 13:17:35 - mmengine - INFO - Iter(train) [ 71100/160000] lr: 5.9337e-03 eta: 13:37:35 time: 0.5505 data_time: 0.0062 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0082 decode.acc_seg: 99.6755 aux.loss_ce: 0.0078 aux.acc_seg: 99.1154 +04/18 13:18:02 - mmengine - INFO - Iter(train) [ 71150/160000] lr: 5.9307e-03 eta: 13:37:08 time: 0.5518 data_time: 0.0070 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0071 decode.acc_seg: 99.6461 aux.loss_ce: 0.0072 aux.acc_seg: 99.0748 +04/18 13:18:30 - mmengine - INFO - Iter(train) [ 71200/160000] lr: 5.9278e-03 eta: 13:36:40 time: 0.5517 data_time: 0.0068 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0080 decode.acc_seg: 99.7775 aux.loss_ce: 0.0081 aux.acc_seg: 99.2649 +04/18 13:18:58 - mmengine - INFO - Iter(train) [ 71250/160000] lr: 5.9248e-03 eta: 13:36:12 time: 0.5516 data_time: 0.0071 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0073 decode.acc_seg: 99.5831 aux.loss_ce: 0.0072 aux.acc_seg: 98.8583 +04/18 13:19:25 - mmengine - INFO - Iter(train) [ 71300/160000] lr: 5.9218e-03 eta: 13:35:45 time: 0.5531 data_time: 0.0069 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0074 decode.acc_seg: 99.6793 aux.loss_ce: 0.0074 aux.acc_seg: 99.3607 +04/18 13:19:53 - mmengine - INFO - Iter(train) [ 71350/160000] lr: 5.9189e-03 eta: 13:35:18 time: 0.5516 data_time: 0.0062 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0084 decode.acc_seg: 99.7224 aux.loss_ce: 0.0078 aux.acc_seg: 99.3318 +04/18 13:20:21 - mmengine - INFO - Iter(train) [ 71400/160000] lr: 5.9159e-03 eta: 13:34:50 time: 0.5523 data_time: 0.0062 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0087 decode.acc_seg: 99.6738 aux.loss_ce: 0.0081 aux.acc_seg: 99.1942 +04/18 13:20:48 - mmengine - INFO - Iter(train) [ 71450/160000] lr: 5.9130e-03 eta: 13:34:22 time: 0.5529 data_time: 0.0067 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0084 decode.acc_seg: 99.6629 aux.loss_ce: 0.0080 aux.acc_seg: 99.0391 +04/18 13:21:16 - mmengine - INFO - Iter(train) [ 71500/160000] lr: 5.9100e-03 eta: 13:33:55 time: 0.5531 data_time: 0.0060 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0072 decode.acc_seg: 99.7650 aux.loss_ce: 0.0074 aux.acc_seg: 99.3747 +04/18 13:21:44 - mmengine - INFO - Iter(train) [ 71550/160000] lr: 5.9071e-03 eta: 13:33:27 time: 0.5533 data_time: 0.0061 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0074 decode.acc_seg: 99.7768 aux.loss_ce: 0.0076 aux.acc_seg: 99.4062 +04/18 13:22:11 - mmengine - INFO - Iter(train) [ 71600/160000] lr: 5.9041e-03 eta: 13:33:00 time: 0.5510 data_time: 0.0062 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0070 decode.acc_seg: 99.7050 aux.loss_ce: 0.0072 aux.acc_seg: 99.2457 +04/18 13:22:39 - mmengine - INFO - Iter(train) [ 71650/160000] lr: 5.9012e-03 eta: 13:32:32 time: 0.5524 data_time: 0.0075 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0074 decode.acc_seg: 99.7500 aux.loss_ce: 0.0080 aux.acc_seg: 99.1837 +04/18 13:23:06 - mmengine - INFO - Iter(train) [ 71700/160000] lr: 5.8982e-03 eta: 13:32:05 time: 0.5519 data_time: 0.0063 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0083 decode.acc_seg: 99.7140 aux.loss_ce: 0.0079 aux.acc_seg: 99.2368 +04/18 13:23:34 - mmengine - INFO - Iter(train) [ 71750/160000] lr: 5.8953e-03 eta: 13:31:37 time: 0.5511 data_time: 0.0066 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0076 decode.acc_seg: 99.6075 aux.loss_ce: 0.0073 aux.acc_seg: 99.2095 +04/18 13:24:02 - mmengine - INFO - Iter(train) [ 71800/160000] lr: 5.8923e-03 eta: 13:31:10 time: 0.5530 data_time: 0.0067 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0082 decode.acc_seg: 99.6995 aux.loss_ce: 0.0074 aux.acc_seg: 99.3881 +04/18 13:24:29 - mmengine - INFO - Iter(train) [ 71850/160000] lr: 5.8893e-03 eta: 13:30:42 time: 0.5511 data_time: 0.0064 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0080 decode.acc_seg: 99.7274 aux.loss_ce: 0.0074 aux.acc_seg: 99.4112 +04/18 13:24:57 - mmengine - INFO - Iter(train) [ 71900/160000] lr: 5.8864e-03 eta: 13:30:14 time: 0.5520 data_time: 0.0061 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0079 decode.acc_seg: 99.5666 aux.loss_ce: 0.0079 aux.acc_seg: 98.9028 +04/18 13:25:25 - mmengine - INFO - Iter(train) [ 71950/160000] lr: 5.8834e-03 eta: 13:29:47 time: 0.5530 data_time: 0.0067 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0080 decode.acc_seg: 99.7374 aux.loss_ce: 0.0077 aux.acc_seg: 99.2367 +04/18 13:25:52 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 13:25:52 - mmengine - INFO - Iter(train) [ 72000/160000] lr: 5.8805e-03 eta: 13:29:19 time: 0.5529 data_time: 0.0064 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0085 decode.acc_seg: 99.7155 aux.loss_ce: 0.0083 aux.acc_seg: 99.0468 +04/18 13:26:20 - mmengine - INFO - Iter(train) [ 72050/160000] lr: 5.8775e-03 eta: 13:28:52 time: 0.5527 data_time: 0.0061 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0086 decode.acc_seg: 99.6279 aux.loss_ce: 0.0080 aux.acc_seg: 99.0722 +04/18 13:26:47 - mmengine - INFO - Iter(train) [ 72100/160000] lr: 5.8746e-03 eta: 13:28:24 time: 0.5514 data_time: 0.0068 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0080 decode.acc_seg: 99.6089 aux.loss_ce: 0.0074 aux.acc_seg: 99.1879 +04/18 13:27:15 - mmengine - INFO - Iter(train) [ 72150/160000] lr: 5.8716e-03 eta: 13:27:57 time: 0.5522 data_time: 0.0063 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0077 decode.acc_seg: 99.7412 aux.loss_ce: 0.0080 aux.acc_seg: 99.2281 +04/18 13:27:43 - mmengine - INFO - Iter(train) [ 72200/160000] lr: 5.8687e-03 eta: 13:27:29 time: 0.5529 data_time: 0.0064 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0080 decode.acc_seg: 99.5743 aux.loss_ce: 0.0078 aux.acc_seg: 99.1511 +04/18 13:28:10 - mmengine - INFO - Iter(train) [ 72250/160000] lr: 5.8657e-03 eta: 13:27:02 time: 0.5520 data_time: 0.0064 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0072 decode.acc_seg: 99.6676 aux.loss_ce: 0.0072 aux.acc_seg: 99.2157 +04/18 13:28:38 - mmengine - INFO - Iter(train) [ 72300/160000] lr: 5.8627e-03 eta: 13:26:34 time: 0.5533 data_time: 0.0060 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0079 decode.acc_seg: 99.6785 aux.loss_ce: 0.0072 aux.acc_seg: 99.3123 +04/18 13:29:06 - mmengine - INFO - Iter(train) [ 72350/160000] lr: 5.8598e-03 eta: 13:26:06 time: 0.5529 data_time: 0.0070 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0088 decode.acc_seg: 99.6458 aux.loss_ce: 0.0081 aux.acc_seg: 99.1272 +04/18 13:29:33 - mmengine - INFO - Iter(train) [ 72400/160000] lr: 5.8568e-03 eta: 13:25:39 time: 0.5621 data_time: 0.0071 memory: 7635 loss: 0.0186 decode.loss_ce: 0.0102 decode.acc_seg: 99.5074 aux.loss_ce: 0.0084 aux.acc_seg: 99.0326 +04/18 13:30:01 - mmengine - INFO - Iter(train) [ 72450/160000] lr: 5.8539e-03 eta: 13:25:12 time: 0.5519 data_time: 0.0063 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0079 decode.acc_seg: 99.6523 aux.loss_ce: 0.0077 aux.acc_seg: 99.0721 +04/18 13:30:29 - mmengine - INFO - Iter(train) [ 72500/160000] lr: 5.8509e-03 eta: 13:24:44 time: 0.5510 data_time: 0.0060 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0081 decode.acc_seg: 99.6679 aux.loss_ce: 0.0078 aux.acc_seg: 99.3586 +04/18 13:30:56 - mmengine - INFO - Iter(train) [ 72550/160000] lr: 5.8480e-03 eta: 13:24:16 time: 0.5515 data_time: 0.0061 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0067 decode.acc_seg: 99.7038 aux.loss_ce: 0.0071 aux.acc_seg: 99.1717 +04/18 13:31:24 - mmengine - INFO - Iter(train) [ 72600/160000] lr: 5.8450e-03 eta: 13:23:49 time: 0.5533 data_time: 0.0059 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0070 decode.acc_seg: 99.6857 aux.loss_ce: 0.0074 aux.acc_seg: 98.9161 +04/18 13:31:52 - mmengine - INFO - Iter(train) [ 72650/160000] lr: 5.8420e-03 eta: 13:23:21 time: 0.5540 data_time: 0.0071 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0081 decode.acc_seg: 99.7041 aux.loss_ce: 0.0080 aux.acc_seg: 99.2405 +04/18 13:32:19 - mmengine - INFO - Iter(train) [ 72700/160000] lr: 5.8391e-03 eta: 13:22:54 time: 0.5524 data_time: 0.0066 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0081 decode.acc_seg: 99.6450 aux.loss_ce: 0.0079 aux.acc_seg: 98.9261 +04/18 13:32:47 - mmengine - INFO - Iter(train) [ 72750/160000] lr: 5.8361e-03 eta: 13:22:26 time: 0.5525 data_time: 0.0069 memory: 7635 loss: 0.0207 decode.loss_ce: 0.0118 decode.acc_seg: 99.3946 aux.loss_ce: 0.0090 aux.acc_seg: 99.1690 +04/18 13:33:14 - mmengine - INFO - Iter(train) [ 72800/160000] lr: 5.8332e-03 eta: 13:21:59 time: 0.5519 data_time: 0.0058 memory: 7635 loss: 0.0190 decode.loss_ce: 0.0103 decode.acc_seg: 99.4988 aux.loss_ce: 0.0088 aux.acc_seg: 98.8207 +04/18 13:33:42 - mmengine - INFO - Iter(train) [ 72850/160000] lr: 5.8302e-03 eta: 13:21:31 time: 0.5530 data_time: 0.0068 memory: 7635 loss: 0.0201 decode.loss_ce: 0.0111 decode.acc_seg: 99.6811 aux.loss_ce: 0.0090 aux.acc_seg: 99.2533 +04/18 13:34:10 - mmengine - INFO - Iter(train) [ 72900/160000] lr: 5.8272e-03 eta: 13:21:04 time: 0.5532 data_time: 0.0060 memory: 7635 loss: 0.0194 decode.loss_ce: 0.0108 decode.acc_seg: 99.4739 aux.loss_ce: 0.0086 aux.acc_seg: 99.0827 +04/18 13:34:37 - mmengine - INFO - Iter(train) [ 72950/160000] lr: 5.8243e-03 eta: 13:20:36 time: 0.5532 data_time: 0.0072 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0092 decode.acc_seg: 99.6920 aux.loss_ce: 0.0086 aux.acc_seg: 99.1225 +04/18 13:35:05 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 13:35:05 - mmengine - INFO - Iter(train) [ 73000/160000] lr: 5.8213e-03 eta: 13:20:08 time: 0.5531 data_time: 0.0065 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0085 decode.acc_seg: 99.6268 aux.loss_ce: 0.0081 aux.acc_seg: 99.2789 +04/18 13:35:33 - mmengine - INFO - Iter(train) [ 73050/160000] lr: 5.8184e-03 eta: 13:19:41 time: 0.5532 data_time: 0.0064 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0084 decode.acc_seg: 99.6858 aux.loss_ce: 0.0081 aux.acc_seg: 99.2652 +04/18 13:36:00 - mmengine - INFO - Iter(train) [ 73100/160000] lr: 5.8154e-03 eta: 13:19:13 time: 0.5517 data_time: 0.0071 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0081 decode.acc_seg: 99.6793 aux.loss_ce: 0.0078 aux.acc_seg: 99.1027 +04/18 13:36:28 - mmengine - INFO - Iter(train) [ 73150/160000] lr: 5.8125e-03 eta: 13:18:46 time: 0.5520 data_time: 0.0069 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0081 decode.acc_seg: 99.5923 aux.loss_ce: 0.0077 aux.acc_seg: 99.0482 +04/18 13:36:55 - mmengine - INFO - Iter(train) [ 73200/160000] lr: 5.8095e-03 eta: 13:18:18 time: 0.5532 data_time: 0.0062 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0090 decode.acc_seg: 99.6848 aux.loss_ce: 0.0082 aux.acc_seg: 99.1988 +04/18 13:37:23 - mmengine - INFO - Iter(train) [ 73250/160000] lr: 5.8065e-03 eta: 13:17:51 time: 0.5534 data_time: 0.0063 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0086 decode.acc_seg: 99.7367 aux.loss_ce: 0.0079 aux.acc_seg: 99.3934 +04/18 13:37:51 - mmengine - INFO - Iter(train) [ 73300/160000] lr: 5.8036e-03 eta: 13:17:23 time: 0.5531 data_time: 0.0061 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0083 decode.acc_seg: 99.7201 aux.loss_ce: 0.0078 aux.acc_seg: 99.3034 +04/18 13:38:18 - mmengine - INFO - Iter(train) [ 73350/160000] lr: 5.8006e-03 eta: 13:16:56 time: 0.5530 data_time: 0.0072 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0079 decode.acc_seg: 99.7479 aux.loss_ce: 0.0075 aux.acc_seg: 99.3849 +04/18 13:38:46 - mmengine - INFO - Iter(train) [ 73400/160000] lr: 5.7976e-03 eta: 13:16:28 time: 0.5514 data_time: 0.0061 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0100 decode.acc_seg: 99.3410 aux.loss_ce: 0.0089 aux.acc_seg: 98.5963 +04/18 13:39:14 - mmengine - INFO - Iter(train) [ 73450/160000] lr: 5.7947e-03 eta: 13:16:01 time: 0.5611 data_time: 0.0071 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0088 decode.acc_seg: 99.7078 aux.loss_ce: 0.0079 aux.acc_seg: 99.0885 +04/18 13:39:41 - mmengine - INFO - Iter(train) [ 73500/160000] lr: 5.7917e-03 eta: 13:15:33 time: 0.5535 data_time: 0.0065 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0084 decode.acc_seg: 99.6415 aux.loss_ce: 0.0082 aux.acc_seg: 99.0867 +04/18 13:40:09 - mmengine - INFO - Iter(train) [ 73550/160000] lr: 5.7888e-03 eta: 13:15:05 time: 0.5513 data_time: 0.0058 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0078 decode.acc_seg: 99.6291 aux.loss_ce: 0.0073 aux.acc_seg: 99.2481 +04/18 13:40:37 - mmengine - INFO - Iter(train) [ 73600/160000] lr: 5.7858e-03 eta: 13:14:38 time: 0.5520 data_time: 0.0071 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0083 decode.acc_seg: 99.6764 aux.loss_ce: 0.0082 aux.acc_seg: 99.1766 +04/18 13:41:04 - mmengine - INFO - Iter(train) [ 73650/160000] lr: 5.7828e-03 eta: 13:14:10 time: 0.5522 data_time: 0.0063 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0079 decode.acc_seg: 99.6997 aux.loss_ce: 0.0077 aux.acc_seg: 99.1902 +04/18 13:41:32 - mmengine - INFO - Iter(train) [ 73700/160000] lr: 5.7799e-03 eta: 13:13:43 time: 0.5514 data_time: 0.0063 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0089 decode.acc_seg: 99.5741 aux.loss_ce: 0.0083 aux.acc_seg: 99.1199 +04/18 13:42:00 - mmengine - INFO - Iter(train) [ 73750/160000] lr: 5.7769e-03 eta: 13:13:15 time: 0.5530 data_time: 0.0065 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0084 decode.acc_seg: 99.6207 aux.loss_ce: 0.0078 aux.acc_seg: 98.9251 +04/18 13:42:27 - mmengine - INFO - Iter(train) [ 73800/160000] lr: 5.7740e-03 eta: 13:12:48 time: 0.5527 data_time: 0.0066 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0067 decode.acc_seg: 99.7057 aux.loss_ce: 0.0073 aux.acc_seg: 99.1655 +04/18 13:42:55 - mmengine - INFO - Iter(train) [ 73850/160000] lr: 5.7710e-03 eta: 13:12:20 time: 0.5541 data_time: 0.0060 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0072 decode.acc_seg: 99.6880 aux.loss_ce: 0.0073 aux.acc_seg: 99.2225 +04/18 13:43:22 - mmengine - INFO - Iter(train) [ 73900/160000] lr: 5.7680e-03 eta: 13:11:53 time: 0.5520 data_time: 0.0063 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0091 decode.acc_seg: 99.7266 aux.loss_ce: 0.0077 aux.acc_seg: 99.3498 +04/18 13:43:50 - mmengine - INFO - Iter(train) [ 73950/160000] lr: 5.7651e-03 eta: 13:11:25 time: 0.5538 data_time: 0.0065 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0076 decode.acc_seg: 99.6967 aux.loss_ce: 0.0075 aux.acc_seg: 99.1593 +04/18 13:44:18 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 13:44:18 - mmengine - INFO - Iter(train) [ 74000/160000] lr: 5.7621e-03 eta: 13:10:58 time: 0.5507 data_time: 0.0064 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0073 decode.acc_seg: 99.7504 aux.loss_ce: 0.0075 aux.acc_seg: 99.2293 +04/18 13:44:45 - mmengine - INFO - Iter(train) [ 74050/160000] lr: 5.7591e-03 eta: 13:10:30 time: 0.5523 data_time: 0.0065 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0074 decode.acc_seg: 99.7253 aux.loss_ce: 0.0074 aux.acc_seg: 99.3150 +04/18 13:45:13 - mmengine - INFO - Iter(train) [ 74100/160000] lr: 5.7562e-03 eta: 13:10:02 time: 0.5513 data_time: 0.0064 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0072 decode.acc_seg: 99.7438 aux.loss_ce: 0.0073 aux.acc_seg: 99.2490 +04/18 13:45:41 - mmengine - INFO - Iter(train) [ 74150/160000] lr: 5.7532e-03 eta: 13:09:35 time: 0.5530 data_time: 0.0066 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0069 decode.acc_seg: 99.6561 aux.loss_ce: 0.0072 aux.acc_seg: 99.1373 +04/18 13:46:08 - mmengine - INFO - Iter(train) [ 74200/160000] lr: 5.7503e-03 eta: 13:09:07 time: 0.5527 data_time: 0.0068 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0070 decode.acc_seg: 99.6842 aux.loss_ce: 0.0073 aux.acc_seg: 99.4061 +04/18 13:46:36 - mmengine - INFO - Iter(train) [ 74250/160000] lr: 5.7473e-03 eta: 13:08:40 time: 0.5513 data_time: 0.0061 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0071 decode.acc_seg: 99.7683 aux.loss_ce: 0.0074 aux.acc_seg: 99.3338 +04/18 13:47:04 - mmengine - INFO - Iter(train) [ 74300/160000] lr: 5.7443e-03 eta: 13:08:12 time: 0.5521 data_time: 0.0067 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0073 decode.acc_seg: 99.7014 aux.loss_ce: 0.0073 aux.acc_seg: 99.2202 +04/18 13:47:31 - mmengine - INFO - Iter(train) [ 74350/160000] lr: 5.7414e-03 eta: 13:07:45 time: 0.5522 data_time: 0.0071 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0076 decode.acc_seg: 99.7216 aux.loss_ce: 0.0076 aux.acc_seg: 99.2443 +04/18 13:47:59 - mmengine - INFO - Iter(train) [ 74400/160000] lr: 5.7384e-03 eta: 13:07:17 time: 0.5533 data_time: 0.0066 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0069 decode.acc_seg: 99.7744 aux.loss_ce: 0.0072 aux.acc_seg: 99.3647 +04/18 13:48:27 - mmengine - INFO - Iter(train) [ 74450/160000] lr: 5.7354e-03 eta: 13:06:50 time: 0.5514 data_time: 0.0063 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0077 decode.acc_seg: 99.6039 aux.loss_ce: 0.0078 aux.acc_seg: 99.1303 +04/18 13:48:54 - mmengine - INFO - Iter(train) [ 74500/160000] lr: 5.7325e-03 eta: 13:06:22 time: 0.5527 data_time: 0.0060 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0075 decode.acc_seg: 99.6873 aux.loss_ce: 0.0075 aux.acc_seg: 99.2661 +04/18 13:49:22 - mmengine - INFO - Iter(train) [ 74550/160000] lr: 5.7295e-03 eta: 13:05:55 time: 0.5601 data_time: 0.0067 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0071 decode.acc_seg: 99.7502 aux.loss_ce: 0.0072 aux.acc_seg: 99.4673 +04/18 13:49:50 - mmengine - INFO - Iter(train) [ 74600/160000] lr: 5.7265e-03 eta: 13:05:27 time: 0.5527 data_time: 0.0064 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0074 decode.acc_seg: 99.6539 aux.loss_ce: 0.0074 aux.acc_seg: 99.1780 +04/18 13:50:17 - mmengine - INFO - Iter(train) [ 74650/160000] lr: 5.7236e-03 eta: 13:05:00 time: 0.5522 data_time: 0.0062 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0073 decode.acc_seg: 99.6793 aux.loss_ce: 0.0076 aux.acc_seg: 99.0724 +04/18 13:50:45 - mmengine - INFO - Iter(train) [ 74700/160000] lr: 5.7206e-03 eta: 13:04:32 time: 0.5520 data_time: 0.0061 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0073 decode.acc_seg: 99.7586 aux.loss_ce: 0.0072 aux.acc_seg: 99.3429 +04/18 13:51:13 - mmengine - INFO - Iter(train) [ 74750/160000] lr: 5.7176e-03 eta: 13:04:05 time: 0.5539 data_time: 0.0067 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0068 decode.acc_seg: 99.7895 aux.loss_ce: 0.0070 aux.acc_seg: 99.3579 +04/18 13:51:40 - mmengine - INFO - Iter(train) [ 74800/160000] lr: 5.7147e-03 eta: 13:03:37 time: 0.5555 data_time: 0.0066 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0072 decode.acc_seg: 99.6549 aux.loss_ce: 0.0069 aux.acc_seg: 99.0606 +04/18 13:52:08 - mmengine - INFO - Iter(train) [ 74850/160000] lr: 5.7117e-03 eta: 13:03:10 time: 0.5523 data_time: 0.0065 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0072 decode.acc_seg: 99.6537 aux.loss_ce: 0.0075 aux.acc_seg: 99.2857 +04/18 13:52:36 - mmengine - INFO - Iter(train) [ 74900/160000] lr: 5.7088e-03 eta: 13:02:42 time: 0.5531 data_time: 0.0065 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0077 decode.acc_seg: 99.6949 aux.loss_ce: 0.0080 aux.acc_seg: 99.3430 +04/18 13:53:03 - mmengine - INFO - Iter(train) [ 74950/160000] lr: 5.7058e-03 eta: 13:02:15 time: 0.5525 data_time: 0.0067 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0076 decode.acc_seg: 99.6485 aux.loss_ce: 0.0076 aux.acc_seg: 99.1668 +04/18 13:53:31 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 13:53:31 - mmengine - INFO - Iter(train) [ 75000/160000] lr: 5.7028e-03 eta: 13:01:47 time: 0.5534 data_time: 0.0062 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0070 decode.acc_seg: 99.7306 aux.loss_ce: 0.0069 aux.acc_seg: 99.3453 +04/18 13:53:59 - mmengine - INFO - Iter(train) [ 75050/160000] lr: 5.6999e-03 eta: 13:01:20 time: 0.5542 data_time: 0.0065 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0072 decode.acc_seg: 99.6854 aux.loss_ce: 0.0077 aux.acc_seg: 99.3029 +04/18 13:54:26 - mmengine - INFO - Iter(train) [ 75100/160000] lr: 5.6969e-03 eta: 13:00:52 time: 0.5522 data_time: 0.0065 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0080 decode.acc_seg: 99.6496 aux.loss_ce: 0.0079 aux.acc_seg: 99.2293 +04/18 13:54:54 - mmengine - INFO - Iter(train) [ 75150/160000] lr: 5.6939e-03 eta: 13:00:24 time: 0.5518 data_time: 0.0063 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0086 decode.acc_seg: 99.7335 aux.loss_ce: 0.0076 aux.acc_seg: 99.2230 +04/18 13:55:22 - mmengine - INFO - Iter(train) [ 75200/160000] lr: 5.6910e-03 eta: 12:59:57 time: 0.5517 data_time: 0.0063 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0089 decode.acc_seg: 99.6819 aux.loss_ce: 0.0083 aux.acc_seg: 99.0951 +04/18 13:55:49 - mmengine - INFO - Iter(train) [ 75250/160000] lr: 5.6880e-03 eta: 12:59:29 time: 0.5526 data_time: 0.0069 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0080 decode.acc_seg: 99.6603 aux.loss_ce: 0.0081 aux.acc_seg: 99.1907 +04/18 13:56:17 - mmengine - INFO - Iter(train) [ 75300/160000] lr: 5.6850e-03 eta: 12:59:02 time: 0.5514 data_time: 0.0062 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0082 decode.acc_seg: 99.6013 aux.loss_ce: 0.0076 aux.acc_seg: 99.2129 +04/18 13:56:44 - mmengine - INFO - Iter(train) [ 75350/160000] lr: 5.6821e-03 eta: 12:58:34 time: 0.5522 data_time: 0.0064 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0087 decode.acc_seg: 99.7513 aux.loss_ce: 0.0085 aux.acc_seg: 99.2340 +04/18 13:57:12 - mmengine - INFO - Iter(train) [ 75400/160000] lr: 5.6791e-03 eta: 12:58:07 time: 0.5529 data_time: 0.0069 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0082 decode.acc_seg: 99.7286 aux.loss_ce: 0.0081 aux.acc_seg: 99.3635 +04/18 13:57:40 - mmengine - INFO - Iter(train) [ 75450/160000] lr: 5.6761e-03 eta: 12:57:39 time: 0.5534 data_time: 0.0064 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0075 decode.acc_seg: 99.7062 aux.loss_ce: 0.0076 aux.acc_seg: 99.2773 +04/18 13:58:07 - mmengine - INFO - Iter(train) [ 75500/160000] lr: 5.6731e-03 eta: 12:57:12 time: 0.5529 data_time: 0.0065 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0085 decode.acc_seg: 99.6880 aux.loss_ce: 0.0079 aux.acc_seg: 99.1724 +04/18 13:58:35 - mmengine - INFO - Iter(train) [ 75550/160000] lr: 5.6702e-03 eta: 12:56:44 time: 0.5509 data_time: 0.0061 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0091 decode.acc_seg: 99.5934 aux.loss_ce: 0.0082 aux.acc_seg: 99.0724 +04/18 13:59:03 - mmengine - INFO - Iter(train) [ 75600/160000] lr: 5.6672e-03 eta: 12:56:17 time: 0.5612 data_time: 0.0064 memory: 7635 loss: 0.0187 decode.loss_ce: 0.0098 decode.acc_seg: 99.6265 aux.loss_ce: 0.0088 aux.acc_seg: 99.2264 +04/18 13:59:31 - mmengine - INFO - Iter(train) [ 75650/160000] lr: 5.6642e-03 eta: 12:55:49 time: 0.5617 data_time: 0.0067 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0086 decode.acc_seg: 99.6624 aux.loss_ce: 0.0076 aux.acc_seg: 99.1029 +04/18 13:59:58 - mmengine - INFO - Iter(train) [ 75700/160000] lr: 5.6613e-03 eta: 12:55:22 time: 0.5527 data_time: 0.0067 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0084 decode.acc_seg: 99.6754 aux.loss_ce: 0.0077 aux.acc_seg: 99.2651 +04/18 14:00:26 - mmengine - INFO - Iter(train) [ 75750/160000] lr: 5.6583e-03 eta: 12:54:54 time: 0.5508 data_time: 0.0063 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0076 decode.acc_seg: 99.6983 aux.loss_ce: 0.0072 aux.acc_seg: 99.2643 +04/18 14:00:54 - mmengine - INFO - Iter(train) [ 75800/160000] lr: 5.6553e-03 eta: 12:54:27 time: 0.5529 data_time: 0.0073 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0074 decode.acc_seg: 99.6488 aux.loss_ce: 0.0078 aux.acc_seg: 99.1465 +04/18 14:01:21 - mmengine - INFO - Iter(train) [ 75850/160000] lr: 5.6524e-03 eta: 12:53:59 time: 0.5518 data_time: 0.0066 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0082 decode.acc_seg: 99.5838 aux.loss_ce: 0.0080 aux.acc_seg: 99.0612 +04/18 14:01:49 - mmengine - INFO - Iter(train) [ 75900/160000] lr: 5.6494e-03 eta: 12:53:32 time: 0.5517 data_time: 0.0063 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0074 decode.acc_seg: 99.7929 aux.loss_ce: 0.0072 aux.acc_seg: 99.5135 +04/18 14:02:17 - mmengine - INFO - Iter(train) [ 75950/160000] lr: 5.6464e-03 eta: 12:53:04 time: 0.5544 data_time: 0.0070 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0083 decode.acc_seg: 99.6447 aux.loss_ce: 0.0080 aux.acc_seg: 99.0114 +04/18 14:02:44 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 14:02:44 - mmengine - INFO - Iter(train) [ 76000/160000] lr: 5.6435e-03 eta: 12:52:37 time: 0.5505 data_time: 0.0064 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0079 decode.acc_seg: 99.6880 aux.loss_ce: 0.0079 aux.acc_seg: 99.0818 +04/18 14:03:12 - mmengine - INFO - Iter(train) [ 76050/160000] lr: 5.6405e-03 eta: 12:52:09 time: 0.5538 data_time: 0.0065 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0078 decode.acc_seg: 99.7631 aux.loss_ce: 0.0082 aux.acc_seg: 99.4489 +04/18 14:03:39 - mmengine - INFO - Iter(train) [ 76100/160000] lr: 5.6375e-03 eta: 12:51:41 time: 0.5519 data_time: 0.0066 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0072 decode.acc_seg: 99.6859 aux.loss_ce: 0.0072 aux.acc_seg: 99.2662 +04/18 14:04:07 - mmengine - INFO - Iter(train) [ 76150/160000] lr: 5.6346e-03 eta: 12:51:14 time: 0.5537 data_time: 0.0062 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0074 decode.acc_seg: 99.7334 aux.loss_ce: 0.0077 aux.acc_seg: 99.4172 +04/18 14:04:35 - mmengine - INFO - Iter(train) [ 76200/160000] lr: 5.6316e-03 eta: 12:50:46 time: 0.5527 data_time: 0.0066 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0071 decode.acc_seg: 99.7077 aux.loss_ce: 0.0075 aux.acc_seg: 99.2603 +04/18 14:05:02 - mmengine - INFO - Iter(train) [ 76250/160000] lr: 5.6286e-03 eta: 12:50:19 time: 0.5537 data_time: 0.0061 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0075 decode.acc_seg: 99.6560 aux.loss_ce: 0.0073 aux.acc_seg: 99.2674 +04/18 14:05:30 - mmengine - INFO - Iter(train) [ 76300/160000] lr: 5.6256e-03 eta: 12:49:51 time: 0.5521 data_time: 0.0062 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0068 decode.acc_seg: 99.6660 aux.loss_ce: 0.0073 aux.acc_seg: 98.9780 +04/18 14:05:58 - mmengine - INFO - Iter(train) [ 76350/160000] lr: 5.6227e-03 eta: 12:49:24 time: 0.5534 data_time: 0.0067 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0103 decode.acc_seg: 99.7115 aux.loss_ce: 0.0088 aux.acc_seg: 99.1650 +04/18 14:06:25 - mmengine - INFO - Iter(train) [ 76400/160000] lr: 5.6197e-03 eta: 12:48:56 time: 0.5525 data_time: 0.0065 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0084 decode.acc_seg: 99.6535 aux.loss_ce: 0.0085 aux.acc_seg: 99.1377 +04/18 14:06:53 - mmengine - INFO - Iter(train) [ 76450/160000] lr: 5.6167e-03 eta: 12:48:29 time: 0.5524 data_time: 0.0065 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0090 decode.acc_seg: 99.6810 aux.loss_ce: 0.0085 aux.acc_seg: 99.1442 +04/18 14:07:21 - mmengine - INFO - Iter(train) [ 76500/160000] lr: 5.6138e-03 eta: 12:48:01 time: 0.5529 data_time: 0.0065 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0078 decode.acc_seg: 99.6929 aux.loss_ce: 0.0078 aux.acc_seg: 99.1059 +04/18 14:07:48 - mmengine - INFO - Iter(train) [ 76550/160000] lr: 5.6108e-03 eta: 12:47:34 time: 0.5533 data_time: 0.0074 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0079 decode.acc_seg: 99.6104 aux.loss_ce: 0.0074 aux.acc_seg: 99.3080 +04/18 14:08:16 - mmengine - INFO - Iter(train) [ 76600/160000] lr: 5.6078e-03 eta: 12:47:06 time: 0.5530 data_time: 0.0072 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0087 decode.acc_seg: 99.6520 aux.loss_ce: 0.0079 aux.acc_seg: 99.2299 +04/18 14:08:44 - mmengine - INFO - Iter(train) [ 76650/160000] lr: 5.6048e-03 eta: 12:46:39 time: 0.5530 data_time: 0.0071 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0076 decode.acc_seg: 99.7184 aux.loss_ce: 0.0078 aux.acc_seg: 99.2104 +04/18 14:09:11 - mmengine - INFO - Iter(train) [ 76700/160000] lr: 5.6019e-03 eta: 12:46:11 time: 0.5547 data_time: 0.0069 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0071 decode.acc_seg: 99.7515 aux.loss_ce: 0.0073 aux.acc_seg: 99.2829 +04/18 14:09:39 - mmengine - INFO - Iter(train) [ 76750/160000] lr: 5.5989e-03 eta: 12:45:44 time: 0.5530 data_time: 0.0067 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0080 decode.acc_seg: 99.7278 aux.loss_ce: 0.0082 aux.acc_seg: 99.1805 +04/18 14:10:07 - mmengine - INFO - Iter(train) [ 76800/160000] lr: 5.5959e-03 eta: 12:45:16 time: 0.5519 data_time: 0.0059 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0086 decode.acc_seg: 99.6566 aux.loss_ce: 0.0082 aux.acc_seg: 99.1994 +04/18 14:10:34 - mmengine - INFO - Iter(train) [ 76850/160000] lr: 5.5930e-03 eta: 12:44:49 time: 0.5539 data_time: 0.0068 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0077 decode.acc_seg: 99.7783 aux.loss_ce: 0.0076 aux.acc_seg: 99.3851 +04/18 14:11:02 - mmengine - INFO - Iter(train) [ 76900/160000] lr: 5.5900e-03 eta: 12:44:21 time: 0.5537 data_time: 0.0069 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0089 decode.acc_seg: 99.5828 aux.loss_ce: 0.0083 aux.acc_seg: 99.1943 +04/18 14:11:30 - mmengine - INFO - Iter(train) [ 76950/160000] lr: 5.5870e-03 eta: 12:43:54 time: 0.5526 data_time: 0.0068 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0084 decode.acc_seg: 99.7047 aux.loss_ce: 0.0079 aux.acc_seg: 99.2355 +04/18 14:11:58 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 14:11:58 - mmengine - INFO - Iter(train) [ 77000/160000] lr: 5.5840e-03 eta: 12:43:26 time: 0.5531 data_time: 0.0062 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0094 decode.acc_seg: 99.5891 aux.loss_ce: 0.0079 aux.acc_seg: 99.2018 +04/18 14:12:25 - mmengine - INFO - Iter(train) [ 77050/160000] lr: 5.5811e-03 eta: 12:42:59 time: 0.5541 data_time: 0.0073 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0097 decode.acc_seg: 99.5165 aux.loss_ce: 0.0083 aux.acc_seg: 98.8960 +04/18 14:12:53 - mmengine - INFO - Iter(train) [ 77100/160000] lr: 5.5781e-03 eta: 12:42:31 time: 0.5531 data_time: 0.0061 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0083 decode.acc_seg: 99.7272 aux.loss_ce: 0.0082 aux.acc_seg: 99.2239 +04/18 14:13:21 - mmengine - INFO - Iter(train) [ 77150/160000] lr: 5.5751e-03 eta: 12:42:04 time: 0.5539 data_time: 0.0064 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0076 decode.acc_seg: 99.7736 aux.loss_ce: 0.0074 aux.acc_seg: 99.4412 +04/18 14:13:48 - mmengine - INFO - Iter(train) [ 77200/160000] lr: 5.5721e-03 eta: 12:41:36 time: 0.5531 data_time: 0.0071 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0096 decode.acc_seg: 99.7185 aux.loss_ce: 0.0081 aux.acc_seg: 99.3146 +04/18 14:14:16 - mmengine - INFO - Iter(train) [ 77250/160000] lr: 5.5692e-03 eta: 12:41:09 time: 0.5540 data_time: 0.0064 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0075 decode.acc_seg: 99.6916 aux.loss_ce: 0.0074 aux.acc_seg: 98.9935 +04/18 14:14:44 - mmengine - INFO - Iter(train) [ 77300/160000] lr: 5.5662e-03 eta: 12:40:41 time: 0.5532 data_time: 0.0073 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0077 decode.acc_seg: 99.6530 aux.loss_ce: 0.0072 aux.acc_seg: 99.3362 +04/18 14:15:11 - mmengine - INFO - Iter(train) [ 77350/160000] lr: 5.5632e-03 eta: 12:40:14 time: 0.5533 data_time: 0.0073 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0074 decode.acc_seg: 99.7839 aux.loss_ce: 0.0075 aux.acc_seg: 99.4560 +04/18 14:15:39 - mmengine - INFO - Iter(train) [ 77400/160000] lr: 5.5602e-03 eta: 12:39:46 time: 0.5534 data_time: 0.0065 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0078 decode.acc_seg: 99.5231 aux.loss_ce: 0.0074 aux.acc_seg: 99.0528 +04/18 14:16:07 - mmengine - INFO - Iter(train) [ 77450/160000] lr: 5.5573e-03 eta: 12:39:18 time: 0.5532 data_time: 0.0062 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0075 decode.acc_seg: 99.6844 aux.loss_ce: 0.0076 aux.acc_seg: 99.3047 +04/18 14:16:34 - mmengine - INFO - Iter(train) [ 77500/160000] lr: 5.5543e-03 eta: 12:38:51 time: 0.5543 data_time: 0.0063 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0073 decode.acc_seg: 99.8195 aux.loss_ce: 0.0074 aux.acc_seg: 99.4947 +04/18 14:17:02 - mmengine - INFO - Iter(train) [ 77550/160000] lr: 5.5513e-03 eta: 12:38:23 time: 0.5514 data_time: 0.0060 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0077 decode.acc_seg: 99.6561 aux.loss_ce: 0.0077 aux.acc_seg: 99.2005 +04/18 14:17:30 - mmengine - INFO - Iter(train) [ 77600/160000] lr: 5.5483e-03 eta: 12:37:56 time: 0.5542 data_time: 0.0064 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0077 decode.acc_seg: 99.6532 aux.loss_ce: 0.0081 aux.acc_seg: 99.0906 +04/18 14:17:57 - mmengine - INFO - Iter(train) [ 77650/160000] lr: 5.5454e-03 eta: 12:37:28 time: 0.5527 data_time: 0.0065 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0082 decode.acc_seg: 99.6670 aux.loss_ce: 0.0079 aux.acc_seg: 99.1526 +04/18 14:18:25 - mmengine - INFO - Iter(train) [ 77700/160000] lr: 5.5424e-03 eta: 12:37:01 time: 0.5537 data_time: 0.0063 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0086 decode.acc_seg: 99.6713 aux.loss_ce: 0.0078 aux.acc_seg: 99.3465 +04/18 14:18:53 - mmengine - INFO - Iter(train) [ 77750/160000] lr: 5.5394e-03 eta: 12:36:34 time: 0.5541 data_time: 0.0067 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0079 decode.acc_seg: 99.6013 aux.loss_ce: 0.0079 aux.acc_seg: 99.0311 +04/18 14:19:21 - mmengine - INFO - Iter(train) [ 77800/160000] lr: 5.5364e-03 eta: 12:36:06 time: 0.5526 data_time: 0.0057 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0074 decode.acc_seg: 99.7191 aux.loss_ce: 0.0076 aux.acc_seg: 99.1973 +04/18 14:19:48 - mmengine - INFO - Iter(train) [ 77850/160000] lr: 5.5335e-03 eta: 12:35:39 time: 0.5515 data_time: 0.0063 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0080 decode.acc_seg: 99.6448 aux.loss_ce: 0.0081 aux.acc_seg: 99.0735 +04/18 14:20:16 - mmengine - INFO - Iter(train) [ 77900/160000] lr: 5.5305e-03 eta: 12:35:11 time: 0.5545 data_time: 0.0061 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0074 decode.acc_seg: 99.7859 aux.loss_ce: 0.0075 aux.acc_seg: 99.3713 +04/18 14:20:44 - mmengine - INFO - Iter(train) [ 77950/160000] lr: 5.5275e-03 eta: 12:34:44 time: 0.5525 data_time: 0.0065 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0077 decode.acc_seg: 99.7121 aux.loss_ce: 0.0080 aux.acc_seg: 99.0719 +04/18 14:21:11 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 14:21:11 - mmengine - INFO - Iter(train) [ 78000/160000] lr: 5.5245e-03 eta: 12:34:16 time: 0.5513 data_time: 0.0062 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0082 decode.acc_seg: 99.6497 aux.loss_ce: 0.0077 aux.acc_seg: 99.0651 +04/18 14:21:39 - mmengine - INFO - Iter(train) [ 78050/160000] lr: 5.5216e-03 eta: 12:33:49 time: 0.5553 data_time: 0.0064 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0075 decode.acc_seg: 99.6351 aux.loss_ce: 0.0076 aux.acc_seg: 99.0310 +04/18 14:22:07 - mmengine - INFO - Iter(train) [ 78100/160000] lr: 5.5186e-03 eta: 12:33:21 time: 0.5536 data_time: 0.0069 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0088 decode.acc_seg: 99.7007 aux.loss_ce: 0.0081 aux.acc_seg: 99.2872 +04/18 14:22:34 - mmengine - INFO - Iter(train) [ 78150/160000] lr: 5.5156e-03 eta: 12:32:53 time: 0.5527 data_time: 0.0067 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0076 decode.acc_seg: 99.6365 aux.loss_ce: 0.0076 aux.acc_seg: 99.1067 +04/18 14:23:02 - mmengine - INFO - Iter(train) [ 78200/160000] lr: 5.5126e-03 eta: 12:32:26 time: 0.5523 data_time: 0.0067 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0080 decode.acc_seg: 99.7033 aux.loss_ce: 0.0085 aux.acc_seg: 99.1899 +04/18 14:23:30 - mmengine - INFO - Iter(train) [ 78250/160000] lr: 5.5096e-03 eta: 12:31:58 time: 0.5516 data_time: 0.0070 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0071 decode.acc_seg: 99.7140 aux.loss_ce: 0.0072 aux.acc_seg: 99.2377 +04/18 14:23:57 - mmengine - INFO - Iter(train) [ 78300/160000] lr: 5.5067e-03 eta: 12:31:31 time: 0.5529 data_time: 0.0060 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0077 decode.acc_seg: 99.6404 aux.loss_ce: 0.0074 aux.acc_seg: 99.1546 +04/18 14:24:25 - mmengine - INFO - Iter(train) [ 78350/160000] lr: 5.5037e-03 eta: 12:31:03 time: 0.5542 data_time: 0.0066 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0069 decode.acc_seg: 99.7976 aux.loss_ce: 0.0071 aux.acc_seg: 99.3376 +04/18 14:24:53 - mmengine - INFO - Iter(train) [ 78400/160000] lr: 5.5007e-03 eta: 12:30:36 time: 0.5523 data_time: 0.0074 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0080 decode.acc_seg: 99.7155 aux.loss_ce: 0.0077 aux.acc_seg: 99.3938 +04/18 14:25:20 - mmengine - INFO - Iter(train) [ 78450/160000] lr: 5.4977e-03 eta: 12:30:08 time: 0.5541 data_time: 0.0061 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0088 decode.acc_seg: 99.6658 aux.loss_ce: 0.0080 aux.acc_seg: 99.1904 +04/18 14:25:48 - mmengine - INFO - Iter(train) [ 78500/160000] lr: 5.4948e-03 eta: 12:29:41 time: 0.5531 data_time: 0.0067 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0082 decode.acc_seg: 99.7759 aux.loss_ce: 0.0082 aux.acc_seg: 99.3614 +04/18 14:26:16 - mmengine - INFO - Iter(train) [ 78550/160000] lr: 5.4918e-03 eta: 12:29:13 time: 0.5521 data_time: 0.0065 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0073 decode.acc_seg: 99.6882 aux.loss_ce: 0.0074 aux.acc_seg: 99.4184 +04/18 14:26:43 - mmengine - INFO - Iter(train) [ 78600/160000] lr: 5.4888e-03 eta: 12:28:46 time: 0.5546 data_time: 0.0070 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0076 decode.acc_seg: 99.7941 aux.loss_ce: 0.0077 aux.acc_seg: 99.3733 +04/18 14:27:11 - mmengine - INFO - Iter(train) [ 78650/160000] lr: 5.4858e-03 eta: 12:28:18 time: 0.5536 data_time: 0.0068 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0074 decode.acc_seg: 99.6781 aux.loss_ce: 0.0075 aux.acc_seg: 99.1870 +04/18 14:27:39 - mmengine - INFO - Iter(train) [ 78700/160000] lr: 5.4828e-03 eta: 12:27:51 time: 0.5530 data_time: 0.0066 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0073 decode.acc_seg: 99.7382 aux.loss_ce: 0.0078 aux.acc_seg: 99.1807 +04/18 14:28:06 - mmengine - INFO - Iter(train) [ 78750/160000] lr: 5.4799e-03 eta: 12:27:23 time: 0.5541 data_time: 0.0066 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0079 decode.acc_seg: 99.7094 aux.loss_ce: 0.0078 aux.acc_seg: 99.2322 +04/18 14:28:34 - mmengine - INFO - Iter(train) [ 78800/160000] lr: 5.4769e-03 eta: 12:26:56 time: 0.5540 data_time: 0.0066 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0075 decode.acc_seg: 99.6613 aux.loss_ce: 0.0074 aux.acc_seg: 99.0807 +04/18 14:29:02 - mmengine - INFO - Iter(train) [ 78850/160000] lr: 5.4739e-03 eta: 12:26:28 time: 0.5516 data_time: 0.0064 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0080 decode.acc_seg: 99.7021 aux.loss_ce: 0.0078 aux.acc_seg: 99.1442 +04/18 14:29:30 - mmengine - INFO - Iter(train) [ 78900/160000] lr: 5.4709e-03 eta: 12:26:01 time: 0.5537 data_time: 0.0065 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0077 decode.acc_seg: 99.6992 aux.loss_ce: 0.0075 aux.acc_seg: 99.3137 +04/18 14:29:57 - mmengine - INFO - Iter(train) [ 78950/160000] lr: 5.4679e-03 eta: 12:25:33 time: 0.5531 data_time: 0.0074 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0077 decode.acc_seg: 99.6041 aux.loss_ce: 0.0077 aux.acc_seg: 99.0901 +04/18 14:30:25 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 14:30:25 - mmengine - INFO - Iter(train) [ 79000/160000] lr: 5.4650e-03 eta: 12:25:06 time: 0.5531 data_time: 0.0072 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0074 decode.acc_seg: 99.7106 aux.loss_ce: 0.0071 aux.acc_seg: 99.2305 +04/18 14:30:53 - mmengine - INFO - Iter(train) [ 79050/160000] lr: 5.4620e-03 eta: 12:24:38 time: 0.5528 data_time: 0.0067 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0074 decode.acc_seg: 99.7114 aux.loss_ce: 0.0074 aux.acc_seg: 99.4676 +04/18 14:31:21 - mmengine - INFO - Iter(train) [ 79100/160000] lr: 5.4590e-03 eta: 12:24:11 time: 0.5531 data_time: 0.0065 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0074 decode.acc_seg: 99.6846 aux.loss_ce: 0.0078 aux.acc_seg: 99.1245 +04/18 14:31:48 - mmengine - INFO - Iter(train) [ 79150/160000] lr: 5.4560e-03 eta: 12:23:43 time: 0.5542 data_time: 0.0063 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0074 decode.acc_seg: 99.6811 aux.loss_ce: 0.0079 aux.acc_seg: 99.1734 +04/18 14:32:16 - mmengine - INFO - Iter(train) [ 79200/160000] lr: 5.4530e-03 eta: 12:23:16 time: 0.5547 data_time: 0.0075 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0073 decode.acc_seg: 99.6526 aux.loss_ce: 0.0074 aux.acc_seg: 99.3362 +04/18 14:32:44 - mmengine - INFO - Iter(train) [ 79250/160000] lr: 5.4501e-03 eta: 12:22:48 time: 0.5550 data_time: 0.0072 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0068 decode.acc_seg: 99.7242 aux.loss_ce: 0.0074 aux.acc_seg: 99.3617 +04/18 14:33:11 - mmengine - INFO - Iter(train) [ 79300/160000] lr: 5.4471e-03 eta: 12:22:21 time: 0.5529 data_time: 0.0063 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0068 decode.acc_seg: 99.7284 aux.loss_ce: 0.0073 aux.acc_seg: 99.1774 +04/18 14:33:39 - mmengine - INFO - Iter(train) [ 79350/160000] lr: 5.4441e-03 eta: 12:21:53 time: 0.5535 data_time: 0.0065 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0072 decode.acc_seg: 99.7021 aux.loss_ce: 0.0072 aux.acc_seg: 99.0726 +04/18 14:34:07 - mmengine - INFO - Iter(train) [ 79400/160000] lr: 5.4411e-03 eta: 12:21:26 time: 0.5528 data_time: 0.0065 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0073 decode.acc_seg: 99.7086 aux.loss_ce: 0.0081 aux.acc_seg: 99.2033 +04/18 14:34:34 - mmengine - INFO - Iter(train) [ 79450/160000] lr: 5.4381e-03 eta: 12:20:58 time: 0.5532 data_time: 0.0063 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0068 decode.acc_seg: 99.6589 aux.loss_ce: 0.0071 aux.acc_seg: 99.1291 +04/18 14:35:02 - mmengine - INFO - Iter(train) [ 79500/160000] lr: 5.4351e-03 eta: 12:20:31 time: 0.5532 data_time: 0.0066 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0079 decode.acc_seg: 99.6635 aux.loss_ce: 0.0079 aux.acc_seg: 99.1943 +04/18 14:35:30 - mmengine - INFO - Iter(train) [ 79550/160000] lr: 5.4322e-03 eta: 12:20:03 time: 0.5551 data_time: 0.0075 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0078 decode.acc_seg: 99.7507 aux.loss_ce: 0.0078 aux.acc_seg: 99.2860 +04/18 14:35:57 - mmengine - INFO - Iter(train) [ 79600/160000] lr: 5.4292e-03 eta: 12:19:36 time: 0.5517 data_time: 0.0069 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0089 decode.acc_seg: 99.7174 aux.loss_ce: 0.0080 aux.acc_seg: 99.2744 +04/18 14:36:25 - mmengine - INFO - Iter(train) [ 79650/160000] lr: 5.4262e-03 eta: 12:19:08 time: 0.5535 data_time: 0.0064 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0089 decode.acc_seg: 99.6916 aux.loss_ce: 0.0082 aux.acc_seg: 99.2051 +04/18 14:36:53 - mmengine - INFO - Iter(train) [ 79700/160000] lr: 5.4232e-03 eta: 12:18:41 time: 0.5540 data_time: 0.0067 memory: 7635 loss: 0.0208 decode.loss_ce: 0.0115 decode.acc_seg: 99.3102 aux.loss_ce: 0.0094 aux.acc_seg: 98.8945 +04/18 14:37:20 - mmengine - INFO - Iter(train) [ 79750/160000] lr: 5.4202e-03 eta: 12:18:13 time: 0.5535 data_time: 0.0061 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0089 decode.acc_seg: 99.5147 aux.loss_ce: 0.0083 aux.acc_seg: 99.0007 +04/18 14:37:48 - mmengine - INFO - Iter(train) [ 79800/160000] lr: 5.4172e-03 eta: 12:17:46 time: 0.5529 data_time: 0.0071 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0106 decode.acc_seg: 99.5624 aux.loss_ce: 0.0085 aux.acc_seg: 99.2125 +04/18 14:38:16 - mmengine - INFO - Iter(train) [ 79850/160000] lr: 5.4143e-03 eta: 12:17:18 time: 0.5536 data_time: 0.0066 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0078 decode.acc_seg: 99.6892 aux.loss_ce: 0.0078 aux.acc_seg: 99.3053 +04/18 14:38:44 - mmengine - INFO - Iter(train) [ 79900/160000] lr: 5.4113e-03 eta: 12:16:51 time: 0.5626 data_time: 0.0064 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0072 decode.acc_seg: 99.7086 aux.loss_ce: 0.0075 aux.acc_seg: 99.1970 +04/18 14:39:11 - mmengine - INFO - Iter(train) [ 79950/160000] lr: 5.4083e-03 eta: 12:16:23 time: 0.5522 data_time: 0.0060 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0089 decode.acc_seg: 99.5859 aux.loss_ce: 0.0082 aux.acc_seg: 99.0210 +04/18 14:39:39 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 14:39:39 - mmengine - INFO - Iter(train) [ 80000/160000] lr: 5.4053e-03 eta: 12:15:56 time: 0.5541 data_time: 0.0065 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0079 decode.acc_seg: 99.7456 aux.loss_ce: 0.0077 aux.acc_seg: 99.3921 +04/18 14:39:39 - mmengine - INFO - Saving checkpoint at 80000 iterations +04/18 14:39:43 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:07 time: 0.0461 data_time: 0.0014 memory: 1657 +04/18 14:39:45 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:04 time: 0.0463 data_time: 0.0015 memory: 1657 +04/18 14:39:48 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:02 time: 0.0466 data_time: 0.0013 memory: 1657 +04/18 14:39:50 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.0457 data_time: 0.0015 memory: 1657 +04/18 14:39:50 - mmengine - INFO - per class results: +04/18 14:39:50 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 99.09 | 99.53 | 99.54 | 99.56 | 99.53 | +| contrast | 80.33 | 89.42 | 89.09 | 88.77 | 89.42 | ++------------+-------+-------+--------+-----------+--------+ +04/18 14:39:50 - mmengine - INFO - Iter(val) [200/200] aAcc: 99.1300 mIoU: 89.7100 mAcc: 94.4700 mFscore: 94.3200 mPrecision: 94.1700 mRecall: 94.4700 data_time: 0.0016 time: 0.0467 +04/18 14:40:18 - mmengine - INFO - Iter(train) [ 80050/160000] lr: 5.4023e-03 eta: 12:15:28 time: 0.5529 data_time: 0.0067 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0075 decode.acc_seg: 99.7221 aux.loss_ce: 0.0074 aux.acc_seg: 99.2517 +04/18 14:40:46 - mmengine - INFO - Iter(train) [ 80100/160000] lr: 5.3993e-03 eta: 12:15:01 time: 0.5548 data_time: 0.0065 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0093 decode.acc_seg: 99.7221 aux.loss_ce: 0.0081 aux.acc_seg: 99.3535 +04/18 14:41:13 - mmengine - INFO - Iter(train) [ 80150/160000] lr: 5.3964e-03 eta: 12:14:33 time: 0.5540 data_time: 0.0070 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0077 decode.acc_seg: 99.6485 aux.loss_ce: 0.0084 aux.acc_seg: 99.0163 +04/18 14:41:41 - mmengine - INFO - Iter(train) [ 80200/160000] lr: 5.3934e-03 eta: 12:14:06 time: 0.5531 data_time: 0.0058 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0082 decode.acc_seg: 99.6730 aux.loss_ce: 0.0077 aux.acc_seg: 99.1230 +04/18 14:42:09 - mmengine - INFO - Iter(train) [ 80250/160000] lr: 5.3904e-03 eta: 12:13:38 time: 0.5530 data_time: 0.0062 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0078 decode.acc_seg: 99.7345 aux.loss_ce: 0.0071 aux.acc_seg: 99.4181 +04/18 14:42:36 - mmengine - INFO - Iter(train) [ 80300/160000] lr: 5.3874e-03 eta: 12:13:11 time: 0.5534 data_time: 0.0064 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0078 decode.acc_seg: 99.6847 aux.loss_ce: 0.0075 aux.acc_seg: 99.0840 +04/18 14:43:04 - mmengine - INFO - Iter(train) [ 80350/160000] lr: 5.3844e-03 eta: 12:12:43 time: 0.5529 data_time: 0.0065 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0083 decode.acc_seg: 99.7005 aux.loss_ce: 0.0080 aux.acc_seg: 99.3281 +04/18 14:43:32 - mmengine - INFO - Iter(train) [ 80400/160000] lr: 5.3814e-03 eta: 12:12:16 time: 0.5527 data_time: 0.0062 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0084 decode.acc_seg: 99.7069 aux.loss_ce: 0.0082 aux.acc_seg: 99.2474 +04/18 14:43:59 - mmengine - INFO - Iter(train) [ 80450/160000] lr: 5.3784e-03 eta: 12:11:48 time: 0.5531 data_time: 0.0070 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0072 decode.acc_seg: 99.6648 aux.loss_ce: 0.0075 aux.acc_seg: 99.0881 +04/18 14:44:27 - mmengine - INFO - Iter(train) [ 80500/160000] lr: 5.3755e-03 eta: 12:11:21 time: 0.5526 data_time: 0.0060 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0078 decode.acc_seg: 99.6853 aux.loss_ce: 0.0077 aux.acc_seg: 99.1436 +04/18 14:44:55 - mmengine - INFO - Iter(train) [ 80550/160000] lr: 5.3725e-03 eta: 12:10:53 time: 0.5547 data_time: 0.0072 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0070 decode.acc_seg: 99.7228 aux.loss_ce: 0.0068 aux.acc_seg: 99.3650 +04/18 14:45:23 - mmengine - INFO - Iter(train) [ 80600/160000] lr: 5.3695e-03 eta: 12:10:26 time: 0.5531 data_time: 0.0064 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0070 decode.acc_seg: 99.7419 aux.loss_ce: 0.0070 aux.acc_seg: 99.3039 +04/18 14:45:50 - mmengine - INFO - Iter(train) [ 80650/160000] lr: 5.3665e-03 eta: 12:09:58 time: 0.5534 data_time: 0.0066 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0077 decode.acc_seg: 99.6632 aux.loss_ce: 0.0075 aux.acc_seg: 99.2009 +04/18 14:46:18 - mmengine - INFO - Iter(train) [ 80700/160000] lr: 5.3635e-03 eta: 12:09:31 time: 0.5541 data_time: 0.0078 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0074 decode.acc_seg: 99.6882 aux.loss_ce: 0.0075 aux.acc_seg: 99.1914 +04/18 14:46:46 - mmengine - INFO - Iter(train) [ 80750/160000] lr: 5.3605e-03 eta: 12:09:03 time: 0.5539 data_time: 0.0061 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0071 decode.acc_seg: 99.6524 aux.loss_ce: 0.0075 aux.acc_seg: 98.9561 +04/18 14:47:13 - mmengine - INFO - Iter(train) [ 80800/160000] lr: 5.3575e-03 eta: 12:08:36 time: 0.5532 data_time: 0.0068 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0072 decode.acc_seg: 99.7296 aux.loss_ce: 0.0071 aux.acc_seg: 99.1956 +04/18 14:47:41 - mmengine - INFO - Iter(train) [ 80850/160000] lr: 5.3545e-03 eta: 12:08:08 time: 0.5528 data_time: 0.0067 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0070 decode.acc_seg: 99.6763 aux.loss_ce: 0.0070 aux.acc_seg: 99.1087 +04/18 14:48:09 - mmengine - INFO - Iter(train) [ 80900/160000] lr: 5.3516e-03 eta: 12:07:41 time: 0.5536 data_time: 0.0073 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0080 decode.acc_seg: 99.7311 aux.loss_ce: 0.0085 aux.acc_seg: 99.1544 +04/18 14:48:36 - mmengine - INFO - Iter(train) [ 80950/160000] lr: 5.3486e-03 eta: 12:07:13 time: 0.5522 data_time: 0.0064 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0074 decode.acc_seg: 99.7464 aux.loss_ce: 0.0071 aux.acc_seg: 99.3919 +04/18 14:49:04 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 14:49:04 - mmengine - INFO - Iter(train) [ 81000/160000] lr: 5.3456e-03 eta: 12:06:46 time: 0.5528 data_time: 0.0068 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0076 decode.acc_seg: 99.5862 aux.loss_ce: 0.0077 aux.acc_seg: 99.0725 +04/18 14:49:32 - mmengine - INFO - Iter(train) [ 81050/160000] lr: 5.3426e-03 eta: 12:06:18 time: 0.5526 data_time: 0.0072 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0067 decode.acc_seg: 99.7510 aux.loss_ce: 0.0070 aux.acc_seg: 99.4220 +04/18 14:50:00 - mmengine - INFO - Iter(train) [ 81100/160000] lr: 5.3396e-03 eta: 12:05:51 time: 0.5517 data_time: 0.0062 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0076 decode.acc_seg: 99.6530 aux.loss_ce: 0.0078 aux.acc_seg: 99.1486 +04/18 14:50:27 - mmengine - INFO - Iter(train) [ 81150/160000] lr: 5.3366e-03 eta: 12:05:23 time: 0.5533 data_time: 0.0061 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0073 decode.acc_seg: 99.7410 aux.loss_ce: 0.0071 aux.acc_seg: 99.3360 +04/18 14:50:55 - mmengine - INFO - Iter(train) [ 81200/160000] lr: 5.3336e-03 eta: 12:04:56 time: 0.5545 data_time: 0.0079 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0073 decode.acc_seg: 99.6817 aux.loss_ce: 0.0073 aux.acc_seg: 99.2269 +04/18 14:51:23 - mmengine - INFO - Iter(train) [ 81250/160000] lr: 5.3306e-03 eta: 12:04:28 time: 0.5537 data_time: 0.0062 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0074 decode.acc_seg: 99.7697 aux.loss_ce: 0.0077 aux.acc_seg: 99.3053 +04/18 14:51:50 - mmengine - INFO - Iter(train) [ 81300/160000] lr: 5.3277e-03 eta: 12:04:01 time: 0.5535 data_time: 0.0071 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0073 decode.acc_seg: 99.6440 aux.loss_ce: 0.0073 aux.acc_seg: 99.1206 +04/18 14:52:18 - mmengine - INFO - Iter(train) [ 81350/160000] lr: 5.3247e-03 eta: 12:03:33 time: 0.5528 data_time: 0.0065 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0073 decode.acc_seg: 99.7443 aux.loss_ce: 0.0074 aux.acc_seg: 99.0421 +04/18 14:52:46 - mmengine - INFO - Iter(train) [ 81400/160000] lr: 5.3217e-03 eta: 12:03:06 time: 0.5533 data_time: 0.0067 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0075 decode.acc_seg: 99.7526 aux.loss_ce: 0.0074 aux.acc_seg: 99.3560 +04/18 14:53:13 - mmengine - INFO - Iter(train) [ 81450/160000] lr: 5.3187e-03 eta: 12:02:38 time: 0.5535 data_time: 0.0064 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0079 decode.acc_seg: 99.6834 aux.loss_ce: 0.0076 aux.acc_seg: 99.1532 +04/18 14:53:41 - mmengine - INFO - Iter(train) [ 81500/160000] lr: 5.3157e-03 eta: 12:02:10 time: 0.5549 data_time: 0.0073 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0073 decode.acc_seg: 99.7202 aux.loss_ce: 0.0072 aux.acc_seg: 99.2405 +04/18 14:54:09 - mmengine - INFO - Iter(train) [ 81550/160000] lr: 5.3127e-03 eta: 12:01:43 time: 0.5525 data_time: 0.0063 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0085 decode.acc_seg: 99.7066 aux.loss_ce: 0.0078 aux.acc_seg: 99.3187 +04/18 14:54:36 - mmengine - INFO - Iter(train) [ 81600/160000] lr: 5.3097e-03 eta: 12:01:15 time: 0.5529 data_time: 0.0069 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0082 decode.acc_seg: 99.6075 aux.loss_ce: 0.0081 aux.acc_seg: 99.0890 +04/18 14:55:04 - mmengine - INFO - Iter(train) [ 81650/160000] lr: 5.3067e-03 eta: 12:00:48 time: 0.5524 data_time: 0.0063 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0076 decode.acc_seg: 99.7453 aux.loss_ce: 0.0079 aux.acc_seg: 99.2027 +04/18 14:55:32 - mmengine - INFO - Iter(train) [ 81700/160000] lr: 5.3037e-03 eta: 12:00:20 time: 0.5525 data_time: 0.0061 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0074 decode.acc_seg: 99.7040 aux.loss_ce: 0.0075 aux.acc_seg: 99.0768 +04/18 14:55:59 - mmengine - INFO - Iter(train) [ 81750/160000] lr: 5.3007e-03 eta: 11:59:53 time: 0.5535 data_time: 0.0063 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0069 decode.acc_seg: 99.6970 aux.loss_ce: 0.0073 aux.acc_seg: 99.2434 +04/18 14:56:27 - mmengine - INFO - Iter(train) [ 81800/160000] lr: 5.2978e-03 eta: 11:59:25 time: 0.5534 data_time: 0.0068 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0078 decode.acc_seg: 99.6740 aux.loss_ce: 0.0078 aux.acc_seg: 99.1517 +04/18 14:56:55 - mmengine - INFO - Iter(train) [ 81850/160000] lr: 5.2948e-03 eta: 11:58:58 time: 0.5533 data_time: 0.0060 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0069 decode.acc_seg: 99.7755 aux.loss_ce: 0.0067 aux.acc_seg: 99.4009 +04/18 14:57:23 - mmengine - INFO - Iter(train) [ 81900/160000] lr: 5.2918e-03 eta: 11:58:30 time: 0.5551 data_time: 0.0066 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0075 decode.acc_seg: 99.6841 aux.loss_ce: 0.0072 aux.acc_seg: 99.2474 +04/18 14:57:50 - mmengine - INFO - Iter(train) [ 81950/160000] lr: 5.2888e-03 eta: 11:58:03 time: 0.5528 data_time: 0.0061 memory: 7635 loss: 0.0131 decode.loss_ce: 0.0065 decode.acc_seg: 99.8104 aux.loss_ce: 0.0065 aux.acc_seg: 99.4801 +04/18 14:58:18 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 14:58:18 - mmengine - INFO - Iter(train) [ 82000/160000] lr: 5.2858e-03 eta: 11:57:35 time: 0.5540 data_time: 0.0064 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0071 decode.acc_seg: 99.7541 aux.loss_ce: 0.0071 aux.acc_seg: 99.4308 +04/18 14:58:46 - mmengine - INFO - Iter(train) [ 82050/160000] lr: 5.2828e-03 eta: 11:57:08 time: 0.5534 data_time: 0.0067 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0078 decode.acc_seg: 99.6637 aux.loss_ce: 0.0082 aux.acc_seg: 99.0720 +04/18 14:59:14 - mmengine - INFO - Iter(train) [ 82100/160000] lr: 5.2798e-03 eta: 11:56:41 time: 0.5539 data_time: 0.0064 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0078 decode.acc_seg: 99.6006 aux.loss_ce: 0.0078 aux.acc_seg: 99.0711 +04/18 14:59:41 - mmengine - INFO - Iter(train) [ 82150/160000] lr: 5.2768e-03 eta: 11:56:13 time: 0.5531 data_time: 0.0065 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0072 decode.acc_seg: 99.7355 aux.loss_ce: 0.0077 aux.acc_seg: 99.1730 +04/18 15:00:09 - mmengine - INFO - Iter(train) [ 82200/160000] lr: 5.2738e-03 eta: 11:55:45 time: 0.5544 data_time: 0.0062 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0080 decode.acc_seg: 99.6276 aux.loss_ce: 0.0077 aux.acc_seg: 99.0049 +04/18 15:00:37 - mmengine - INFO - Iter(train) [ 82250/160000] lr: 5.2708e-03 eta: 11:55:18 time: 0.5538 data_time: 0.0067 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0086 decode.acc_seg: 99.6283 aux.loss_ce: 0.0075 aux.acc_seg: 99.3985 +04/18 15:01:04 - mmengine - INFO - Iter(train) [ 82300/160000] lr: 5.2678e-03 eta: 11:54:50 time: 0.5550 data_time: 0.0066 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0075 decode.acc_seg: 99.6887 aux.loss_ce: 0.0075 aux.acc_seg: 99.0437 +04/18 15:01:32 - mmengine - INFO - Iter(train) [ 82350/160000] lr: 5.2648e-03 eta: 11:54:23 time: 0.5541 data_time: 0.0070 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0071 decode.acc_seg: 99.6777 aux.loss_ce: 0.0074 aux.acc_seg: 99.2339 +04/18 15:02:00 - mmengine - INFO - Iter(train) [ 82400/160000] lr: 5.2618e-03 eta: 11:53:55 time: 0.5538 data_time: 0.0065 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0073 decode.acc_seg: 99.6851 aux.loss_ce: 0.0076 aux.acc_seg: 99.2001 +04/18 15:02:27 - mmengine - INFO - Iter(train) [ 82450/160000] lr: 5.2589e-03 eta: 11:53:28 time: 0.5537 data_time: 0.0069 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0089 decode.acc_seg: 99.7082 aux.loss_ce: 0.0083 aux.acc_seg: 99.3728 +04/18 15:02:55 - mmengine - INFO - Iter(train) [ 82500/160000] lr: 5.2559e-03 eta: 11:53:00 time: 0.5536 data_time: 0.0067 memory: 7635 loss: 0.0246 decode.loss_ce: 0.0147 decode.acc_seg: 99.1263 aux.loss_ce: 0.0099 aux.acc_seg: 98.6214 +04/18 15:03:23 - mmengine - INFO - Iter(train) [ 82550/160000] lr: 5.2529e-03 eta: 11:52:33 time: 0.5535 data_time: 0.0069 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0099 decode.acc_seg: 99.6870 aux.loss_ce: 0.0089 aux.acc_seg: 99.1781 +04/18 15:03:50 - mmengine - INFO - Iter(train) [ 82600/160000] lr: 5.2499e-03 eta: 11:52:05 time: 0.5532 data_time: 0.0060 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0088 decode.acc_seg: 99.6472 aux.loss_ce: 0.0081 aux.acc_seg: 99.2063 +04/18 15:04:18 - mmengine - INFO - Iter(train) [ 82650/160000] lr: 5.2469e-03 eta: 11:51:38 time: 0.5618 data_time: 0.0068 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0091 decode.acc_seg: 99.6465 aux.loss_ce: 0.0079 aux.acc_seg: 99.1975 +04/18 15:04:46 - mmengine - INFO - Iter(train) [ 82700/160000] lr: 5.2439e-03 eta: 11:51:10 time: 0.5523 data_time: 0.0064 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0077 decode.acc_seg: 99.7294 aux.loss_ce: 0.0074 aux.acc_seg: 99.3320 +04/18 15:05:13 - mmengine - INFO - Iter(train) [ 82750/160000] lr: 5.2409e-03 eta: 11:50:43 time: 0.5535 data_time: 0.0073 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0087 decode.acc_seg: 99.6490 aux.loss_ce: 0.0080 aux.acc_seg: 99.2598 +04/18 15:05:41 - mmengine - INFO - Iter(train) [ 82800/160000] lr: 5.2379e-03 eta: 11:50:15 time: 0.5528 data_time: 0.0067 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0094 decode.acc_seg: 99.6695 aux.loss_ce: 0.0082 aux.acc_seg: 99.2054 +04/18 15:06:09 - mmengine - INFO - Iter(train) [ 82850/160000] lr: 5.2349e-03 eta: 11:49:47 time: 0.5548 data_time: 0.0071 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0081 decode.acc_seg: 99.6936 aux.loss_ce: 0.0077 aux.acc_seg: 99.1436 +04/18 15:06:36 - mmengine - INFO - Iter(train) [ 82900/160000] lr: 5.2319e-03 eta: 11:49:20 time: 0.5537 data_time: 0.0063 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0079 decode.acc_seg: 99.6571 aux.loss_ce: 0.0077 aux.acc_seg: 99.1663 +04/18 15:07:04 - mmengine - INFO - Iter(train) [ 82950/160000] lr: 5.2289e-03 eta: 11:48:52 time: 0.5527 data_time: 0.0065 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0075 decode.acc_seg: 99.6861 aux.loss_ce: 0.0074 aux.acc_seg: 99.4237 +04/18 15:07:32 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 15:07:32 - mmengine - INFO - Iter(train) [ 83000/160000] lr: 5.2259e-03 eta: 11:48:25 time: 0.5541 data_time: 0.0064 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0074 decode.acc_seg: 99.6954 aux.loss_ce: 0.0074 aux.acc_seg: 99.1837 +04/18 15:07:59 - mmengine - INFO - Iter(train) [ 83050/160000] lr: 5.2229e-03 eta: 11:47:57 time: 0.5541 data_time: 0.0066 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0075 decode.acc_seg: 99.6863 aux.loss_ce: 0.0078 aux.acc_seg: 99.1245 +04/18 15:08:27 - mmengine - INFO - Iter(train) [ 83100/160000] lr: 5.2199e-03 eta: 11:47:30 time: 0.5610 data_time: 0.0064 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0066 decode.acc_seg: 99.7141 aux.loss_ce: 0.0070 aux.acc_seg: 99.1758 +04/18 15:08:55 - mmengine - INFO - Iter(train) [ 83150/160000] lr: 5.2169e-03 eta: 11:47:02 time: 0.5549 data_time: 0.0065 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0070 decode.acc_seg: 99.7163 aux.loss_ce: 0.0073 aux.acc_seg: 99.2763 +04/18 15:09:23 - mmengine - INFO - Iter(train) [ 83200/160000] lr: 5.2139e-03 eta: 11:46:35 time: 0.5541 data_time: 0.0064 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0076 decode.acc_seg: 99.6291 aux.loss_ce: 0.0076 aux.acc_seg: 98.9625 +04/18 15:09:50 - mmengine - INFO - Iter(train) [ 83250/160000] lr: 5.2109e-03 eta: 11:46:07 time: 0.5529 data_time: 0.0070 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0067 decode.acc_seg: 99.7611 aux.loss_ce: 0.0070 aux.acc_seg: 99.2933 +04/18 15:10:18 - mmengine - INFO - Iter(train) [ 83300/160000] lr: 5.2079e-03 eta: 11:45:40 time: 0.5558 data_time: 0.0061 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0073 decode.acc_seg: 99.7195 aux.loss_ce: 0.0073 aux.acc_seg: 99.1021 +04/18 15:10:46 - mmengine - INFO - Iter(train) [ 83350/160000] lr: 5.2049e-03 eta: 11:45:12 time: 0.5535 data_time: 0.0065 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0069 decode.acc_seg: 99.6616 aux.loss_ce: 0.0075 aux.acc_seg: 99.2410 +04/18 15:11:14 - mmengine - INFO - Iter(train) [ 83400/160000] lr: 5.2019e-03 eta: 11:44:45 time: 0.5545 data_time: 0.0073 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0074 decode.acc_seg: 99.7234 aux.loss_ce: 0.0079 aux.acc_seg: 99.1467 +04/18 15:11:41 - mmengine - INFO - Iter(train) [ 83450/160000] lr: 5.1989e-03 eta: 11:44:17 time: 0.5533 data_time: 0.0070 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0067 decode.acc_seg: 99.7294 aux.loss_ce: 0.0072 aux.acc_seg: 99.3606 +04/18 15:12:09 - mmengine - INFO - Iter(train) [ 83500/160000] lr: 5.1959e-03 eta: 11:43:50 time: 0.5541 data_time: 0.0062 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0075 decode.acc_seg: 99.6798 aux.loss_ce: 0.0077 aux.acc_seg: 99.1134 +04/18 15:12:37 - mmengine - INFO - Iter(train) [ 83550/160000] lr: 5.1929e-03 eta: 11:43:22 time: 0.5539 data_time: 0.0066 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0076 decode.acc_seg: 99.7350 aux.loss_ce: 0.0082 aux.acc_seg: 99.1143 +04/18 15:13:04 - mmengine - INFO - Iter(train) [ 83600/160000] lr: 5.1900e-03 eta: 11:42:55 time: 0.5539 data_time: 0.0067 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0072 decode.acc_seg: 99.6842 aux.loss_ce: 0.0076 aux.acc_seg: 99.1166 +04/18 15:13:32 - mmengine - INFO - Iter(train) [ 83650/160000] lr: 5.1870e-03 eta: 11:42:27 time: 0.5526 data_time: 0.0070 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0074 decode.acc_seg: 99.6420 aux.loss_ce: 0.0077 aux.acc_seg: 99.1487 +04/18 15:14:00 - mmengine - INFO - Iter(train) [ 83700/160000] lr: 5.1840e-03 eta: 11:42:00 time: 0.5528 data_time: 0.0066 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0072 decode.acc_seg: 99.6781 aux.loss_ce: 0.0071 aux.acc_seg: 99.2856 +04/18 15:14:27 - mmengine - INFO - Iter(train) [ 83750/160000] lr: 5.1810e-03 eta: 11:41:32 time: 0.5549 data_time: 0.0072 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0087 decode.acc_seg: 99.6272 aux.loss_ce: 0.0083 aux.acc_seg: 99.1007 +04/18 15:14:55 - mmengine - INFO - Iter(train) [ 83800/160000] lr: 5.1780e-03 eta: 11:41:05 time: 0.5538 data_time: 0.0062 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0077 decode.acc_seg: 99.6845 aux.loss_ce: 0.0074 aux.acc_seg: 99.2111 +04/18 15:15:23 - mmengine - INFO - Iter(train) [ 83850/160000] lr: 5.1750e-03 eta: 11:40:37 time: 0.5553 data_time: 0.0066 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0085 decode.acc_seg: 99.7717 aux.loss_ce: 0.0081 aux.acc_seg: 99.3909 +04/18 15:15:51 - mmengine - INFO - Iter(train) [ 83900/160000] lr: 5.1720e-03 eta: 11:40:10 time: 0.5549 data_time: 0.0060 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0081 decode.acc_seg: 99.7340 aux.loss_ce: 0.0073 aux.acc_seg: 99.3512 +04/18 15:16:18 - mmengine - INFO - Iter(train) [ 83950/160000] lr: 5.1690e-03 eta: 11:39:42 time: 0.5532 data_time: 0.0064 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0083 decode.acc_seg: 99.7579 aux.loss_ce: 0.0083 aux.acc_seg: 99.1828 +04/18 15:16:46 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 15:16:46 - mmengine - INFO - Iter(train) [ 84000/160000] lr: 5.1660e-03 eta: 11:39:15 time: 0.5530 data_time: 0.0062 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0076 decode.acc_seg: 99.7683 aux.loss_ce: 0.0075 aux.acc_seg: 99.2795 +04/18 15:17:14 - mmengine - INFO - Iter(train) [ 84050/160000] lr: 5.1630e-03 eta: 11:38:47 time: 0.5528 data_time: 0.0063 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0075 decode.acc_seg: 99.7567 aux.loss_ce: 0.0076 aux.acc_seg: 99.4097 +04/18 15:17:41 - mmengine - INFO - Iter(train) [ 84100/160000] lr: 5.1600e-03 eta: 11:38:19 time: 0.5553 data_time: 0.0066 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0076 decode.acc_seg: 99.7196 aux.loss_ce: 0.0077 aux.acc_seg: 99.2937 +04/18 15:18:09 - mmengine - INFO - Iter(train) [ 84150/160000] lr: 5.1570e-03 eta: 11:37:52 time: 0.5541 data_time: 0.0065 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0073 decode.acc_seg: 99.6591 aux.loss_ce: 0.0076 aux.acc_seg: 99.2619 +04/18 15:18:37 - mmengine - INFO - Iter(train) [ 84200/160000] lr: 5.1540e-03 eta: 11:37:25 time: 0.5541 data_time: 0.0062 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0075 decode.acc_seg: 99.7203 aux.loss_ce: 0.0069 aux.acc_seg: 99.3240 +04/18 15:19:05 - mmengine - INFO - Iter(train) [ 84250/160000] lr: 5.1510e-03 eta: 11:36:57 time: 0.5528 data_time: 0.0066 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0079 decode.acc_seg: 99.7210 aux.loss_ce: 0.0072 aux.acc_seg: 99.3785 +04/18 15:19:32 - mmengine - INFO - Iter(train) [ 84300/160000] lr: 5.1480e-03 eta: 11:36:30 time: 0.5540 data_time: 0.0066 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0097 decode.acc_seg: 99.6539 aux.loss_ce: 0.0085 aux.acc_seg: 99.0205 +04/18 15:20:00 - mmengine - INFO - Iter(train) [ 84350/160000] lr: 5.1450e-03 eta: 11:36:02 time: 0.5548 data_time: 0.0068 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0077 decode.acc_seg: 99.5545 aux.loss_ce: 0.0077 aux.acc_seg: 98.9246 +04/18 15:20:28 - mmengine - INFO - Iter(train) [ 84400/160000] lr: 5.1420e-03 eta: 11:35:35 time: 0.5524 data_time: 0.0067 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0088 decode.acc_seg: 99.6613 aux.loss_ce: 0.0080 aux.acc_seg: 99.2380 +04/18 15:20:55 - mmengine - INFO - Iter(train) [ 84450/160000] lr: 5.1390e-03 eta: 11:35:07 time: 0.5533 data_time: 0.0065 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0073 decode.acc_seg: 99.7334 aux.loss_ce: 0.0073 aux.acc_seg: 99.2209 +04/18 15:21:23 - mmengine - INFO - Iter(train) [ 84500/160000] lr: 5.1360e-03 eta: 11:34:39 time: 0.5533 data_time: 0.0067 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0072 decode.acc_seg: 99.7145 aux.loss_ce: 0.0071 aux.acc_seg: 99.2967 +04/18 15:21:51 - mmengine - INFO - Iter(train) [ 84550/160000] lr: 5.1330e-03 eta: 11:34:12 time: 0.5532 data_time: 0.0062 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0070 decode.acc_seg: 99.6774 aux.loss_ce: 0.0075 aux.acc_seg: 98.9341 +04/18 15:22:18 - mmengine - INFO - Iter(train) [ 84600/160000] lr: 5.1300e-03 eta: 11:33:44 time: 0.5535 data_time: 0.0069 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0080 decode.acc_seg: 99.6275 aux.loss_ce: 0.0082 aux.acc_seg: 99.0343 +04/18 15:22:46 - mmengine - INFO - Iter(train) [ 84650/160000] lr: 5.1270e-03 eta: 11:33:17 time: 0.5525 data_time: 0.0065 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0075 decode.acc_seg: 99.6758 aux.loss_ce: 0.0077 aux.acc_seg: 98.8980 +04/18 15:23:14 - mmengine - INFO - Iter(train) [ 84700/160000] lr: 5.1239e-03 eta: 11:32:49 time: 0.5528 data_time: 0.0061 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0101 decode.acc_seg: 99.6681 aux.loss_ce: 0.0084 aux.acc_seg: 99.2866 +04/18 15:23:42 - mmengine - INFO - Iter(train) [ 84750/160000] lr: 5.1209e-03 eta: 11:32:22 time: 0.5524 data_time: 0.0066 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0080 decode.acc_seg: 99.6438 aux.loss_ce: 0.0078 aux.acc_seg: 99.1773 +04/18 15:24:09 - mmengine - INFO - Iter(train) [ 84800/160000] lr: 5.1179e-03 eta: 11:31:54 time: 0.5625 data_time: 0.0059 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0074 decode.acc_seg: 99.7628 aux.loss_ce: 0.0077 aux.acc_seg: 99.3494 +04/18 15:24:37 - mmengine - INFO - Iter(train) [ 84850/160000] lr: 5.1149e-03 eta: 11:31:27 time: 0.5527 data_time: 0.0063 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0070 decode.acc_seg: 99.7654 aux.loss_ce: 0.0069 aux.acc_seg: 99.4826 +04/18 15:25:05 - mmengine - INFO - Iter(train) [ 84900/160000] lr: 5.1119e-03 eta: 11:30:59 time: 0.5560 data_time: 0.0061 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0079 decode.acc_seg: 99.6209 aux.loss_ce: 0.0080 aux.acc_seg: 98.9641 +04/18 15:25:32 - mmengine - INFO - Iter(train) [ 84950/160000] lr: 5.1089e-03 eta: 11:30:32 time: 0.5538 data_time: 0.0070 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0069 decode.acc_seg: 99.7670 aux.loss_ce: 0.0071 aux.acc_seg: 99.3470 +04/18 15:26:00 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 15:26:00 - mmengine - INFO - Iter(train) [ 85000/160000] lr: 5.1059e-03 eta: 11:30:04 time: 0.5531 data_time: 0.0069 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0082 decode.acc_seg: 99.6502 aux.loss_ce: 0.0082 aux.acc_seg: 98.8712 +04/18 15:26:28 - mmengine - INFO - Iter(train) [ 85050/160000] lr: 5.1029e-03 eta: 11:29:37 time: 0.5550 data_time: 0.0065 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0077 decode.acc_seg: 99.5624 aux.loss_ce: 0.0072 aux.acc_seg: 99.3143 +04/18 15:26:56 - mmengine - INFO - Iter(train) [ 85100/160000] lr: 5.0999e-03 eta: 11:29:09 time: 0.5558 data_time: 0.0065 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0091 decode.acc_seg: 99.5568 aux.loss_ce: 0.0082 aux.acc_seg: 99.1658 +04/18 15:27:23 - mmengine - INFO - Iter(train) [ 85150/160000] lr: 5.0969e-03 eta: 11:28:42 time: 0.5537 data_time: 0.0072 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0076 decode.acc_seg: 99.7728 aux.loss_ce: 0.0075 aux.acc_seg: 99.3503 +04/18 15:27:51 - mmengine - INFO - Iter(train) [ 85200/160000] lr: 5.0939e-03 eta: 11:28:14 time: 0.5544 data_time: 0.0072 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0082 decode.acc_seg: 99.6938 aux.loss_ce: 0.0077 aux.acc_seg: 99.1798 +04/18 15:28:19 - mmengine - INFO - Iter(train) [ 85250/160000] lr: 5.0909e-03 eta: 11:27:47 time: 0.5535 data_time: 0.0063 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0089 decode.acc_seg: 99.6203 aux.loss_ce: 0.0081 aux.acc_seg: 99.1729 +04/18 15:28:47 - mmengine - INFO - Iter(train) [ 85300/160000] lr: 5.0879e-03 eta: 11:27:19 time: 0.5635 data_time: 0.0068 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0074 decode.acc_seg: 99.7382 aux.loss_ce: 0.0076 aux.acc_seg: 99.4714 +04/18 15:29:14 - mmengine - INFO - Iter(train) [ 85350/160000] lr: 5.0849e-03 eta: 11:26:52 time: 0.5533 data_time: 0.0066 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0075 decode.acc_seg: 99.6572 aux.loss_ce: 0.0077 aux.acc_seg: 99.1914 +04/18 15:29:42 - mmengine - INFO - Iter(train) [ 85400/160000] lr: 5.0819e-03 eta: 11:26:24 time: 0.5531 data_time: 0.0066 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0075 decode.acc_seg: 99.7693 aux.loss_ce: 0.0071 aux.acc_seg: 99.3942 +04/18 15:30:10 - mmengine - INFO - Iter(train) [ 85450/160000] lr: 5.0789e-03 eta: 11:25:57 time: 0.5522 data_time: 0.0068 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0069 decode.acc_seg: 99.6550 aux.loss_ce: 0.0071 aux.acc_seg: 99.2377 +04/18 15:30:37 - mmengine - INFO - Iter(train) [ 85500/160000] lr: 5.0759e-03 eta: 11:25:29 time: 0.5536 data_time: 0.0063 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0074 decode.acc_seg: 99.7290 aux.loss_ce: 0.0076 aux.acc_seg: 99.3373 +04/18 15:31:05 - mmengine - INFO - Iter(train) [ 85550/160000] lr: 5.0729e-03 eta: 11:25:02 time: 0.5521 data_time: 0.0065 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0076 decode.acc_seg: 99.6910 aux.loss_ce: 0.0077 aux.acc_seg: 99.3348 +04/18 15:31:33 - mmengine - INFO - Iter(train) [ 85600/160000] lr: 5.0699e-03 eta: 11:24:34 time: 0.5540 data_time: 0.0072 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0079 decode.acc_seg: 99.7341 aux.loss_ce: 0.0076 aux.acc_seg: 99.1769 +04/18 15:32:00 - mmengine - INFO - Iter(train) [ 85650/160000] lr: 5.0669e-03 eta: 11:24:07 time: 0.5543 data_time: 0.0066 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0082 decode.acc_seg: 99.5660 aux.loss_ce: 0.0080 aux.acc_seg: 99.0636 +04/18 15:32:28 - mmengine - INFO - Iter(train) [ 85700/160000] lr: 5.0639e-03 eta: 11:23:39 time: 0.5530 data_time: 0.0063 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0072 decode.acc_seg: 99.7385 aux.loss_ce: 0.0073 aux.acc_seg: 99.3439 +04/18 15:32:56 - mmengine - INFO - Iter(train) [ 85750/160000] lr: 5.0609e-03 eta: 11:23:12 time: 0.5541 data_time: 0.0065 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0069 decode.acc_seg: 99.8116 aux.loss_ce: 0.0075 aux.acc_seg: 99.3443 +04/18 15:33:23 - mmengine - INFO - Iter(train) [ 85800/160000] lr: 5.0578e-03 eta: 11:22:44 time: 0.5526 data_time: 0.0066 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0074 decode.acc_seg: 99.6855 aux.loss_ce: 0.0074 aux.acc_seg: 99.2805 +04/18 15:33:51 - mmengine - INFO - Iter(train) [ 85850/160000] lr: 5.0548e-03 eta: 11:22:16 time: 0.5528 data_time: 0.0073 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0078 decode.acc_seg: 99.7166 aux.loss_ce: 0.0078 aux.acc_seg: 99.2358 +04/18 15:34:19 - mmengine - INFO - Iter(train) [ 85900/160000] lr: 5.0518e-03 eta: 11:21:49 time: 0.5523 data_time: 0.0068 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0071 decode.acc_seg: 99.6441 aux.loss_ce: 0.0072 aux.acc_seg: 99.1346 +04/18 15:34:47 - mmengine - INFO - Iter(train) [ 85950/160000] lr: 5.0488e-03 eta: 11:21:21 time: 0.5537 data_time: 0.0073 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0074 decode.acc_seg: 99.6896 aux.loss_ce: 0.0077 aux.acc_seg: 99.3023 +04/18 15:35:14 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 15:35:14 - mmengine - INFO - Iter(train) [ 86000/160000] lr: 5.0458e-03 eta: 11:20:54 time: 0.5534 data_time: 0.0063 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0072 decode.acc_seg: 99.7148 aux.loss_ce: 0.0075 aux.acc_seg: 99.0259 +04/18 15:35:42 - mmengine - INFO - Iter(train) [ 86050/160000] lr: 5.0428e-03 eta: 11:20:26 time: 0.5533 data_time: 0.0064 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0070 decode.acc_seg: 99.7795 aux.loss_ce: 0.0073 aux.acc_seg: 99.4060 +04/18 15:36:10 - mmengine - INFO - Iter(train) [ 86100/160000] lr: 5.0398e-03 eta: 11:19:59 time: 0.5544 data_time: 0.0070 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0079 decode.acc_seg: 99.5991 aux.loss_ce: 0.0077 aux.acc_seg: 99.0725 +04/18 15:36:37 - mmengine - INFO - Iter(train) [ 86150/160000] lr: 5.0368e-03 eta: 11:19:31 time: 0.5537 data_time: 0.0064 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0065 decode.acc_seg: 99.8203 aux.loss_ce: 0.0070 aux.acc_seg: 99.3801 +04/18 15:37:05 - mmengine - INFO - Iter(train) [ 86200/160000] lr: 5.0338e-03 eta: 11:19:04 time: 0.5538 data_time: 0.0065 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0072 decode.acc_seg: 99.7426 aux.loss_ce: 0.0073 aux.acc_seg: 99.3389 +04/18 15:37:33 - mmengine - INFO - Iter(train) [ 86250/160000] lr: 5.0308e-03 eta: 11:18:36 time: 0.5532 data_time: 0.0070 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0070 decode.acc_seg: 99.7746 aux.loss_ce: 0.0073 aux.acc_seg: 99.3625 +04/18 15:38:00 - mmengine - INFO - Iter(train) [ 86300/160000] lr: 5.0278e-03 eta: 11:18:09 time: 0.5531 data_time: 0.0069 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0073 decode.acc_seg: 99.6654 aux.loss_ce: 0.0072 aux.acc_seg: 99.1960 +04/18 15:38:28 - mmengine - INFO - Iter(train) [ 86350/160000] lr: 5.0248e-03 eta: 11:17:41 time: 0.5529 data_time: 0.0068 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0067 decode.acc_seg: 99.7266 aux.loss_ce: 0.0068 aux.acc_seg: 99.2655 +04/18 15:38:56 - mmengine - INFO - Iter(train) [ 86400/160000] lr: 5.0218e-03 eta: 11:17:14 time: 0.5542 data_time: 0.0075 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0074 decode.acc_seg: 99.6494 aux.loss_ce: 0.0077 aux.acc_seg: 98.9733 +04/18 15:39:24 - mmengine - INFO - Iter(train) [ 86450/160000] lr: 5.0187e-03 eta: 11:16:46 time: 0.5533 data_time: 0.0065 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0071 decode.acc_seg: 99.7817 aux.loss_ce: 0.0074 aux.acc_seg: 99.4470 +04/18 15:39:52 - mmengine - INFO - Iter(train) [ 86500/160000] lr: 5.0157e-03 eta: 11:16:19 time: 0.5537 data_time: 0.0068 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0068 decode.acc_seg: 99.7622 aux.loss_ce: 0.0067 aux.acc_seg: 99.3397 +04/18 15:40:19 - mmengine - INFO - Iter(train) [ 86550/160000] lr: 5.0127e-03 eta: 11:15:51 time: 0.5546 data_time: 0.0074 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0068 decode.acc_seg: 99.7285 aux.loss_ce: 0.0070 aux.acc_seg: 99.3051 +04/18 15:40:47 - mmengine - INFO - Iter(train) [ 86600/160000] lr: 5.0097e-03 eta: 11:15:24 time: 0.5543 data_time: 0.0064 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0066 decode.acc_seg: 99.7585 aux.loss_ce: 0.0075 aux.acc_seg: 99.2744 +04/18 15:41:15 - mmengine - INFO - Iter(train) [ 86650/160000] lr: 5.0067e-03 eta: 11:14:56 time: 0.5522 data_time: 0.0063 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0068 decode.acc_seg: 99.6712 aux.loss_ce: 0.0073 aux.acc_seg: 99.2598 +04/18 15:41:42 - mmengine - INFO - Iter(train) [ 86700/160000] lr: 5.0037e-03 eta: 11:14:29 time: 0.5550 data_time: 0.0069 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0073 decode.acc_seg: 99.7520 aux.loss_ce: 0.0075 aux.acc_seg: 99.2879 +04/18 15:42:10 - mmengine - INFO - Iter(train) [ 86750/160000] lr: 5.0007e-03 eta: 11:14:01 time: 0.5543 data_time: 0.0070 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0069 decode.acc_seg: 99.7194 aux.loss_ce: 0.0071 aux.acc_seg: 99.3279 +04/18 15:42:38 - mmengine - INFO - Iter(train) [ 86800/160000] lr: 4.9977e-03 eta: 11:13:34 time: 0.5544 data_time: 0.0064 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0069 decode.acc_seg: 99.6115 aux.loss_ce: 0.0075 aux.acc_seg: 98.7851 +04/18 15:43:06 - mmengine - INFO - Iter(train) [ 86850/160000] lr: 4.9947e-03 eta: 11:13:06 time: 0.5544 data_time: 0.0061 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0063 decode.acc_seg: 99.7292 aux.loss_ce: 0.0067 aux.acc_seg: 99.2359 +04/18 15:43:33 - mmengine - INFO - Iter(train) [ 86900/160000] lr: 4.9916e-03 eta: 11:12:39 time: 0.5528 data_time: 0.0062 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0072 decode.acc_seg: 99.6617 aux.loss_ce: 0.0074 aux.acc_seg: 99.1792 +04/18 15:44:01 - mmengine - INFO - Iter(train) [ 86950/160000] lr: 4.9886e-03 eta: 11:12:11 time: 0.5533 data_time: 0.0059 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0073 decode.acc_seg: 99.7668 aux.loss_ce: 0.0077 aux.acc_seg: 99.3586 +04/18 15:44:29 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 15:44:29 - mmengine - INFO - Iter(train) [ 87000/160000] lr: 4.9856e-03 eta: 11:11:44 time: 0.5541 data_time: 0.0071 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0066 decode.acc_seg: 99.7103 aux.loss_ce: 0.0068 aux.acc_seg: 99.3274 +04/18 15:44:56 - mmengine - INFO - Iter(train) [ 87050/160000] lr: 4.9826e-03 eta: 11:11:16 time: 0.5550 data_time: 0.0063 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0072 decode.acc_seg: 99.6572 aux.loss_ce: 0.0075 aux.acc_seg: 98.9265 +04/18 15:45:24 - mmengine - INFO - Iter(train) [ 87100/160000] lr: 4.9796e-03 eta: 11:10:49 time: 0.5545 data_time: 0.0067 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0080 decode.acc_seg: 99.6841 aux.loss_ce: 0.0076 aux.acc_seg: 99.2471 +04/18 15:45:52 - mmengine - INFO - Iter(train) [ 87150/160000] lr: 4.9766e-03 eta: 11:10:21 time: 0.5533 data_time: 0.0063 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0074 decode.acc_seg: 99.7620 aux.loss_ce: 0.0071 aux.acc_seg: 99.3556 +04/18 15:46:19 - mmengine - INFO - Iter(train) [ 87200/160000] lr: 4.9736e-03 eta: 11:09:53 time: 0.5535 data_time: 0.0062 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0075 decode.acc_seg: 99.6668 aux.loss_ce: 0.0076 aux.acc_seg: 99.0356 +04/18 15:46:47 - mmengine - INFO - Iter(train) [ 87250/160000] lr: 4.9706e-03 eta: 11:09:26 time: 0.5544 data_time: 0.0066 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0074 decode.acc_seg: 99.7335 aux.loss_ce: 0.0072 aux.acc_seg: 99.2883 +04/18 15:47:15 - mmengine - INFO - Iter(train) [ 87300/160000] lr: 4.9676e-03 eta: 11:08:58 time: 0.5532 data_time: 0.0061 memory: 7635 loss: 0.0219 decode.loss_ce: 0.0124 decode.acc_seg: 99.3621 aux.loss_ce: 0.0095 aux.acc_seg: 99.0041 +04/18 15:47:43 - mmengine - INFO - Iter(train) [ 87350/160000] lr: 4.9645e-03 eta: 11:08:31 time: 0.5531 data_time: 0.0067 memory: 7635 loss: 0.0231 decode.loss_ce: 0.0136 decode.acc_seg: 99.5957 aux.loss_ce: 0.0095 aux.acc_seg: 99.2460 +04/18 15:48:10 - mmengine - INFO - Iter(train) [ 87400/160000] lr: 4.9615e-03 eta: 11:08:03 time: 0.5546 data_time: 0.0066 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0102 decode.acc_seg: 99.7156 aux.loss_ce: 0.0086 aux.acc_seg: 99.2757 +04/18 15:48:38 - mmengine - INFO - Iter(train) [ 87450/160000] lr: 4.9585e-03 eta: 11:07:36 time: 0.5541 data_time: 0.0070 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0089 decode.acc_seg: 99.8028 aux.loss_ce: 0.0080 aux.acc_seg: 99.5077 +04/18 15:49:06 - mmengine - INFO - Iter(train) [ 87500/160000] lr: 4.9555e-03 eta: 11:07:08 time: 0.5541 data_time: 0.0066 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0087 decode.acc_seg: 99.6773 aux.loss_ce: 0.0080 aux.acc_seg: 99.0063 +04/18 15:49:34 - mmengine - INFO - Iter(train) [ 87550/160000] lr: 4.9525e-03 eta: 11:06:41 time: 0.5524 data_time: 0.0065 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0080 decode.acc_seg: 99.8137 aux.loss_ce: 0.0074 aux.acc_seg: 99.5594 +04/18 15:50:01 - mmengine - INFO - Iter(train) [ 87600/160000] lr: 4.9495e-03 eta: 11:06:13 time: 0.5531 data_time: 0.0069 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0076 decode.acc_seg: 99.7032 aux.loss_ce: 0.0075 aux.acc_seg: 99.2753 +04/18 15:50:29 - mmengine - INFO - Iter(train) [ 87650/160000] lr: 4.9465e-03 eta: 11:05:46 time: 0.5525 data_time: 0.0063 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0086 decode.acc_seg: 99.7018 aux.loss_ce: 0.0078 aux.acc_seg: 99.2401 +04/18 15:50:57 - mmengine - INFO - Iter(train) [ 87700/160000] lr: 4.9434e-03 eta: 11:05:18 time: 0.5522 data_time: 0.0062 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0072 decode.acc_seg: 99.7213 aux.loss_ce: 0.0070 aux.acc_seg: 99.3948 +04/18 15:51:24 - mmengine - INFO - Iter(train) [ 87750/160000] lr: 4.9404e-03 eta: 11:04:51 time: 0.5536 data_time: 0.0063 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0083 decode.acc_seg: 99.6844 aux.loss_ce: 0.0074 aux.acc_seg: 99.4108 +04/18 15:51:52 - mmengine - INFO - Iter(train) [ 87800/160000] lr: 4.9374e-03 eta: 11:04:23 time: 0.5544 data_time: 0.0077 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0070 decode.acc_seg: 99.7691 aux.loss_ce: 0.0076 aux.acc_seg: 99.2150 +04/18 15:52:20 - mmengine - INFO - Iter(train) [ 87850/160000] lr: 4.9344e-03 eta: 11:03:56 time: 0.5553 data_time: 0.0070 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0069 decode.acc_seg: 99.7274 aux.loss_ce: 0.0071 aux.acc_seg: 99.1962 +04/18 15:52:48 - mmengine - INFO - Iter(train) [ 87900/160000] lr: 4.9314e-03 eta: 11:03:28 time: 0.5538 data_time: 0.0070 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0083 decode.acc_seg: 99.6800 aux.loss_ce: 0.0080 aux.acc_seg: 99.1891 +04/18 15:53:15 - mmengine - INFO - Iter(train) [ 87950/160000] lr: 4.9284e-03 eta: 11:03:01 time: 0.5535 data_time: 0.0064 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0072 decode.acc_seg: 99.7053 aux.loss_ce: 0.0070 aux.acc_seg: 99.3465 +04/18 15:53:43 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 15:53:43 - mmengine - INFO - Iter(train) [ 88000/160000] lr: 4.9254e-03 eta: 11:02:33 time: 0.5528 data_time: 0.0067 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0074 decode.acc_seg: 99.7036 aux.loss_ce: 0.0072 aux.acc_seg: 99.0846 +04/18 15:54:11 - mmengine - INFO - Iter(train) [ 88050/160000] lr: 4.9223e-03 eta: 11:02:06 time: 0.5529 data_time: 0.0066 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0084 decode.acc_seg: 99.4520 aux.loss_ce: 0.0082 aux.acc_seg: 99.1279 +04/18 15:54:38 - mmengine - INFO - Iter(train) [ 88100/160000] lr: 4.9193e-03 eta: 11:01:38 time: 0.5519 data_time: 0.0064 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0070 decode.acc_seg: 99.6961 aux.loss_ce: 0.0071 aux.acc_seg: 99.1393 +04/18 15:55:06 - mmengine - INFO - Iter(train) [ 88150/160000] lr: 4.9163e-03 eta: 11:01:11 time: 0.5549 data_time: 0.0067 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0074 decode.acc_seg: 99.7077 aux.loss_ce: 0.0074 aux.acc_seg: 99.4089 +04/18 15:55:34 - mmengine - INFO - Iter(train) [ 88200/160000] lr: 4.9133e-03 eta: 11:00:43 time: 0.5541 data_time: 0.0065 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0076 decode.acc_seg: 99.7110 aux.loss_ce: 0.0077 aux.acc_seg: 99.1214 +04/18 15:56:02 - mmengine - INFO - Iter(train) [ 88250/160000] lr: 4.9103e-03 eta: 11:00:16 time: 0.5534 data_time: 0.0060 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0080 decode.acc_seg: 99.6861 aux.loss_ce: 0.0077 aux.acc_seg: 99.0960 +04/18 15:56:29 - mmengine - INFO - Iter(train) [ 88300/160000] lr: 4.9073e-03 eta: 10:59:48 time: 0.5533 data_time: 0.0066 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0071 decode.acc_seg: 99.7285 aux.loss_ce: 0.0070 aux.acc_seg: 99.1835 +04/18 15:56:57 - mmengine - INFO - Iter(train) [ 88350/160000] lr: 4.9042e-03 eta: 10:59:20 time: 0.5532 data_time: 0.0065 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0064 decode.acc_seg: 99.7697 aux.loss_ce: 0.0070 aux.acc_seg: 99.2548 +04/18 15:57:25 - mmengine - INFO - Iter(train) [ 88400/160000] lr: 4.9012e-03 eta: 10:58:53 time: 0.5537 data_time: 0.0070 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0077 decode.acc_seg: 99.7449 aux.loss_ce: 0.0078 aux.acc_seg: 99.2819 +04/18 15:57:52 - mmengine - INFO - Iter(train) [ 88450/160000] lr: 4.8982e-03 eta: 10:58:25 time: 0.5517 data_time: 0.0064 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0074 decode.acc_seg: 99.7245 aux.loss_ce: 0.0074 aux.acc_seg: 99.2621 +04/18 15:58:20 - mmengine - INFO - Iter(train) [ 88500/160000] lr: 4.8952e-03 eta: 10:57:58 time: 0.5541 data_time: 0.0067 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0072 decode.acc_seg: 99.7017 aux.loss_ce: 0.0074 aux.acc_seg: 99.0706 +04/18 15:58:48 - mmengine - INFO - Iter(train) [ 88550/160000] lr: 4.8922e-03 eta: 10:57:31 time: 0.5539 data_time: 0.0064 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0086 decode.acc_seg: 99.5434 aux.loss_ce: 0.0077 aux.acc_seg: 99.1254 +04/18 15:59:16 - mmengine - INFO - Iter(train) [ 88600/160000] lr: 4.8891e-03 eta: 10:57:03 time: 0.5532 data_time: 0.0063 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0080 decode.acc_seg: 99.7069 aux.loss_ce: 0.0082 aux.acc_seg: 98.9732 +04/18 15:59:44 - mmengine - INFO - Iter(train) [ 88650/160000] lr: 4.8861e-03 eta: 10:56:35 time: 0.5542 data_time: 0.0063 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0076 decode.acc_seg: 99.7394 aux.loss_ce: 0.0076 aux.acc_seg: 99.3766 +04/18 16:00:11 - mmengine - INFO - Iter(train) [ 88700/160000] lr: 4.8831e-03 eta: 10:56:08 time: 0.5540 data_time: 0.0062 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0073 decode.acc_seg: 99.6796 aux.loss_ce: 0.0073 aux.acc_seg: 99.2290 +04/18 16:00:39 - mmengine - INFO - Iter(train) [ 88750/160000] lr: 4.8801e-03 eta: 10:55:40 time: 0.5541 data_time: 0.0066 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0068 decode.acc_seg: 99.7298 aux.loss_ce: 0.0069 aux.acc_seg: 99.1305 +04/18 16:01:07 - mmengine - INFO - Iter(train) [ 88800/160000] lr: 4.8771e-03 eta: 10:55:13 time: 0.5526 data_time: 0.0066 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0077 decode.acc_seg: 99.5977 aux.loss_ce: 0.0077 aux.acc_seg: 99.0127 +04/18 16:01:34 - mmengine - INFO - Iter(train) [ 88850/160000] lr: 4.8741e-03 eta: 10:54:45 time: 0.5544 data_time: 0.0066 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0093 decode.acc_seg: 99.5275 aux.loss_ce: 0.0084 aux.acc_seg: 99.0695 +04/18 16:02:02 - mmengine - INFO - Iter(train) [ 88900/160000] lr: 4.8710e-03 eta: 10:54:18 time: 0.5525 data_time: 0.0060 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0074 decode.acc_seg: 99.6779 aux.loss_ce: 0.0078 aux.acc_seg: 99.1474 +04/18 16:02:30 - mmengine - INFO - Iter(train) [ 88950/160000] lr: 4.8680e-03 eta: 10:53:50 time: 0.5552 data_time: 0.0072 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0081 decode.acc_seg: 99.6627 aux.loss_ce: 0.0080 aux.acc_seg: 99.1783 +04/18 16:02:58 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 16:02:58 - mmengine - INFO - Iter(train) [ 89000/160000] lr: 4.8650e-03 eta: 10:53:23 time: 0.5524 data_time: 0.0067 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0067 decode.acc_seg: 99.7494 aux.loss_ce: 0.0069 aux.acc_seg: 99.2952 +04/18 16:03:25 - mmengine - INFO - Iter(train) [ 89050/160000] lr: 4.8620e-03 eta: 10:52:55 time: 0.5539 data_time: 0.0073 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0078 decode.acc_seg: 99.6391 aux.loss_ce: 0.0079 aux.acc_seg: 99.3408 +04/18 16:03:53 - mmengine - INFO - Iter(train) [ 89100/160000] lr: 4.8590e-03 eta: 10:52:28 time: 0.5538 data_time: 0.0061 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0090 decode.acc_seg: 99.6519 aux.loss_ce: 0.0085 aux.acc_seg: 99.1333 +04/18 16:04:21 - mmengine - INFO - Iter(train) [ 89150/160000] lr: 4.8559e-03 eta: 10:52:00 time: 0.5530 data_time: 0.0060 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0080 decode.acc_seg: 99.6053 aux.loss_ce: 0.0078 aux.acc_seg: 99.1386 +04/18 16:04:48 - mmengine - INFO - Iter(train) [ 89200/160000] lr: 4.8529e-03 eta: 10:51:33 time: 0.5539 data_time: 0.0065 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0081 decode.acc_seg: 99.7516 aux.loss_ce: 0.0079 aux.acc_seg: 99.3876 +04/18 16:05:16 - mmengine - INFO - Iter(train) [ 89250/160000] lr: 4.8499e-03 eta: 10:51:05 time: 0.5555 data_time: 0.0065 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0073 decode.acc_seg: 99.6994 aux.loss_ce: 0.0074 aux.acc_seg: 99.2689 +04/18 16:05:44 - mmengine - INFO - Iter(train) [ 89300/160000] lr: 4.8469e-03 eta: 10:50:38 time: 0.5549 data_time: 0.0067 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0076 decode.acc_seg: 99.6260 aux.loss_ce: 0.0074 aux.acc_seg: 99.0972 +04/18 16:06:12 - mmengine - INFO - Iter(train) [ 89350/160000] lr: 4.8438e-03 eta: 10:50:10 time: 0.5546 data_time: 0.0062 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0071 decode.acc_seg: 99.6920 aux.loss_ce: 0.0075 aux.acc_seg: 99.0008 +04/18 16:06:39 - mmengine - INFO - Iter(train) [ 89400/160000] lr: 4.8408e-03 eta: 10:49:43 time: 0.5534 data_time: 0.0066 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0066 decode.acc_seg: 99.6706 aux.loss_ce: 0.0069 aux.acc_seg: 99.0152 +04/18 16:07:07 - mmengine - INFO - Iter(train) [ 89450/160000] lr: 4.8378e-03 eta: 10:49:15 time: 0.5539 data_time: 0.0074 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0073 decode.acc_seg: 99.7231 aux.loss_ce: 0.0075 aux.acc_seg: 99.2122 +04/18 16:07:35 - mmengine - INFO - Iter(train) [ 89500/160000] lr: 4.8348e-03 eta: 10:48:47 time: 0.5540 data_time: 0.0066 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0069 decode.acc_seg: 99.7463 aux.loss_ce: 0.0068 aux.acc_seg: 99.3187 +04/18 16:08:03 - mmengine - INFO - Iter(train) [ 89550/160000] lr: 4.8318e-03 eta: 10:48:20 time: 0.5644 data_time: 0.0068 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0070 decode.acc_seg: 99.7202 aux.loss_ce: 0.0073 aux.acc_seg: 99.1396 +04/18 16:08:30 - mmengine - INFO - Iter(train) [ 89600/160000] lr: 4.8287e-03 eta: 10:47:53 time: 0.5543 data_time: 0.0071 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0066 decode.acc_seg: 99.7922 aux.loss_ce: 0.0070 aux.acc_seg: 99.4041 +04/18 16:08:58 - mmengine - INFO - Iter(train) [ 89650/160000] lr: 4.8257e-03 eta: 10:47:25 time: 0.5531 data_time: 0.0069 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0068 decode.acc_seg: 99.7718 aux.loss_ce: 0.0072 aux.acc_seg: 99.2868 +04/18 16:09:26 - mmengine - INFO - Iter(train) [ 89700/160000] lr: 4.8227e-03 eta: 10:46:58 time: 0.5551 data_time: 0.0080 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0066 decode.acc_seg: 99.6888 aux.loss_ce: 0.0069 aux.acc_seg: 99.3211 +04/18 16:09:54 - mmengine - INFO - Iter(train) [ 89750/160000] lr: 4.8197e-03 eta: 10:46:30 time: 0.5533 data_time: 0.0068 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0076 decode.acc_seg: 99.5870 aux.loss_ce: 0.0075 aux.acc_seg: 98.8684 +04/18 16:10:21 - mmengine - INFO - Iter(train) [ 89800/160000] lr: 4.8166e-03 eta: 10:46:02 time: 0.5538 data_time: 0.0066 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0081 decode.acc_seg: 99.8167 aux.loss_ce: 0.0072 aux.acc_seg: 99.5503 +04/18 16:10:49 - mmengine - INFO - Iter(train) [ 89850/160000] lr: 4.8136e-03 eta: 10:45:35 time: 0.5539 data_time: 0.0064 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0085 decode.acc_seg: 99.7402 aux.loss_ce: 0.0085 aux.acc_seg: 99.2199 +04/18 16:11:17 - mmengine - INFO - Iter(train) [ 89900/160000] lr: 4.8106e-03 eta: 10:45:07 time: 0.5541 data_time: 0.0062 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0090 decode.acc_seg: 99.6343 aux.loss_ce: 0.0075 aux.acc_seg: 99.0417 +04/18 16:11:44 - mmengine - INFO - Iter(train) [ 89950/160000] lr: 4.8076e-03 eta: 10:44:40 time: 0.5520 data_time: 0.0062 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0092 decode.acc_seg: 99.6977 aux.loss_ce: 0.0081 aux.acc_seg: 99.2770 +04/18 16:12:12 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 16:12:12 - mmengine - INFO - Iter(train) [ 90000/160000] lr: 4.8045e-03 eta: 10:44:12 time: 0.5534 data_time: 0.0065 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0084 decode.acc_seg: 99.6110 aux.loss_ce: 0.0079 aux.acc_seg: 98.7445 +04/18 16:12:12 - mmengine - INFO - Saving checkpoint at 90000 iterations +04/18 16:12:16 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:07 time: 0.0469 data_time: 0.0016 memory: 1657 +04/18 16:12:18 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:04 time: 0.0467 data_time: 0.0013 memory: 1657 +04/18 16:12:21 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:02 time: 0.0468 data_time: 0.0014 memory: 1657 +04/18 16:12:23 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.0451 data_time: 0.0012 memory: 1657 +04/18 16:12:23 - mmengine - INFO - per class results: +04/18 16:12:23 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 98.99 | 99.43 | 99.49 | 99.55 | 99.43 | +| contrast | 78.56 | 89.25 | 87.99 | 86.76 | 89.25 | ++------------+-------+-------+--------+-----------+--------+ +04/18 16:12:23 - mmengine - INFO - Iter(val) [200/200] aAcc: 99.0300 mIoU: 88.7700 mAcc: 94.3400 mFscore: 93.7400 mPrecision: 93.1600 mRecall: 94.3400 data_time: 0.0016 time: 0.0468 +04/18 16:12:51 - mmengine - INFO - Iter(train) [ 90050/160000] lr: 4.8015e-03 eta: 10:43:45 time: 0.5541 data_time: 0.0073 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0088 decode.acc_seg: 99.7517 aux.loss_ce: 0.0079 aux.acc_seg: 99.4351 +04/18 16:13:19 - mmengine - INFO - Iter(train) [ 90100/160000] lr: 4.7985e-03 eta: 10:43:17 time: 0.5540 data_time: 0.0068 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0080 decode.acc_seg: 99.7671 aux.loss_ce: 0.0075 aux.acc_seg: 99.3597 +04/18 16:13:46 - mmengine - INFO - Iter(train) [ 90150/160000] lr: 4.7955e-03 eta: 10:42:50 time: 0.5537 data_time: 0.0068 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0089 decode.acc_seg: 99.6665 aux.loss_ce: 0.0081 aux.acc_seg: 99.1957 +04/18 16:14:14 - mmengine - INFO - Iter(train) [ 90200/160000] lr: 4.7924e-03 eta: 10:42:22 time: 0.5526 data_time: 0.0057 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0072 decode.acc_seg: 99.6493 aux.loss_ce: 0.0070 aux.acc_seg: 99.1049 +04/18 16:14:42 - mmengine - INFO - Iter(train) [ 90250/160000] lr: 4.7894e-03 eta: 10:41:55 time: 0.5531 data_time: 0.0066 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0081 decode.acc_seg: 99.6822 aux.loss_ce: 0.0083 aux.acc_seg: 99.0961 +04/18 16:15:10 - mmengine - INFO - Iter(train) [ 90300/160000] lr: 4.7864e-03 eta: 10:41:27 time: 0.5532 data_time: 0.0068 memory: 7635 loss: 0.0182 decode.loss_ce: 0.0098 decode.acc_seg: 99.6121 aux.loss_ce: 0.0084 aux.acc_seg: 99.2744 +04/18 16:15:37 - mmengine - INFO - Iter(train) [ 90350/160000] lr: 4.7834e-03 eta: 10:41:00 time: 0.5539 data_time: 0.0065 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0068 decode.acc_seg: 99.6592 aux.loss_ce: 0.0067 aux.acc_seg: 99.2595 +04/18 16:16:05 - mmengine - INFO - Iter(train) [ 90400/160000] lr: 4.7803e-03 eta: 10:40:32 time: 0.5538 data_time: 0.0071 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0070 decode.acc_seg: 99.7413 aux.loss_ce: 0.0075 aux.acc_seg: 99.2717 +04/18 16:16:33 - mmengine - INFO - Iter(train) [ 90450/160000] lr: 4.7773e-03 eta: 10:40:04 time: 0.5532 data_time: 0.0066 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0072 decode.acc_seg: 99.7602 aux.loss_ce: 0.0075 aux.acc_seg: 99.4248 +04/18 16:17:00 - mmengine - INFO - Iter(train) [ 90500/160000] lr: 4.7743e-03 eta: 10:39:37 time: 0.5643 data_time: 0.0076 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0075 decode.acc_seg: 99.6515 aux.loss_ce: 0.0080 aux.acc_seg: 99.1366 +04/18 16:17:28 - mmengine - INFO - Iter(train) [ 90550/160000] lr: 4.7713e-03 eta: 10:39:09 time: 0.5543 data_time: 0.0070 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0078 decode.acc_seg: 99.6815 aux.loss_ce: 0.0075 aux.acc_seg: 99.3804 +04/18 16:17:56 - mmengine - INFO - Iter(train) [ 90600/160000] lr: 4.7682e-03 eta: 10:38:42 time: 0.5634 data_time: 0.0065 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0077 decode.acc_seg: 99.7453 aux.loss_ce: 0.0082 aux.acc_seg: 99.2341 +04/18 16:18:24 - mmengine - INFO - Iter(train) [ 90650/160000] lr: 4.7652e-03 eta: 10:38:15 time: 0.5551 data_time: 0.0066 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0073 decode.acc_seg: 99.7275 aux.loss_ce: 0.0076 aux.acc_seg: 99.4253 +04/18 16:18:52 - mmengine - INFO - Iter(train) [ 90700/160000] lr: 4.7622e-03 eta: 10:37:47 time: 0.5541 data_time: 0.0063 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0081 decode.acc_seg: 99.6181 aux.loss_ce: 0.0077 aux.acc_seg: 99.0476 +04/18 16:19:19 - mmengine - INFO - Iter(train) [ 90750/160000] lr: 4.7592e-03 eta: 10:37:20 time: 0.5537 data_time: 0.0061 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0093 decode.acc_seg: 99.5783 aux.loss_ce: 0.0081 aux.acc_seg: 99.0585 +04/18 16:19:47 - mmengine - INFO - Iter(train) [ 90800/160000] lr: 4.7561e-03 eta: 10:36:52 time: 0.5549 data_time: 0.0072 memory: 7635 loss: 0.0185 decode.loss_ce: 0.0101 decode.acc_seg: 99.7410 aux.loss_ce: 0.0084 aux.acc_seg: 99.3400 +04/18 16:20:15 - mmengine - INFO - Iter(train) [ 90850/160000] lr: 4.7531e-03 eta: 10:36:24 time: 0.5552 data_time: 0.0066 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0081 decode.acc_seg: 99.6765 aux.loss_ce: 0.0080 aux.acc_seg: 99.1732 +04/18 16:20:42 - mmengine - INFO - Iter(train) [ 90900/160000] lr: 4.7501e-03 eta: 10:35:57 time: 0.5535 data_time: 0.0064 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0081 decode.acc_seg: 99.6199 aux.loss_ce: 0.0074 aux.acc_seg: 99.3221 +04/18 16:21:10 - mmengine - INFO - Iter(train) [ 90950/160000] lr: 4.7470e-03 eta: 10:35:29 time: 0.5540 data_time: 0.0062 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0088 decode.acc_seg: 99.5703 aux.loss_ce: 0.0079 aux.acc_seg: 99.0297 +04/18 16:21:38 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 16:21:38 - mmengine - INFO - Iter(train) [ 91000/160000] lr: 4.7440e-03 eta: 10:35:02 time: 0.5558 data_time: 0.0069 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0085 decode.acc_seg: 99.5575 aux.loss_ce: 0.0081 aux.acc_seg: 99.2346 +04/18 16:22:05 - mmengine - INFO - Iter(train) [ 91050/160000] lr: 4.7410e-03 eta: 10:34:34 time: 0.5534 data_time: 0.0071 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0092 decode.acc_seg: 99.6915 aux.loss_ce: 0.0084 aux.acc_seg: 99.2305 +04/18 16:22:33 - mmengine - INFO - Iter(train) [ 91100/160000] lr: 4.7380e-03 eta: 10:34:07 time: 0.5547 data_time: 0.0065 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0072 decode.acc_seg: 99.6708 aux.loss_ce: 0.0072 aux.acc_seg: 99.2778 +04/18 16:23:01 - mmengine - INFO - Iter(train) [ 91150/160000] lr: 4.7349e-03 eta: 10:33:39 time: 0.5542 data_time: 0.0067 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0072 decode.acc_seg: 99.7328 aux.loss_ce: 0.0074 aux.acc_seg: 99.3443 +04/18 16:23:29 - mmengine - INFO - Iter(train) [ 91200/160000] lr: 4.7319e-03 eta: 10:33:12 time: 0.5562 data_time: 0.0079 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0068 decode.acc_seg: 99.7856 aux.loss_ce: 0.0068 aux.acc_seg: 99.3442 +04/18 16:23:56 - mmengine - INFO - Iter(train) [ 91250/160000] lr: 4.7289e-03 eta: 10:32:44 time: 0.5530 data_time: 0.0060 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0072 decode.acc_seg: 99.7177 aux.loss_ce: 0.0073 aux.acc_seg: 99.3159 +04/18 16:24:24 - mmengine - INFO - Iter(train) [ 91300/160000] lr: 4.7258e-03 eta: 10:32:17 time: 0.5543 data_time: 0.0070 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0083 decode.acc_seg: 99.7583 aux.loss_ce: 0.0074 aux.acc_seg: 99.3390 +04/18 16:24:52 - mmengine - INFO - Iter(train) [ 91350/160000] lr: 4.7228e-03 eta: 10:31:49 time: 0.5560 data_time: 0.0063 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0088 decode.acc_seg: 99.7421 aux.loss_ce: 0.0084 aux.acc_seg: 99.2332 +04/18 16:25:19 - mmengine - INFO - Iter(train) [ 91400/160000] lr: 4.7198e-03 eta: 10:31:21 time: 0.5546 data_time: 0.0067 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0075 decode.acc_seg: 99.7305 aux.loss_ce: 0.0076 aux.acc_seg: 99.3952 +04/18 16:25:47 - mmengine - INFO - Iter(train) [ 91450/160000] lr: 4.7168e-03 eta: 10:30:54 time: 0.5536 data_time: 0.0065 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0087 decode.acc_seg: 99.7017 aux.loss_ce: 0.0080 aux.acc_seg: 99.1828 +04/18 16:26:15 - mmengine - INFO - Iter(train) [ 91500/160000] lr: 4.7137e-03 eta: 10:30:26 time: 0.5532 data_time: 0.0070 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0071 decode.acc_seg: 99.7032 aux.loss_ce: 0.0073 aux.acc_seg: 99.1995 +04/18 16:26:43 - mmengine - INFO - Iter(train) [ 91550/160000] lr: 4.7107e-03 eta: 10:29:59 time: 0.5549 data_time: 0.0060 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0073 decode.acc_seg: 99.7380 aux.loss_ce: 0.0075 aux.acc_seg: 99.1542 +04/18 16:27:10 - mmengine - INFO - Iter(train) [ 91600/160000] lr: 4.7077e-03 eta: 10:29:31 time: 0.5535 data_time: 0.0067 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0073 decode.acc_seg: 99.6866 aux.loss_ce: 0.0074 aux.acc_seg: 99.2903 +04/18 16:27:38 - mmengine - INFO - Iter(train) [ 91650/160000] lr: 4.7046e-03 eta: 10:29:04 time: 0.5541 data_time: 0.0067 memory: 7635 loss: 0.0132 decode.loss_ce: 0.0065 decode.acc_seg: 99.7046 aux.loss_ce: 0.0067 aux.acc_seg: 99.1618 +04/18 16:28:06 - mmengine - INFO - Iter(train) [ 91700/160000] lr: 4.7016e-03 eta: 10:28:36 time: 0.5545 data_time: 0.0062 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0067 decode.acc_seg: 99.7621 aux.loss_ce: 0.0069 aux.acc_seg: 99.4017 +04/18 16:28:34 - mmengine - INFO - Iter(train) [ 91750/160000] lr: 4.6986e-03 eta: 10:28:09 time: 0.5532 data_time: 0.0061 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0074 decode.acc_seg: 99.6921 aux.loss_ce: 0.0076 aux.acc_seg: 99.2706 +04/18 16:29:02 - mmengine - INFO - Iter(train) [ 91800/160000] lr: 4.6955e-03 eta: 10:27:41 time: 0.5538 data_time: 0.0069 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0073 decode.acc_seg: 99.7231 aux.loss_ce: 0.0073 aux.acc_seg: 99.1256 +04/18 16:29:29 - mmengine - INFO - Iter(train) [ 91850/160000] lr: 4.6925e-03 eta: 10:27:14 time: 0.5539 data_time: 0.0068 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0071 decode.acc_seg: 99.7126 aux.loss_ce: 0.0073 aux.acc_seg: 99.1622 +04/18 16:29:57 - mmengine - INFO - Iter(train) [ 91900/160000] lr: 4.6895e-03 eta: 10:26:46 time: 0.5536 data_time: 0.0068 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0068 decode.acc_seg: 99.7128 aux.loss_ce: 0.0068 aux.acc_seg: 99.2221 +04/18 16:30:25 - mmengine - INFO - Iter(train) [ 91950/160000] lr: 4.6864e-03 eta: 10:26:19 time: 0.5540 data_time: 0.0067 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0081 decode.acc_seg: 99.5994 aux.loss_ce: 0.0077 aux.acc_seg: 99.2487 +04/18 16:30:53 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 16:30:53 - mmengine - INFO - Iter(train) [ 92000/160000] lr: 4.6834e-03 eta: 10:25:51 time: 0.5554 data_time: 0.0074 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0068 decode.acc_seg: 99.6864 aux.loss_ce: 0.0069 aux.acc_seg: 99.2065 +04/18 16:31:20 - mmengine - INFO - Iter(train) [ 92050/160000] lr: 4.6804e-03 eta: 10:25:24 time: 0.5554 data_time: 0.0072 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0065 decode.acc_seg: 99.7125 aux.loss_ce: 0.0069 aux.acc_seg: 99.1871 +04/18 16:31:48 - mmengine - INFO - Iter(train) [ 92100/160000] lr: 4.6773e-03 eta: 10:24:56 time: 0.5545 data_time: 0.0065 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0070 decode.acc_seg: 99.6840 aux.loss_ce: 0.0077 aux.acc_seg: 99.2039 +04/18 16:32:16 - mmengine - INFO - Iter(train) [ 92150/160000] lr: 4.6743e-03 eta: 10:24:29 time: 0.5534 data_time: 0.0069 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0069 decode.acc_seg: 99.7873 aux.loss_ce: 0.0072 aux.acc_seg: 99.4144 +04/18 16:32:43 - mmengine - INFO - Iter(train) [ 92200/160000] lr: 4.6713e-03 eta: 10:24:01 time: 0.5551 data_time: 0.0072 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0067 decode.acc_seg: 99.7953 aux.loss_ce: 0.0069 aux.acc_seg: 99.4600 +04/18 16:33:11 - mmengine - INFO - Iter(train) [ 92250/160000] lr: 4.6682e-03 eta: 10:23:34 time: 0.5552 data_time: 0.0069 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0073 decode.acc_seg: 99.6154 aux.loss_ce: 0.0075 aux.acc_seg: 99.2498 +04/18 16:33:39 - mmengine - INFO - Iter(train) [ 92300/160000] lr: 4.6652e-03 eta: 10:23:06 time: 0.5536 data_time: 0.0062 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0073 decode.acc_seg: 99.7125 aux.loss_ce: 0.0073 aux.acc_seg: 99.1919 +04/18 16:34:07 - mmengine - INFO - Iter(train) [ 92350/160000] lr: 4.6622e-03 eta: 10:22:39 time: 0.5560 data_time: 0.0071 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0073 decode.acc_seg: 99.7135 aux.loss_ce: 0.0074 aux.acc_seg: 99.3181 +04/18 16:34:34 - mmengine - INFO - Iter(train) [ 92400/160000] lr: 4.6591e-03 eta: 10:22:11 time: 0.5533 data_time: 0.0065 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0071 decode.acc_seg: 99.5176 aux.loss_ce: 0.0074 aux.acc_seg: 99.0997 +04/18 16:35:02 - mmengine - INFO - Iter(train) [ 92450/160000] lr: 4.6561e-03 eta: 10:21:44 time: 0.5544 data_time: 0.0068 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0076 decode.acc_seg: 99.6918 aux.loss_ce: 0.0076 aux.acc_seg: 99.3507 +04/18 16:35:30 - mmengine - INFO - Iter(train) [ 92500/160000] lr: 4.6531e-03 eta: 10:21:16 time: 0.5534 data_time: 0.0061 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0069 decode.acc_seg: 99.7289 aux.loss_ce: 0.0069 aux.acc_seg: 99.3187 +04/18 16:35:58 - mmengine - INFO - Iter(train) [ 92550/160000] lr: 4.6500e-03 eta: 10:20:48 time: 0.5542 data_time: 0.0070 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0071 decode.acc_seg: 99.5818 aux.loss_ce: 0.0071 aux.acc_seg: 98.9644 +04/18 16:36:25 - mmengine - INFO - Iter(train) [ 92600/160000] lr: 4.6470e-03 eta: 10:20:21 time: 0.5542 data_time: 0.0069 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0066 decode.acc_seg: 99.7667 aux.loss_ce: 0.0068 aux.acc_seg: 99.2359 +04/18 16:36:53 - mmengine - INFO - Iter(train) [ 92650/160000] lr: 4.6439e-03 eta: 10:19:53 time: 0.5530 data_time: 0.0065 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0089 decode.acc_seg: 99.7192 aux.loss_ce: 0.0080 aux.acc_seg: 99.1530 +04/18 16:37:21 - mmengine - INFO - Iter(train) [ 92700/160000] lr: 4.6409e-03 eta: 10:19:26 time: 0.5541 data_time: 0.0065 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0073 decode.acc_seg: 99.6608 aux.loss_ce: 0.0073 aux.acc_seg: 99.2210 +04/18 16:37:48 - mmengine - INFO - Iter(train) [ 92750/160000] lr: 4.6379e-03 eta: 10:18:58 time: 0.5545 data_time: 0.0065 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0077 decode.acc_seg: 99.6756 aux.loss_ce: 0.0074 aux.acc_seg: 99.1357 +04/18 16:38:16 - mmengine - INFO - Iter(train) [ 92800/160000] lr: 4.6348e-03 eta: 10:18:31 time: 0.5544 data_time: 0.0071 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0076 decode.acc_seg: 99.6621 aux.loss_ce: 0.0077 aux.acc_seg: 99.0524 +04/18 16:38:44 - mmengine - INFO - Iter(train) [ 92850/160000] lr: 4.6318e-03 eta: 10:18:03 time: 0.5539 data_time: 0.0070 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0084 decode.acc_seg: 99.7317 aux.loss_ce: 0.0078 aux.acc_seg: 99.3055 +04/18 16:39:12 - mmengine - INFO - Iter(train) [ 92900/160000] lr: 4.6288e-03 eta: 10:17:36 time: 0.5529 data_time: 0.0072 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0089 decode.acc_seg: 99.7147 aux.loss_ce: 0.0083 aux.acc_seg: 99.0110 +04/18 16:39:40 - mmengine - INFO - Iter(train) [ 92950/160000] lr: 4.6257e-03 eta: 10:17:08 time: 0.5548 data_time: 0.0063 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0072 decode.acc_seg: 99.6904 aux.loss_ce: 0.0073 aux.acc_seg: 99.1941 +04/18 16:40:07 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 16:40:07 - mmengine - INFO - Iter(train) [ 93000/160000] lr: 4.6227e-03 eta: 10:16:41 time: 0.5559 data_time: 0.0069 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0077 decode.acc_seg: 99.7231 aux.loss_ce: 0.0076 aux.acc_seg: 99.0689 +04/18 16:40:35 - mmengine - INFO - Iter(train) [ 93050/160000] lr: 4.6197e-03 eta: 10:16:13 time: 0.5521 data_time: 0.0063 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0067 decode.acc_seg: 99.7027 aux.loss_ce: 0.0075 aux.acc_seg: 99.0228 +04/18 16:41:03 - mmengine - INFO - Iter(train) [ 93100/160000] lr: 4.6166e-03 eta: 10:15:46 time: 0.5525 data_time: 0.0069 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0073 decode.acc_seg: 99.6995 aux.loss_ce: 0.0076 aux.acc_seg: 99.1519 +04/18 16:41:30 - mmengine - INFO - Iter(train) [ 93150/160000] lr: 4.6136e-03 eta: 10:15:18 time: 0.5539 data_time: 0.0065 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0078 decode.acc_seg: 99.7832 aux.loss_ce: 0.0076 aux.acc_seg: 99.2928 +04/18 16:41:58 - mmengine - INFO - Iter(train) [ 93200/160000] lr: 4.6105e-03 eta: 10:14:51 time: 0.5528 data_time: 0.0067 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0064 decode.acc_seg: 99.6873 aux.loss_ce: 0.0068 aux.acc_seg: 99.1235 +04/18 16:42:26 - mmengine - INFO - Iter(train) [ 93250/160000] lr: 4.6075e-03 eta: 10:14:23 time: 0.5548 data_time: 0.0066 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0068 decode.acc_seg: 99.6369 aux.loss_ce: 0.0071 aux.acc_seg: 99.0698 +04/18 16:42:54 - mmengine - INFO - Iter(train) [ 93300/160000] lr: 4.6045e-03 eta: 10:13:56 time: 0.5528 data_time: 0.0069 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0070 decode.acc_seg: 99.7134 aux.loss_ce: 0.0072 aux.acc_seg: 99.1560 +04/18 16:43:21 - mmengine - INFO - Iter(train) [ 93350/160000] lr: 4.6014e-03 eta: 10:13:28 time: 0.5539 data_time: 0.0072 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0073 decode.acc_seg: 99.6781 aux.loss_ce: 0.0074 aux.acc_seg: 99.1940 +04/18 16:43:49 - mmengine - INFO - Iter(train) [ 93400/160000] lr: 4.5984e-03 eta: 10:13:00 time: 0.5544 data_time: 0.0081 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0069 decode.acc_seg: 99.6734 aux.loss_ce: 0.0074 aux.acc_seg: 99.1181 +04/18 16:44:17 - mmengine - INFO - Iter(train) [ 93450/160000] lr: 4.5953e-03 eta: 10:12:33 time: 0.5534 data_time: 0.0066 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0077 decode.acc_seg: 99.7462 aux.loss_ce: 0.0075 aux.acc_seg: 99.3246 +04/18 16:44:45 - mmengine - INFO - Iter(train) [ 93500/160000] lr: 4.5923e-03 eta: 10:12:05 time: 0.5547 data_time: 0.0064 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0068 decode.acc_seg: 99.7552 aux.loss_ce: 0.0070 aux.acc_seg: 99.3195 +04/18 16:45:12 - mmengine - INFO - Iter(train) [ 93550/160000] lr: 4.5893e-03 eta: 10:11:38 time: 0.5537 data_time: 0.0072 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0070 decode.acc_seg: 99.7588 aux.loss_ce: 0.0072 aux.acc_seg: 99.3825 +04/18 16:45:40 - mmengine - INFO - Iter(train) [ 93600/160000] lr: 4.5862e-03 eta: 10:11:10 time: 0.5536 data_time: 0.0066 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0072 decode.acc_seg: 99.6302 aux.loss_ce: 0.0073 aux.acc_seg: 99.0357 +04/18 16:46:08 - mmengine - INFO - Iter(train) [ 93650/160000] lr: 4.5832e-03 eta: 10:10:43 time: 0.5551 data_time: 0.0063 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0075 decode.acc_seg: 99.7753 aux.loss_ce: 0.0079 aux.acc_seg: 99.3404 +04/18 16:46:35 - mmengine - INFO - Iter(train) [ 93700/160000] lr: 4.5801e-03 eta: 10:10:15 time: 0.5544 data_time: 0.0061 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0066 decode.acc_seg: 99.7736 aux.loss_ce: 0.0067 aux.acc_seg: 99.4164 +04/18 16:47:03 - mmengine - INFO - Iter(train) [ 93750/160000] lr: 4.5771e-03 eta: 10:09:48 time: 0.5539 data_time: 0.0081 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0073 decode.acc_seg: 99.7229 aux.loss_ce: 0.0078 aux.acc_seg: 99.1734 +04/18 16:47:31 - mmengine - INFO - Iter(train) [ 93800/160000] lr: 4.5741e-03 eta: 10:09:20 time: 0.5542 data_time: 0.0068 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0068 decode.acc_seg: 99.7807 aux.loss_ce: 0.0072 aux.acc_seg: 99.2726 +04/18 16:47:59 - mmengine - INFO - Iter(train) [ 93850/160000] lr: 4.5710e-03 eta: 10:08:53 time: 0.5547 data_time: 0.0069 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0067 decode.acc_seg: 99.7389 aux.loss_ce: 0.0071 aux.acc_seg: 99.1090 +04/18 16:48:27 - mmengine - INFO - Iter(train) [ 93900/160000] lr: 4.5680e-03 eta: 10:08:25 time: 0.5543 data_time: 0.0078 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0068 decode.acc_seg: 99.6845 aux.loss_ce: 0.0073 aux.acc_seg: 99.1134 +04/18 16:48:54 - mmengine - INFO - Iter(train) [ 93950/160000] lr: 4.5649e-03 eta: 10:07:58 time: 0.5530 data_time: 0.0066 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0069 decode.acc_seg: 99.6857 aux.loss_ce: 0.0073 aux.acc_seg: 99.0434 +04/18 16:49:22 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 16:49:22 - mmengine - INFO - Iter(train) [ 94000/160000] lr: 4.5619e-03 eta: 10:07:30 time: 0.5557 data_time: 0.0065 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0066 decode.acc_seg: 99.7552 aux.loss_ce: 0.0068 aux.acc_seg: 99.4005 +04/18 16:49:50 - mmengine - INFO - Iter(train) [ 94050/160000] lr: 4.5589e-03 eta: 10:07:03 time: 0.5542 data_time: 0.0072 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0070 decode.acc_seg: 99.6277 aux.loss_ce: 0.0080 aux.acc_seg: 98.4986 +04/18 16:50:18 - mmengine - INFO - Iter(train) [ 94100/160000] lr: 4.5558e-03 eta: 10:06:35 time: 0.5549 data_time: 0.0061 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0069 decode.acc_seg: 99.7393 aux.loss_ce: 0.0069 aux.acc_seg: 99.3925 +04/18 16:50:45 - mmengine - INFO - Iter(train) [ 94150/160000] lr: 4.5528e-03 eta: 10:06:08 time: 0.5538 data_time: 0.0062 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0073 decode.acc_seg: 99.6656 aux.loss_ce: 0.0074 aux.acc_seg: 99.0781 +04/18 16:51:13 - mmengine - INFO - Iter(train) [ 94200/160000] lr: 4.5497e-03 eta: 10:05:40 time: 0.5537 data_time: 0.0068 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0068 decode.acc_seg: 99.6878 aux.loss_ce: 0.0068 aux.acc_seg: 99.2552 +04/18 16:51:41 - mmengine - INFO - Iter(train) [ 94250/160000] lr: 4.5467e-03 eta: 10:05:12 time: 0.5525 data_time: 0.0062 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0069 decode.acc_seg: 99.6560 aux.loss_ce: 0.0072 aux.acc_seg: 99.1100 +04/18 16:52:08 - mmengine - INFO - Iter(train) [ 94300/160000] lr: 4.5436e-03 eta: 10:04:45 time: 0.5532 data_time: 0.0069 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0073 decode.acc_seg: 99.7941 aux.loss_ce: 0.0076 aux.acc_seg: 99.3337 +04/18 16:52:36 - mmengine - INFO - Iter(train) [ 94350/160000] lr: 4.5406e-03 eta: 10:04:17 time: 0.5547 data_time: 0.0066 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0068 decode.acc_seg: 99.6806 aux.loss_ce: 0.0072 aux.acc_seg: 99.1579 +04/18 16:53:04 - mmengine - INFO - Iter(train) [ 94400/160000] lr: 4.5375e-03 eta: 10:03:50 time: 0.5540 data_time: 0.0066 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0063 decode.acc_seg: 99.6963 aux.loss_ce: 0.0072 aux.acc_seg: 99.0783 +04/18 16:53:32 - mmengine - INFO - Iter(train) [ 94450/160000] lr: 4.5345e-03 eta: 10:03:22 time: 0.5536 data_time: 0.0067 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0068 decode.acc_seg: 99.7437 aux.loss_ce: 0.0072 aux.acc_seg: 99.2533 +04/18 16:53:59 - mmengine - INFO - Iter(train) [ 94500/160000] lr: 4.5315e-03 eta: 10:02:55 time: 0.5544 data_time: 0.0071 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0063 decode.acc_seg: 99.8111 aux.loss_ce: 0.0070 aux.acc_seg: 99.3376 +04/18 16:54:27 - mmengine - INFO - Iter(train) [ 94550/160000] lr: 4.5284e-03 eta: 10:02:27 time: 0.5529 data_time: 0.0068 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0066 decode.acc_seg: 99.7833 aux.loss_ce: 0.0067 aux.acc_seg: 99.3541 +04/18 16:54:55 - mmengine - INFO - Iter(train) [ 94600/160000] lr: 4.5254e-03 eta: 10:02:00 time: 0.5534 data_time: 0.0064 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0069 decode.acc_seg: 99.7384 aux.loss_ce: 0.0076 aux.acc_seg: 99.1609 +04/18 16:55:23 - mmengine - INFO - Iter(train) [ 94650/160000] lr: 4.5223e-03 eta: 10:01:32 time: 0.5548 data_time: 0.0064 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0069 decode.acc_seg: 99.7441 aux.loss_ce: 0.0076 aux.acc_seg: 99.2318 +04/18 16:55:50 - mmengine - INFO - Iter(train) [ 94700/160000] lr: 4.5193e-03 eta: 10:01:05 time: 0.5552 data_time: 0.0062 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0069 decode.acc_seg: 99.6926 aux.loss_ce: 0.0073 aux.acc_seg: 99.1946 +04/18 16:56:18 - mmengine - INFO - Iter(train) [ 94750/160000] lr: 4.5162e-03 eta: 10:00:37 time: 0.5530 data_time: 0.0066 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0066 decode.acc_seg: 99.6918 aux.loss_ce: 0.0069 aux.acc_seg: 99.3544 +04/18 16:56:46 - mmengine - INFO - Iter(train) [ 94800/160000] lr: 4.5132e-03 eta: 10:00:09 time: 0.5624 data_time: 0.0072 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0072 decode.acc_seg: 99.6819 aux.loss_ce: 0.0072 aux.acc_seg: 99.2666 +04/18 16:57:13 - mmengine - INFO - Iter(train) [ 94850/160000] lr: 4.5101e-03 eta: 9:59:42 time: 0.5553 data_time: 0.0059 memory: 7635 loss: 0.0124 decode.loss_ce: 0.0059 decode.acc_seg: 99.7990 aux.loss_ce: 0.0065 aux.acc_seg: 99.3868 +04/18 16:57:41 - mmengine - INFO - Iter(train) [ 94900/160000] lr: 4.5071e-03 eta: 9:59:14 time: 0.5533 data_time: 0.0066 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0070 decode.acc_seg: 99.7571 aux.loss_ce: 0.0074 aux.acc_seg: 99.2595 +04/18 16:58:09 - mmengine - INFO - Iter(train) [ 94950/160000] lr: 4.5040e-03 eta: 9:58:47 time: 0.5638 data_time: 0.0060 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0084 decode.acc_seg: 99.6574 aux.loss_ce: 0.0082 aux.acc_seg: 99.1121 +04/18 16:58:37 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 16:58:37 - mmengine - INFO - Iter(train) [ 95000/160000] lr: 4.5010e-03 eta: 9:58:19 time: 0.5526 data_time: 0.0065 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0072 decode.acc_seg: 99.6932 aux.loss_ce: 0.0072 aux.acc_seg: 99.2917 +04/18 16:59:05 - mmengine - INFO - Iter(train) [ 95050/160000] lr: 4.4980e-03 eta: 9:57:52 time: 0.5536 data_time: 0.0064 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0070 decode.acc_seg: 99.7259 aux.loss_ce: 0.0072 aux.acc_seg: 99.1473 +04/18 16:59:32 - mmengine - INFO - Iter(train) [ 95100/160000] lr: 4.4949e-03 eta: 9:57:24 time: 0.5531 data_time: 0.0067 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0066 decode.acc_seg: 99.7579 aux.loss_ce: 0.0068 aux.acc_seg: 99.3444 +04/18 17:00:00 - mmengine - INFO - Iter(train) [ 95150/160000] lr: 4.4919e-03 eta: 9:56:57 time: 0.5531 data_time: 0.0065 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0079 decode.acc_seg: 99.6657 aux.loss_ce: 0.0081 aux.acc_seg: 99.2908 +04/18 17:00:28 - mmengine - INFO - Iter(train) [ 95200/160000] lr: 4.4888e-03 eta: 9:56:29 time: 0.5539 data_time: 0.0072 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0077 decode.acc_seg: 99.5678 aux.loss_ce: 0.0079 aux.acc_seg: 99.0845 +04/18 17:00:55 - mmengine - INFO - Iter(train) [ 95250/160000] lr: 4.4858e-03 eta: 9:56:02 time: 0.5536 data_time: 0.0063 memory: 7635 loss: 0.0183 decode.loss_ce: 0.0103 decode.acc_seg: 99.4879 aux.loss_ce: 0.0080 aux.acc_seg: 99.0593 +04/18 17:01:23 - mmengine - INFO - Iter(train) [ 95300/160000] lr: 4.4827e-03 eta: 9:55:34 time: 0.5533 data_time: 0.0064 memory: 7635 loss: 0.0198 decode.loss_ce: 0.0108 decode.acc_seg: 99.6470 aux.loss_ce: 0.0091 aux.acc_seg: 99.1091 +04/18 17:01:51 - mmengine - INFO - Iter(train) [ 95350/160000] lr: 4.4797e-03 eta: 9:55:07 time: 0.5540 data_time: 0.0064 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0100 decode.acc_seg: 99.6401 aux.loss_ce: 0.0079 aux.acc_seg: 99.2100 +04/18 17:02:19 - mmengine - INFO - Iter(train) [ 95400/160000] lr: 4.4766e-03 eta: 9:54:39 time: 0.5550 data_time: 0.0072 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0091 decode.acc_seg: 99.6162 aux.loss_ce: 0.0080 aux.acc_seg: 99.3341 +04/18 17:02:46 - mmengine - INFO - Iter(train) [ 95450/160000] lr: 4.4736e-03 eta: 9:54:11 time: 0.5549 data_time: 0.0068 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0079 decode.acc_seg: 99.6016 aux.loss_ce: 0.0079 aux.acc_seg: 99.1653 +04/18 17:03:14 - mmengine - INFO - Iter(train) [ 95500/160000] lr: 4.4705e-03 eta: 9:53:44 time: 0.5543 data_time: 0.0068 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0075 decode.acc_seg: 99.7846 aux.loss_ce: 0.0072 aux.acc_seg: 99.3245 +04/18 17:03:42 - mmengine - INFO - Iter(train) [ 95550/160000] lr: 4.4675e-03 eta: 9:53:16 time: 0.5537 data_time: 0.0065 memory: 7635 loss: 0.0195 decode.loss_ce: 0.0105 decode.acc_seg: 99.6505 aux.loss_ce: 0.0090 aux.acc_seg: 99.2621 +04/18 17:04:10 - mmengine - INFO - Iter(train) [ 95600/160000] lr: 4.4644e-03 eta: 9:52:49 time: 0.5546 data_time: 0.0077 memory: 7635 loss: 0.0193 decode.loss_ce: 0.0100 decode.acc_seg: 99.6714 aux.loss_ce: 0.0092 aux.acc_seg: 99.0884 +04/18 17:04:37 - mmengine - INFO - Iter(train) [ 95650/160000] lr: 4.4614e-03 eta: 9:52:21 time: 0.5539 data_time: 0.0066 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0077 decode.acc_seg: 99.6899 aux.loss_ce: 0.0075 aux.acc_seg: 99.3049 +04/18 17:05:05 - mmengine - INFO - Iter(train) [ 95700/160000] lr: 4.4583e-03 eta: 9:51:54 time: 0.5525 data_time: 0.0062 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0072 decode.acc_seg: 99.6909 aux.loss_ce: 0.0077 aux.acc_seg: 99.1796 +04/18 17:05:33 - mmengine - INFO - Iter(train) [ 95750/160000] lr: 4.4553e-03 eta: 9:51:26 time: 0.5537 data_time: 0.0065 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0078 decode.acc_seg: 99.7422 aux.loss_ce: 0.0081 aux.acc_seg: 99.1709 +04/18 17:06:00 - mmengine - INFO - Iter(train) [ 95800/160000] lr: 4.4522e-03 eta: 9:50:59 time: 0.5545 data_time: 0.0068 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0086 decode.acc_seg: 99.4585 aux.loss_ce: 0.0083 aux.acc_seg: 98.8168 +04/18 17:06:28 - mmengine - INFO - Iter(train) [ 95850/160000] lr: 4.4492e-03 eta: 9:50:31 time: 0.5546 data_time: 0.0073 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0085 decode.acc_seg: 99.6819 aux.loss_ce: 0.0079 aux.acc_seg: 99.1382 +04/18 17:06:56 - mmengine - INFO - Iter(train) [ 95900/160000] lr: 4.4461e-03 eta: 9:50:04 time: 0.5539 data_time: 0.0066 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0076 decode.acc_seg: 99.7220 aux.loss_ce: 0.0079 aux.acc_seg: 99.2452 +04/18 17:07:24 - mmengine - INFO - Iter(train) [ 95950/160000] lr: 4.4431e-03 eta: 9:49:36 time: 0.5551 data_time: 0.0068 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0073 decode.acc_seg: 99.7738 aux.loss_ce: 0.0073 aux.acc_seg: 99.3678 +04/18 17:07:52 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 17:07:52 - mmengine - INFO - Iter(train) [ 96000/160000] lr: 4.4400e-03 eta: 9:49:09 time: 0.5632 data_time: 0.0064 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0073 decode.acc_seg: 99.7273 aux.loss_ce: 0.0073 aux.acc_seg: 99.2279 +04/18 17:08:19 - mmengine - INFO - Iter(train) [ 96050/160000] lr: 4.4370e-03 eta: 9:48:41 time: 0.5545 data_time: 0.0074 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0073 decode.acc_seg: 99.6448 aux.loss_ce: 0.0073 aux.acc_seg: 99.2266 +04/18 17:08:47 - mmengine - INFO - Iter(train) [ 96100/160000] lr: 4.4339e-03 eta: 9:48:13 time: 0.5546 data_time: 0.0064 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0082 decode.acc_seg: 99.6461 aux.loss_ce: 0.0075 aux.acc_seg: 99.1035 +04/18 17:09:15 - mmengine - INFO - Iter(train) [ 96150/160000] lr: 4.4309e-03 eta: 9:47:46 time: 0.5545 data_time: 0.0076 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0074 decode.acc_seg: 99.7221 aux.loss_ce: 0.0074 aux.acc_seg: 99.1570 +04/18 17:09:42 - mmengine - INFO - Iter(train) [ 96200/160000] lr: 4.4278e-03 eta: 9:47:18 time: 0.5534 data_time: 0.0063 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0075 decode.acc_seg: 99.7612 aux.loss_ce: 0.0074 aux.acc_seg: 99.3772 +04/18 17:10:10 - mmengine - INFO - Iter(train) [ 96250/160000] lr: 4.4248e-03 eta: 9:46:51 time: 0.5543 data_time: 0.0060 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0078 decode.acc_seg: 99.6583 aux.loss_ce: 0.0081 aux.acc_seg: 99.0301 +04/18 17:10:38 - mmengine - INFO - Iter(train) [ 96300/160000] lr: 4.4217e-03 eta: 9:46:23 time: 0.5542 data_time: 0.0070 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0074 decode.acc_seg: 99.7119 aux.loss_ce: 0.0080 aux.acc_seg: 99.1693 +04/18 17:11:06 - mmengine - INFO - Iter(train) [ 96350/160000] lr: 4.4187e-03 eta: 9:45:56 time: 0.5549 data_time: 0.0064 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0070 decode.acc_seg: 99.6655 aux.loss_ce: 0.0073 aux.acc_seg: 99.1374 +04/18 17:11:33 - mmengine - INFO - Iter(train) [ 96400/160000] lr: 4.4156e-03 eta: 9:45:28 time: 0.5550 data_time: 0.0072 memory: 7635 loss: 0.0127 decode.loss_ce: 0.0061 decode.acc_seg: 99.7654 aux.loss_ce: 0.0065 aux.acc_seg: 99.3459 +04/18 17:12:01 - mmengine - INFO - Iter(train) [ 96450/160000] lr: 4.4125e-03 eta: 9:45:01 time: 0.5531 data_time: 0.0064 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0076 decode.acc_seg: 99.6296 aux.loss_ce: 0.0077 aux.acc_seg: 98.8319 +04/18 17:12:29 - mmengine - INFO - Iter(train) [ 96500/160000] lr: 4.4095e-03 eta: 9:44:33 time: 0.5550 data_time: 0.0081 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0067 decode.acc_seg: 99.7513 aux.loss_ce: 0.0069 aux.acc_seg: 99.2723 +04/18 17:12:57 - mmengine - INFO - Iter(train) [ 96550/160000] lr: 4.4064e-03 eta: 9:44:06 time: 0.5556 data_time: 0.0075 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0073 decode.acc_seg: 99.7525 aux.loss_ce: 0.0077 aux.acc_seg: 99.3343 +04/18 17:13:24 - mmengine - INFO - Iter(train) [ 96600/160000] lr: 4.4034e-03 eta: 9:43:38 time: 0.5554 data_time: 0.0065 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0069 decode.acc_seg: 99.7320 aux.loss_ce: 0.0073 aux.acc_seg: 99.3683 +04/18 17:13:52 - mmengine - INFO - Iter(train) [ 96650/160000] lr: 4.4003e-03 eta: 9:43:10 time: 0.5547 data_time: 0.0063 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0072 decode.acc_seg: 99.7233 aux.loss_ce: 0.0073 aux.acc_seg: 99.2402 +04/18 17:14:20 - mmengine - INFO - Iter(train) [ 96700/160000] lr: 4.3973e-03 eta: 9:42:43 time: 0.5549 data_time: 0.0066 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0066 decode.acc_seg: 99.7185 aux.loss_ce: 0.0069 aux.acc_seg: 99.1770 +04/18 17:14:48 - mmengine - INFO - Iter(train) [ 96750/160000] lr: 4.3942e-03 eta: 9:42:15 time: 0.5568 data_time: 0.0073 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0077 decode.acc_seg: 99.7624 aux.loss_ce: 0.0077 aux.acc_seg: 99.2085 +04/18 17:15:15 - mmengine - INFO - Iter(train) [ 96800/160000] lr: 4.3912e-03 eta: 9:41:48 time: 0.5545 data_time: 0.0069 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0071 decode.acc_seg: 99.5921 aux.loss_ce: 0.0077 aux.acc_seg: 98.8240 +04/18 17:15:43 - mmengine - INFO - Iter(train) [ 96850/160000] lr: 4.3881e-03 eta: 9:41:20 time: 0.5558 data_time: 0.0083 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0068 decode.acc_seg: 99.7164 aux.loss_ce: 0.0069 aux.acc_seg: 99.2104 +04/18 17:16:11 - mmengine - INFO - Iter(train) [ 96900/160000] lr: 4.3851e-03 eta: 9:40:53 time: 0.5544 data_time: 0.0065 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0086 decode.acc_seg: 99.6477 aux.loss_ce: 0.0087 aux.acc_seg: 99.0242 +04/18 17:16:39 - mmengine - INFO - Iter(train) [ 96950/160000] lr: 4.3820e-03 eta: 9:40:25 time: 0.5621 data_time: 0.0067 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0078 decode.acc_seg: 99.6933 aux.loss_ce: 0.0077 aux.acc_seg: 99.1102 +04/18 17:17:06 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 17:17:06 - mmengine - INFO - Iter(train) [ 97000/160000] lr: 4.3789e-03 eta: 9:39:58 time: 0.5567 data_time: 0.0077 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0080 decode.acc_seg: 99.5578 aux.loss_ce: 0.0079 aux.acc_seg: 99.0413 +04/18 17:17:34 - mmengine - INFO - Iter(train) [ 97050/160000] lr: 4.3759e-03 eta: 9:39:30 time: 0.5549 data_time: 0.0069 memory: 7635 loss: 0.0179 decode.loss_ce: 0.0093 decode.acc_seg: 99.7400 aux.loss_ce: 0.0086 aux.acc_seg: 99.1406 +04/18 17:18:02 - mmengine - INFO - Iter(train) [ 97100/160000] lr: 4.3728e-03 eta: 9:39:03 time: 0.5626 data_time: 0.0069 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0078 decode.acc_seg: 99.6466 aux.loss_ce: 0.0074 aux.acc_seg: 99.1657 +04/18 17:18:30 - mmengine - INFO - Iter(train) [ 97150/160000] lr: 4.3698e-03 eta: 9:38:35 time: 0.5541 data_time: 0.0065 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0067 decode.acc_seg: 99.7083 aux.loss_ce: 0.0070 aux.acc_seg: 99.2443 +04/18 17:18:58 - mmengine - INFO - Iter(train) [ 97200/160000] lr: 4.3667e-03 eta: 9:38:08 time: 0.5554 data_time: 0.0072 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0072 decode.acc_seg: 99.7644 aux.loss_ce: 0.0073 aux.acc_seg: 99.2749 +04/18 17:19:25 - mmengine - INFO - Iter(train) [ 97250/160000] lr: 4.3637e-03 eta: 9:37:40 time: 0.5551 data_time: 0.0067 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0063 decode.acc_seg: 99.7007 aux.loss_ce: 0.0067 aux.acc_seg: 99.1323 +04/18 17:19:53 - mmengine - INFO - Iter(train) [ 97300/160000] lr: 4.3606e-03 eta: 9:37:13 time: 0.5536 data_time: 0.0065 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0069 decode.acc_seg: 99.7317 aux.loss_ce: 0.0074 aux.acc_seg: 99.1363 +04/18 17:20:21 - mmengine - INFO - Iter(train) [ 97350/160000] lr: 4.3575e-03 eta: 9:36:45 time: 0.5533 data_time: 0.0066 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0066 decode.acc_seg: 99.6843 aux.loss_ce: 0.0068 aux.acc_seg: 99.1193 +04/18 17:20:48 - mmengine - INFO - Iter(train) [ 97400/160000] lr: 4.3545e-03 eta: 9:36:17 time: 0.5542 data_time: 0.0070 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0073 decode.acc_seg: 99.6987 aux.loss_ce: 0.0078 aux.acc_seg: 99.2604 +04/18 17:21:16 - mmengine - INFO - Iter(train) [ 97450/160000] lr: 4.3514e-03 eta: 9:35:50 time: 0.5532 data_time: 0.0061 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0073 decode.acc_seg: 99.6772 aux.loss_ce: 0.0074 aux.acc_seg: 99.1599 +04/18 17:21:44 - mmengine - INFO - Iter(train) [ 97500/160000] lr: 4.3484e-03 eta: 9:35:22 time: 0.5556 data_time: 0.0066 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0067 decode.acc_seg: 99.7450 aux.loss_ce: 0.0070 aux.acc_seg: 99.1950 +04/18 17:22:12 - mmengine - INFO - Iter(train) [ 97550/160000] lr: 4.3453e-03 eta: 9:34:55 time: 0.5539 data_time: 0.0061 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0069 decode.acc_seg: 99.6667 aux.loss_ce: 0.0069 aux.acc_seg: 99.1509 +04/18 17:22:39 - mmengine - INFO - Iter(train) [ 97600/160000] lr: 4.3422e-03 eta: 9:34:27 time: 0.5554 data_time: 0.0073 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0072 decode.acc_seg: 99.6469 aux.loss_ce: 0.0076 aux.acc_seg: 99.0324 +04/18 17:23:07 - mmengine - INFO - Iter(train) [ 97650/160000] lr: 4.3392e-03 eta: 9:34:00 time: 0.5545 data_time: 0.0068 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0071 decode.acc_seg: 99.7666 aux.loss_ce: 0.0075 aux.acc_seg: 99.3659 +04/18 17:23:35 - mmengine - INFO - Iter(train) [ 97700/160000] lr: 4.3361e-03 eta: 9:33:32 time: 0.5561 data_time: 0.0063 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0075 decode.acc_seg: 99.6807 aux.loss_ce: 0.0075 aux.acc_seg: 99.2209 +04/18 17:24:03 - mmengine - INFO - Iter(train) [ 97750/160000] lr: 4.3331e-03 eta: 9:33:05 time: 0.5536 data_time: 0.0064 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0074 decode.acc_seg: 99.7061 aux.loss_ce: 0.0076 aux.acc_seg: 99.2447 +04/18 17:24:30 - mmengine - INFO - Iter(train) [ 97800/160000] lr: 4.3300e-03 eta: 9:32:37 time: 0.5560 data_time: 0.0072 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0070 decode.acc_seg: 99.6998 aux.loss_ce: 0.0071 aux.acc_seg: 99.1749 +04/18 17:24:58 - mmengine - INFO - Iter(train) [ 97850/160000] lr: 4.3269e-03 eta: 9:32:10 time: 0.5544 data_time: 0.0065 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0073 decode.acc_seg: 99.6622 aux.loss_ce: 0.0074 aux.acc_seg: 99.0404 +04/18 17:25:26 - mmengine - INFO - Iter(train) [ 97900/160000] lr: 4.3239e-03 eta: 9:31:42 time: 0.5540 data_time: 0.0070 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0070 decode.acc_seg: 99.7380 aux.loss_ce: 0.0072 aux.acc_seg: 99.3701 +04/18 17:25:54 - mmengine - INFO - Iter(train) [ 97950/160000] lr: 4.3208e-03 eta: 9:31:14 time: 0.5563 data_time: 0.0064 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0070 decode.acc_seg: 99.7115 aux.loss_ce: 0.0071 aux.acc_seg: 99.0584 +04/18 17:26:21 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 17:26:21 - mmengine - INFO - Iter(train) [ 98000/160000] lr: 4.3178e-03 eta: 9:30:47 time: 0.5547 data_time: 0.0069 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0071 decode.acc_seg: 99.7242 aux.loss_ce: 0.0078 aux.acc_seg: 99.2871 +04/18 17:26:49 - mmengine - INFO - Iter(train) [ 98050/160000] lr: 4.3147e-03 eta: 9:30:19 time: 0.5557 data_time: 0.0071 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0067 decode.acc_seg: 99.7675 aux.loss_ce: 0.0071 aux.acc_seg: 99.3314 +04/18 17:27:17 - mmengine - INFO - Iter(train) [ 98100/160000] lr: 4.3116e-03 eta: 9:29:52 time: 0.5544 data_time: 0.0080 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0068 decode.acc_seg: 99.7635 aux.loss_ce: 0.0071 aux.acc_seg: 99.3004 +04/18 17:27:45 - mmengine - INFO - Iter(train) [ 98150/160000] lr: 4.3086e-03 eta: 9:29:24 time: 0.5616 data_time: 0.0064 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0069 decode.acc_seg: 99.7717 aux.loss_ce: 0.0071 aux.acc_seg: 99.2148 +04/18 17:28:13 - mmengine - INFO - Iter(train) [ 98200/160000] lr: 4.3055e-03 eta: 9:28:57 time: 0.5636 data_time: 0.0074 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0065 decode.acc_seg: 99.7410 aux.loss_ce: 0.0070 aux.acc_seg: 99.2487 +04/18 17:28:40 - mmengine - INFO - Iter(train) [ 98250/160000] lr: 4.3025e-03 eta: 9:28:29 time: 0.5549 data_time: 0.0067 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0069 decode.acc_seg: 99.6316 aux.loss_ce: 0.0073 aux.acc_seg: 99.0919 +04/18 17:29:08 - mmengine - INFO - Iter(train) [ 98300/160000] lr: 4.2994e-03 eta: 9:28:02 time: 0.5546 data_time: 0.0067 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0075 decode.acc_seg: 99.6526 aux.loss_ce: 0.0075 aux.acc_seg: 99.1125 +04/18 17:29:36 - mmengine - INFO - Iter(train) [ 98350/160000] lr: 4.2963e-03 eta: 9:27:34 time: 0.5553 data_time: 0.0067 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0071 decode.acc_seg: 99.6505 aux.loss_ce: 0.0071 aux.acc_seg: 99.1550 +04/18 17:30:04 - mmengine - INFO - Iter(train) [ 98400/160000] lr: 4.2933e-03 eta: 9:27:07 time: 0.5535 data_time: 0.0066 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0075 decode.acc_seg: 99.7371 aux.loss_ce: 0.0078 aux.acc_seg: 99.3375 +04/18 17:30:31 - mmengine - INFO - Iter(train) [ 98450/160000] lr: 4.2902e-03 eta: 9:26:39 time: 0.5554 data_time: 0.0062 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0071 decode.acc_seg: 99.7246 aux.loss_ce: 0.0072 aux.acc_seg: 99.1135 +04/18 17:30:59 - mmengine - INFO - Iter(train) [ 98500/160000] lr: 4.2871e-03 eta: 9:26:12 time: 0.5536 data_time: 0.0066 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0071 decode.acc_seg: 99.6385 aux.loss_ce: 0.0071 aux.acc_seg: 99.0829 +04/18 17:31:27 - mmengine - INFO - Iter(train) [ 98550/160000] lr: 4.2841e-03 eta: 9:25:44 time: 0.5546 data_time: 0.0064 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0071 decode.acc_seg: 99.6452 aux.loss_ce: 0.0072 aux.acc_seg: 99.2522 +04/18 17:31:55 - mmengine - INFO - Iter(train) [ 98600/160000] lr: 4.2810e-03 eta: 9:25:17 time: 0.5544 data_time: 0.0070 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0076 decode.acc_seg: 99.6967 aux.loss_ce: 0.0077 aux.acc_seg: 99.3668 +04/18 17:32:22 - mmengine - INFO - Iter(train) [ 98650/160000] lr: 4.2779e-03 eta: 9:24:49 time: 0.5551 data_time: 0.0068 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0079 decode.acc_seg: 99.5887 aux.loss_ce: 0.0076 aux.acc_seg: 99.0882 +04/18 17:32:50 - mmengine - INFO - Iter(train) [ 98700/160000] lr: 4.2749e-03 eta: 9:24:21 time: 0.5542 data_time: 0.0068 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0076 decode.acc_seg: 99.7547 aux.loss_ce: 0.0074 aux.acc_seg: 99.2813 +04/18 17:33:18 - mmengine - INFO - Iter(train) [ 98750/160000] lr: 4.2718e-03 eta: 9:23:54 time: 0.5552 data_time: 0.0064 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0067 decode.acc_seg: 99.7483 aux.loss_ce: 0.0069 aux.acc_seg: 99.3092 +04/18 17:33:46 - mmengine - INFO - Iter(train) [ 98800/160000] lr: 4.2688e-03 eta: 9:23:26 time: 0.5534 data_time: 0.0067 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0084 decode.acc_seg: 99.7491 aux.loss_ce: 0.0078 aux.acc_seg: 99.3162 +04/18 17:34:13 - mmengine - INFO - Iter(train) [ 98850/160000] lr: 4.2657e-03 eta: 9:22:59 time: 0.5554 data_time: 0.0060 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0082 decode.acc_seg: 99.6961 aux.loss_ce: 0.0080 aux.acc_seg: 99.2959 +04/18 17:34:41 - mmengine - INFO - Iter(train) [ 98900/160000] lr: 4.2626e-03 eta: 9:22:31 time: 0.5543 data_time: 0.0065 memory: 7635 loss: 0.0215 decode.loss_ce: 0.0117 decode.acc_seg: 99.6341 aux.loss_ce: 0.0098 aux.acc_seg: 99.1272 +04/18 17:35:09 - mmengine - INFO - Iter(train) [ 98950/160000] lr: 4.2596e-03 eta: 9:22:04 time: 0.5551 data_time: 0.0070 memory: 7635 loss: 0.0191 decode.loss_ce: 0.0104 decode.acc_seg: 99.6702 aux.loss_ce: 0.0086 aux.acc_seg: 99.2450 +04/18 17:35:36 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 17:35:36 - mmengine - INFO - Iter(train) [ 99000/160000] lr: 4.2565e-03 eta: 9:21:36 time: 0.5561 data_time: 0.0065 memory: 7635 loss: 0.0188 decode.loss_ce: 0.0100 decode.acc_seg: 99.7582 aux.loss_ce: 0.0089 aux.acc_seg: 99.3401 +04/18 17:36:04 - mmengine - INFO - Iter(train) [ 99050/160000] lr: 4.2534e-03 eta: 9:21:09 time: 0.5551 data_time: 0.0063 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0086 decode.acc_seg: 99.6223 aux.loss_ce: 0.0079 aux.acc_seg: 98.9206 +04/18 17:36:32 - mmengine - INFO - Iter(train) [ 99100/160000] lr: 4.2504e-03 eta: 9:20:41 time: 0.5555 data_time: 0.0065 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0079 decode.acc_seg: 99.5854 aux.loss_ce: 0.0076 aux.acc_seg: 99.0214 +04/18 17:37:00 - mmengine - INFO - Iter(train) [ 99150/160000] lr: 4.2473e-03 eta: 9:20:13 time: 0.5547 data_time: 0.0070 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0077 decode.acc_seg: 99.6884 aux.loss_ce: 0.0075 aux.acc_seg: 99.1795 +04/18 17:37:28 - mmengine - INFO - Iter(train) [ 99200/160000] lr: 4.2442e-03 eta: 9:19:46 time: 0.5554 data_time: 0.0060 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0064 decode.acc_seg: 99.6729 aux.loss_ce: 0.0070 aux.acc_seg: 98.8970 +04/18 17:37:55 - mmengine - INFO - Iter(train) [ 99250/160000] lr: 4.2412e-03 eta: 9:19:18 time: 0.5553 data_time: 0.0069 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0070 decode.acc_seg: 99.7354 aux.loss_ce: 0.0072 aux.acc_seg: 99.3119 +04/18 17:38:23 - mmengine - INFO - Iter(train) [ 99300/160000] lr: 4.2381e-03 eta: 9:18:51 time: 0.5542 data_time: 0.0064 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0080 decode.acc_seg: 99.6769 aux.loss_ce: 0.0081 aux.acc_seg: 99.1219 +04/18 17:38:51 - mmengine - INFO - Iter(train) [ 99350/160000] lr: 4.2350e-03 eta: 9:18:23 time: 0.5549 data_time: 0.0066 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0075 decode.acc_seg: 99.7375 aux.loss_ce: 0.0077 aux.acc_seg: 99.3154 +04/18 17:39:19 - mmengine - INFO - Iter(train) [ 99400/160000] lr: 4.2319e-03 eta: 9:17:56 time: 0.5561 data_time: 0.0076 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0090 decode.acc_seg: 99.5961 aux.loss_ce: 0.0082 aux.acc_seg: 99.1252 +04/18 17:39:46 - mmengine - INFO - Iter(train) [ 99450/160000] lr: 4.2289e-03 eta: 9:17:28 time: 0.5535 data_time: 0.0064 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0077 decode.acc_seg: 99.6661 aux.loss_ce: 0.0077 aux.acc_seg: 99.2162 +04/18 17:40:14 - mmengine - INFO - Iter(train) [ 99500/160000] lr: 4.2258e-03 eta: 9:17:01 time: 0.5546 data_time: 0.0068 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0077 decode.acc_seg: 99.5756 aux.loss_ce: 0.0077 aux.acc_seg: 99.0845 +04/18 17:40:42 - mmengine - INFO - Iter(train) [ 99550/160000] lr: 4.2227e-03 eta: 9:16:33 time: 0.5543 data_time: 0.0067 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0073 decode.acc_seg: 99.7802 aux.loss_ce: 0.0073 aux.acc_seg: 99.4399 +04/18 17:41:10 - mmengine - INFO - Iter(train) [ 99600/160000] lr: 4.2197e-03 eta: 9:16:06 time: 0.5573 data_time: 0.0072 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0075 decode.acc_seg: 99.6706 aux.loss_ce: 0.0079 aux.acc_seg: 99.1669 +04/18 17:41:37 - mmengine - INFO - Iter(train) [ 99650/160000] lr: 4.2166e-03 eta: 9:15:38 time: 0.5545 data_time: 0.0068 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0070 decode.acc_seg: 99.7578 aux.loss_ce: 0.0074 aux.acc_seg: 99.3618 +04/18 17:42:05 - mmengine - INFO - Iter(train) [ 99700/160000] lr: 4.2135e-03 eta: 9:15:11 time: 0.5541 data_time: 0.0067 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0068 decode.acc_seg: 99.7287 aux.loss_ce: 0.0073 aux.acc_seg: 99.1139 +04/18 17:42:33 - mmengine - INFO - Iter(train) [ 99750/160000] lr: 4.2105e-03 eta: 9:14:43 time: 0.5559 data_time: 0.0063 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0069 decode.acc_seg: 99.6978 aux.loss_ce: 0.0074 aux.acc_seg: 99.2369 +04/18 17:43:01 - mmengine - INFO - Iter(train) [ 99800/160000] lr: 4.2074e-03 eta: 9:14:15 time: 0.5551 data_time: 0.0066 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0076 decode.acc_seg: 99.7423 aux.loss_ce: 0.0078 aux.acc_seg: 99.2477 +04/18 17:43:28 - mmengine - INFO - Iter(train) [ 99850/160000] lr: 4.2043e-03 eta: 9:13:48 time: 0.5541 data_time: 0.0062 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0066 decode.acc_seg: 99.7376 aux.loss_ce: 0.0070 aux.acc_seg: 99.2662 +04/18 17:43:56 - mmengine - INFO - Iter(train) [ 99900/160000] lr: 4.2013e-03 eta: 9:13:20 time: 0.5547 data_time: 0.0075 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0076 decode.acc_seg: 99.6868 aux.loss_ce: 0.0078 aux.acc_seg: 99.4017 +04/18 17:44:24 - mmengine - INFO - Iter(train) [ 99950/160000] lr: 4.1982e-03 eta: 9:12:53 time: 0.5559 data_time: 0.0064 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0074 decode.acc_seg: 99.7042 aux.loss_ce: 0.0074 aux.acc_seg: 99.1630 +04/18 17:44:52 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 17:44:52 - mmengine - INFO - Iter(train) [100000/160000] lr: 4.1951e-03 eta: 9:12:25 time: 0.5547 data_time: 0.0063 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0069 decode.acc_seg: 99.7372 aux.loss_ce: 0.0077 aux.acc_seg: 99.2908 +04/18 17:44:52 - mmengine - INFO - Saving checkpoint at 100000 iterations +04/18 17:44:56 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:07 time: 0.0462 data_time: 0.0015 memory: 1657 +04/18 17:44:58 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:04 time: 0.0465 data_time: 0.0015 memory: 1657 +04/18 17:45:00 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:02 time: 0.0472 data_time: 0.0015 memory: 1657 +04/18 17:45:03 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.0458 data_time: 0.0014 memory: 1657 +04/18 17:45:03 - mmengine - INFO - per class results: +04/18 17:45:03 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 99.07 | 99.44 | 99.53 | 99.62 | 99.44 | +| contrast | 80.19 | 90.93 | 89.01 | 87.16 | 90.93 | ++------------+-------+-------+--------+-----------+--------+ +04/18 17:45:03 - mmengine - INFO - Iter(val) [200/200] aAcc: 99.1000 mIoU: 89.6300 mAcc: 95.1900 mFscore: 94.2700 mPrecision: 93.3900 mRecall: 95.1900 data_time: 0.0015 time: 0.0466 +04/18 17:45:31 - mmengine - INFO - Iter(train) [100050/160000] lr: 4.1920e-03 eta: 9:11:58 time: 0.5538 data_time: 0.0067 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0080 decode.acc_seg: 99.7095 aux.loss_ce: 0.0080 aux.acc_seg: 99.1308 +04/18 17:45:58 - mmengine - INFO - Iter(train) [100100/160000] lr: 4.1890e-03 eta: 9:11:30 time: 0.5565 data_time: 0.0071 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0071 decode.acc_seg: 99.6163 aux.loss_ce: 0.0075 aux.acc_seg: 99.1165 +04/18 17:46:26 - mmengine - INFO - Iter(train) [100150/160000] lr: 4.1859e-03 eta: 9:11:03 time: 0.5543 data_time: 0.0068 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0069 decode.acc_seg: 99.6577 aux.loss_ce: 0.0077 aux.acc_seg: 99.1933 +04/18 17:46:54 - mmengine - INFO - Iter(train) [100200/160000] lr: 4.1828e-03 eta: 9:10:35 time: 0.5547 data_time: 0.0067 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0065 decode.acc_seg: 99.7790 aux.loss_ce: 0.0073 aux.acc_seg: 99.3054 +04/18 17:47:22 - mmengine - INFO - Iter(train) [100250/160000] lr: 4.1798e-03 eta: 9:10:08 time: 0.5552 data_time: 0.0063 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0071 decode.acc_seg: 99.7389 aux.loss_ce: 0.0079 aux.acc_seg: 99.2993 +04/18 17:47:49 - mmengine - INFO - Iter(train) [100300/160000] lr: 4.1767e-03 eta: 9:09:40 time: 0.5538 data_time: 0.0065 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0070 decode.acc_seg: 99.7589 aux.loss_ce: 0.0070 aux.acc_seg: 99.3131 +04/18 17:48:17 - mmengine - INFO - Iter(train) [100350/160000] lr: 4.1736e-03 eta: 9:09:13 time: 0.5539 data_time: 0.0066 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0071 decode.acc_seg: 99.7858 aux.loss_ce: 0.0077 aux.acc_seg: 99.3250 +04/18 17:48:45 - mmengine - INFO - Iter(train) [100400/160000] lr: 4.1705e-03 eta: 9:08:45 time: 0.5548 data_time: 0.0065 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0072 decode.acc_seg: 99.7776 aux.loss_ce: 0.0074 aux.acc_seg: 99.4458 +04/18 17:49:13 - mmengine - INFO - Iter(train) [100450/160000] lr: 4.1675e-03 eta: 9:08:17 time: 0.5536 data_time: 0.0068 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0077 decode.acc_seg: 99.6532 aux.loss_ce: 0.0079 aux.acc_seg: 99.0548 +04/18 17:49:41 - mmengine - INFO - Iter(train) [100500/160000] lr: 4.1644e-03 eta: 9:07:50 time: 0.5540 data_time: 0.0069 memory: 7635 loss: 0.0169 decode.loss_ce: 0.0090 decode.acc_seg: 99.7375 aux.loss_ce: 0.0079 aux.acc_seg: 99.4618 +04/18 17:50:08 - mmengine - INFO - Iter(train) [100550/160000] lr: 4.1613e-03 eta: 9:07:22 time: 0.5534 data_time: 0.0062 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0082 decode.acc_seg: 99.7165 aux.loss_ce: 0.0078 aux.acc_seg: 99.1873 +04/18 17:50:36 - mmengine - INFO - Iter(train) [100600/160000] lr: 4.1582e-03 eta: 9:06:55 time: 0.5562 data_time: 0.0060 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0096 decode.acc_seg: 99.6803 aux.loss_ce: 0.0081 aux.acc_seg: 99.2851 +04/18 17:51:04 - mmengine - INFO - Iter(train) [100650/160000] lr: 4.1552e-03 eta: 9:06:27 time: 0.5537 data_time: 0.0064 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0082 decode.acc_seg: 99.6935 aux.loss_ce: 0.0077 aux.acc_seg: 99.1413 +04/18 17:51:31 - mmengine - INFO - Iter(train) [100700/160000] lr: 4.1521e-03 eta: 9:06:00 time: 0.5545 data_time: 0.0069 memory: 7635 loss: 0.0176 decode.loss_ce: 0.0091 decode.acc_seg: 99.5483 aux.loss_ce: 0.0086 aux.acc_seg: 98.9243 +04/18 17:51:59 - mmengine - INFO - Iter(train) [100750/160000] lr: 4.1490e-03 eta: 9:05:32 time: 0.5549 data_time: 0.0064 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0084 decode.acc_seg: 99.6213 aux.loss_ce: 0.0083 aux.acc_seg: 99.0323 +04/18 17:52:27 - mmengine - INFO - Iter(train) [100800/160000] lr: 4.1459e-03 eta: 9:05:05 time: 0.5547 data_time: 0.0067 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0083 decode.acc_seg: 99.6117 aux.loss_ce: 0.0078 aux.acc_seg: 99.1339 +04/18 17:52:55 - mmengine - INFO - Iter(train) [100850/160000] lr: 4.1429e-03 eta: 9:04:37 time: 0.5561 data_time: 0.0072 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0074 decode.acc_seg: 99.6839 aux.loss_ce: 0.0078 aux.acc_seg: 99.2059 +04/18 17:53:23 - mmengine - INFO - Iter(train) [100900/160000] lr: 4.1398e-03 eta: 9:04:09 time: 0.5554 data_time: 0.0067 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0075 decode.acc_seg: 99.6908 aux.loss_ce: 0.0073 aux.acc_seg: 99.3784 +04/18 17:53:50 - mmengine - INFO - Iter(train) [100950/160000] lr: 4.1367e-03 eta: 9:03:42 time: 0.5545 data_time: 0.0067 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0077 decode.acc_seg: 99.8081 aux.loss_ce: 0.0076 aux.acc_seg: 99.3566 +04/18 17:54:18 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 17:54:18 - mmengine - INFO - Iter(train) [101000/160000] lr: 4.1336e-03 eta: 9:03:14 time: 0.5547 data_time: 0.0070 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0075 decode.acc_seg: 99.6223 aux.loss_ce: 0.0076 aux.acc_seg: 99.1516 +04/18 17:54:46 - mmengine - INFO - Iter(train) [101050/160000] lr: 4.1306e-03 eta: 9:02:47 time: 0.5568 data_time: 0.0066 memory: 7635 loss: 0.0171 decode.loss_ce: 0.0088 decode.acc_seg: 99.6987 aux.loss_ce: 0.0083 aux.acc_seg: 99.3050 +04/18 17:55:14 - mmengine - INFO - Iter(train) [101100/160000] lr: 4.1275e-03 eta: 9:02:19 time: 0.5555 data_time: 0.0062 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0074 decode.acc_seg: 99.7017 aux.loss_ce: 0.0076 aux.acc_seg: 99.1898 +04/18 17:55:41 - mmengine - INFO - Iter(train) [101150/160000] lr: 4.1244e-03 eta: 9:01:52 time: 0.5542 data_time: 0.0080 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0083 decode.acc_seg: 99.6719 aux.loss_ce: 0.0076 aux.acc_seg: 99.1834 +04/18 17:56:09 - mmengine - INFO - Iter(train) [101200/160000] lr: 4.1213e-03 eta: 9:01:24 time: 0.5546 data_time: 0.0064 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0072 decode.acc_seg: 99.6968 aux.loss_ce: 0.0071 aux.acc_seg: 99.3033 +04/18 17:56:37 - mmengine - INFO - Iter(train) [101250/160000] lr: 4.1182e-03 eta: 9:00:57 time: 0.5557 data_time: 0.0066 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0077 decode.acc_seg: 99.6788 aux.loss_ce: 0.0072 aux.acc_seg: 99.1690 +04/18 17:57:05 - mmengine - INFO - Iter(train) [101300/160000] lr: 4.1152e-03 eta: 9:00:29 time: 0.5537 data_time: 0.0073 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0082 decode.acc_seg: 99.7045 aux.loss_ce: 0.0077 aux.acc_seg: 99.1835 +04/18 17:57:32 - mmengine - INFO - Iter(train) [101350/160000] lr: 4.1121e-03 eta: 9:00:02 time: 0.5534 data_time: 0.0064 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0072 decode.acc_seg: 99.7789 aux.loss_ce: 0.0076 aux.acc_seg: 99.2284 +04/18 17:58:00 - mmengine - INFO - Iter(train) [101400/160000] lr: 4.1090e-03 eta: 8:59:34 time: 0.5628 data_time: 0.0062 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0082 decode.acc_seg: 99.4895 aux.loss_ce: 0.0077 aux.acc_seg: 99.0331 +04/18 17:58:28 - mmengine - INFO - Iter(train) [101450/160000] lr: 4.1059e-03 eta: 8:59:06 time: 0.5548 data_time: 0.0067 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0082 decode.acc_seg: 99.6900 aux.loss_ce: 0.0082 aux.acc_seg: 99.1838 +04/18 17:58:56 - mmengine - INFO - Iter(train) [101500/160000] lr: 4.1029e-03 eta: 8:58:39 time: 0.5533 data_time: 0.0064 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0074 decode.acc_seg: 99.7442 aux.loss_ce: 0.0072 aux.acc_seg: 99.3762 +04/18 17:59:24 - mmengine - INFO - Iter(train) [101550/160000] lr: 4.0998e-03 eta: 8:58:11 time: 0.5647 data_time: 0.0069 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0071 decode.acc_seg: 99.7145 aux.loss_ce: 0.0075 aux.acc_seg: 99.0725 +04/18 17:59:51 - mmengine - INFO - Iter(train) [101600/160000] lr: 4.0967e-03 eta: 8:57:44 time: 0.5545 data_time: 0.0075 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0074 decode.acc_seg: 99.7160 aux.loss_ce: 0.0073 aux.acc_seg: 99.2603 +04/18 18:00:19 - mmengine - INFO - Iter(train) [101650/160000] lr: 4.0936e-03 eta: 8:57:16 time: 0.5540 data_time: 0.0067 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0081 decode.acc_seg: 99.7094 aux.loss_ce: 0.0080 aux.acc_seg: 99.0880 +04/18 18:00:47 - mmengine - INFO - Iter(train) [101700/160000] lr: 4.0905e-03 eta: 8:56:49 time: 0.5558 data_time: 0.0068 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0076 decode.acc_seg: 99.7360 aux.loss_ce: 0.0076 aux.acc_seg: 99.4184 +04/18 18:01:15 - mmengine - INFO - Iter(train) [101750/160000] lr: 4.0875e-03 eta: 8:56:21 time: 0.5538 data_time: 0.0063 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0075 decode.acc_seg: 99.6214 aux.loss_ce: 0.0076 aux.acc_seg: 99.1718 +04/18 18:01:42 - mmengine - INFO - Iter(train) [101800/160000] lr: 4.0844e-03 eta: 8:55:54 time: 0.5538 data_time: 0.0068 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0075 decode.acc_seg: 99.6907 aux.loss_ce: 0.0074 aux.acc_seg: 99.0748 +04/18 18:02:10 - mmengine - INFO - Iter(train) [101850/160000] lr: 4.0813e-03 eta: 8:55:26 time: 0.5539 data_time: 0.0070 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0076 decode.acc_seg: 99.7403 aux.loss_ce: 0.0078 aux.acc_seg: 99.4285 +04/18 18:02:38 - mmengine - INFO - Iter(train) [101900/160000] lr: 4.0782e-03 eta: 8:54:58 time: 0.5543 data_time: 0.0070 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0076 decode.acc_seg: 99.7054 aux.loss_ce: 0.0075 aux.acc_seg: 99.3196 +04/18 18:03:06 - mmengine - INFO - Iter(train) [101950/160000] lr: 4.0751e-03 eta: 8:54:31 time: 0.5563 data_time: 0.0059 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0076 decode.acc_seg: 99.7217 aux.loss_ce: 0.0074 aux.acc_seg: 99.2308 +04/18 18:03:33 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 18:03:33 - mmengine - INFO - Iter(train) [102000/160000] lr: 4.0721e-03 eta: 8:54:03 time: 0.5546 data_time: 0.0056 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0075 decode.acc_seg: 99.6972 aux.loss_ce: 0.0078 aux.acc_seg: 99.2855 +04/18 18:04:01 - mmengine - INFO - Iter(train) [102050/160000] lr: 4.0690e-03 eta: 8:53:36 time: 0.5556 data_time: 0.0066 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0070 decode.acc_seg: 99.7612 aux.loss_ce: 0.0075 aux.acc_seg: 99.3612 +04/18 18:04:29 - mmengine - INFO - Iter(train) [102100/160000] lr: 4.0659e-03 eta: 8:53:08 time: 0.5556 data_time: 0.0069 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0071 decode.acc_seg: 99.7481 aux.loss_ce: 0.0072 aux.acc_seg: 99.3841 +04/18 18:04:57 - mmengine - INFO - Iter(train) [102150/160000] lr: 4.0628e-03 eta: 8:52:41 time: 0.5547 data_time: 0.0071 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0072 decode.acc_seg: 99.7428 aux.loss_ce: 0.0074 aux.acc_seg: 99.4309 +04/18 18:05:25 - mmengine - INFO - Iter(train) [102200/160000] lr: 4.0597e-03 eta: 8:52:13 time: 0.5553 data_time: 0.0069 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0072 decode.acc_seg: 99.7394 aux.loss_ce: 0.0079 aux.acc_seg: 99.1361 +04/18 18:05:52 - mmengine - INFO - Iter(train) [102250/160000] lr: 4.0566e-03 eta: 8:51:46 time: 0.5551 data_time: 0.0073 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0071 decode.acc_seg: 99.7008 aux.loss_ce: 0.0079 aux.acc_seg: 99.1044 +04/18 18:06:20 - mmengine - INFO - Iter(train) [102300/160000] lr: 4.0536e-03 eta: 8:51:18 time: 0.5558 data_time: 0.0067 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0070 decode.acc_seg: 99.7474 aux.loss_ce: 0.0070 aux.acc_seg: 99.4012 +04/18 18:06:48 - mmengine - INFO - Iter(train) [102350/160000] lr: 4.0505e-03 eta: 8:50:51 time: 0.5554 data_time: 0.0069 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0071 decode.acc_seg: 99.6811 aux.loss_ce: 0.0070 aux.acc_seg: 99.3768 +04/18 18:07:16 - mmengine - INFO - Iter(train) [102400/160000] lr: 4.0474e-03 eta: 8:50:23 time: 0.5557 data_time: 0.0065 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0075 decode.acc_seg: 99.5892 aux.loss_ce: 0.0078 aux.acc_seg: 99.0476 +04/18 18:07:43 - mmengine - INFO - Iter(train) [102450/160000] lr: 4.0443e-03 eta: 8:49:56 time: 0.5545 data_time: 0.0068 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0067 decode.acc_seg: 99.7593 aux.loss_ce: 0.0068 aux.acc_seg: 99.3846 +04/18 18:08:11 - mmengine - INFO - Iter(train) [102500/160000] lr: 4.0412e-03 eta: 8:49:28 time: 0.5534 data_time: 0.0071 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0069 decode.acc_seg: 99.7579 aux.loss_ce: 0.0069 aux.acc_seg: 99.1947 +04/18 18:08:39 - mmengine - INFO - Iter(train) [102550/160000] lr: 4.0381e-03 eta: 8:49:01 time: 0.5557 data_time: 0.0073 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0072 decode.acc_seg: 99.7660 aux.loss_ce: 0.0077 aux.acc_seg: 99.3835 +04/18 18:09:07 - mmengine - INFO - Iter(train) [102600/160000] lr: 4.0351e-03 eta: 8:48:33 time: 0.5553 data_time: 0.0072 memory: 7635 loss: 0.0132 decode.loss_ce: 0.0064 decode.acc_seg: 99.7365 aux.loss_ce: 0.0069 aux.acc_seg: 99.1929 +04/18 18:09:35 - mmengine - INFO - Iter(train) [102650/160000] lr: 4.0320e-03 eta: 8:48:05 time: 0.5547 data_time: 0.0077 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0064 decode.acc_seg: 99.7366 aux.loss_ce: 0.0069 aux.acc_seg: 99.3207 +04/18 18:10:02 - mmengine - INFO - Iter(train) [102700/160000] lr: 4.0289e-03 eta: 8:47:38 time: 0.5551 data_time: 0.0065 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0070 decode.acc_seg: 99.7978 aux.loss_ce: 0.0071 aux.acc_seg: 99.5609 +04/18 18:10:30 - mmengine - INFO - Iter(train) [102750/160000] lr: 4.0258e-03 eta: 8:47:10 time: 0.5563 data_time: 0.0063 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0073 decode.acc_seg: 99.6626 aux.loss_ce: 0.0074 aux.acc_seg: 99.0096 +04/18 18:10:58 - mmengine - INFO - Iter(train) [102800/160000] lr: 4.0227e-03 eta: 8:46:43 time: 0.5553 data_time: 0.0066 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0066 decode.acc_seg: 99.7425 aux.loss_ce: 0.0072 aux.acc_seg: 99.2924 +04/18 18:11:26 - mmengine - INFO - Iter(train) [102850/160000] lr: 4.0196e-03 eta: 8:46:15 time: 0.5558 data_time: 0.0063 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0069 decode.acc_seg: 99.6971 aux.loss_ce: 0.0072 aux.acc_seg: 99.1423 +04/18 18:11:54 - mmengine - INFO - Iter(train) [102900/160000] lr: 4.0165e-03 eta: 8:45:48 time: 0.5554 data_time: 0.0061 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0069 decode.acc_seg: 99.7680 aux.loss_ce: 0.0075 aux.acc_seg: 99.3282 +04/18 18:12:21 - mmengine - INFO - Iter(train) [102950/160000] lr: 4.0134e-03 eta: 8:45:20 time: 0.5552 data_time: 0.0064 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0067 decode.acc_seg: 99.7619 aux.loss_ce: 0.0073 aux.acc_seg: 99.3407 +04/18 18:12:49 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 18:12:49 - mmengine - INFO - Iter(train) [103000/160000] lr: 4.0104e-03 eta: 8:44:53 time: 0.5548 data_time: 0.0072 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0079 decode.acc_seg: 99.7692 aux.loss_ce: 0.0080 aux.acc_seg: 99.3644 +04/18 18:13:17 - mmengine - INFO - Iter(train) [103050/160000] lr: 4.0073e-03 eta: 8:44:25 time: 0.5536 data_time: 0.0063 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0071 decode.acc_seg: 99.7481 aux.loss_ce: 0.0071 aux.acc_seg: 99.2605 +04/18 18:13:44 - mmengine - INFO - Iter(train) [103100/160000] lr: 4.0042e-03 eta: 8:43:57 time: 0.5543 data_time: 0.0067 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0066 decode.acc_seg: 99.7490 aux.loss_ce: 0.0071 aux.acc_seg: 99.2566 +04/18 18:14:12 - mmengine - INFO - Iter(train) [103150/160000] lr: 4.0011e-03 eta: 8:43:30 time: 0.5541 data_time: 0.0063 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0076 decode.acc_seg: 99.7028 aux.loss_ce: 0.0080 aux.acc_seg: 99.1325 +04/18 18:14:40 - mmengine - INFO - Iter(train) [103200/160000] lr: 3.9980e-03 eta: 8:43:02 time: 0.5540 data_time: 0.0063 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0070 decode.acc_seg: 99.7586 aux.loss_ce: 0.0076 aux.acc_seg: 99.3856 +04/18 18:15:08 - mmengine - INFO - Iter(train) [103250/160000] lr: 3.9949e-03 eta: 8:42:35 time: 0.5542 data_time: 0.0066 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0065 decode.acc_seg: 99.8170 aux.loss_ce: 0.0068 aux.acc_seg: 99.4820 +04/18 18:15:35 - mmengine - INFO - Iter(train) [103300/160000] lr: 3.9918e-03 eta: 8:42:07 time: 0.5553 data_time: 0.0075 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0077 decode.acc_seg: 99.6673 aux.loss_ce: 0.0083 aux.acc_seg: 99.0726 +04/18 18:16:03 - mmengine - INFO - Iter(train) [103350/160000] lr: 3.9887e-03 eta: 8:41:40 time: 0.5539 data_time: 0.0064 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0070 decode.acc_seg: 99.7534 aux.loss_ce: 0.0077 aux.acc_seg: 99.4044 +04/18 18:16:31 - mmengine - INFO - Iter(train) [103400/160000] lr: 3.9857e-03 eta: 8:41:12 time: 0.5556 data_time: 0.0077 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0074 decode.acc_seg: 99.6698 aux.loss_ce: 0.0081 aux.acc_seg: 98.9725 +04/18 18:16:59 - mmengine - INFO - Iter(train) [103450/160000] lr: 3.9826e-03 eta: 8:40:44 time: 0.5536 data_time: 0.0062 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0067 decode.acc_seg: 99.7816 aux.loss_ce: 0.0073 aux.acc_seg: 99.2882 +04/18 18:17:26 - mmengine - INFO - Iter(train) [103500/160000] lr: 3.9795e-03 eta: 8:40:17 time: 0.5540 data_time: 0.0071 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0081 decode.acc_seg: 99.4746 aux.loss_ce: 0.0076 aux.acc_seg: 99.2682 +04/18 18:17:54 - mmengine - INFO - Iter(train) [103550/160000] lr: 3.9764e-03 eta: 8:39:49 time: 0.5635 data_time: 0.0063 memory: 7635 loss: 0.0166 decode.loss_ce: 0.0084 decode.acc_seg: 99.6692 aux.loss_ce: 0.0082 aux.acc_seg: 99.2133 +04/18 18:18:22 - mmengine - INFO - Iter(train) [103600/160000] lr: 3.9733e-03 eta: 8:39:22 time: 0.5538 data_time: 0.0069 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0077 decode.acc_seg: 99.7363 aux.loss_ce: 0.0079 aux.acc_seg: 99.2662 +04/18 18:18:50 - mmengine - INFO - Iter(train) [103650/160000] lr: 3.9702e-03 eta: 8:38:54 time: 0.5544 data_time: 0.0062 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0072 decode.acc_seg: 99.6899 aux.loss_ce: 0.0070 aux.acc_seg: 99.1610 +04/18 18:19:18 - mmengine - INFO - Iter(train) [103700/160000] lr: 3.9671e-03 eta: 8:38:27 time: 0.5572 data_time: 0.0067 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0072 decode.acc_seg: 99.7503 aux.loss_ce: 0.0074 aux.acc_seg: 99.4403 +04/18 18:19:46 - mmengine - INFO - Iter(train) [103750/160000] lr: 3.9640e-03 eta: 8:37:59 time: 0.5565 data_time: 0.0071 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0071 decode.acc_seg: 99.6954 aux.loss_ce: 0.0079 aux.acc_seg: 99.1589 +04/18 18:20:13 - mmengine - INFO - Iter(train) [103800/160000] lr: 3.9609e-03 eta: 8:37:32 time: 0.5537 data_time: 0.0066 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0076 decode.acc_seg: 99.6759 aux.loss_ce: 0.0080 aux.acc_seg: 99.1508 +04/18 18:20:41 - mmengine - INFO - Iter(train) [103850/160000] lr: 3.9578e-03 eta: 8:37:04 time: 0.5543 data_time: 0.0065 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0092 decode.acc_seg: 99.7488 aux.loss_ce: 0.0078 aux.acc_seg: 99.3685 +04/18 18:21:09 - mmengine - INFO - Iter(train) [103900/160000] lr: 3.9548e-03 eta: 8:36:37 time: 0.5563 data_time: 0.0065 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0072 decode.acc_seg: 99.8124 aux.loss_ce: 0.0073 aux.acc_seg: 99.4577 +04/18 18:21:37 - mmengine - INFO - Iter(train) [103950/160000] lr: 3.9517e-03 eta: 8:36:09 time: 0.5551 data_time: 0.0066 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0071 decode.acc_seg: 99.6668 aux.loss_ce: 0.0074 aux.acc_seg: 98.9861 +04/18 18:22:04 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 18:22:04 - mmengine - INFO - Iter(train) [104000/160000] lr: 3.9486e-03 eta: 8:35:41 time: 0.5547 data_time: 0.0072 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0076 decode.acc_seg: 99.7065 aux.loss_ce: 0.0077 aux.acc_seg: 99.3050 +04/18 18:22:32 - mmengine - INFO - Iter(train) [104050/160000] lr: 3.9455e-03 eta: 8:35:14 time: 0.5551 data_time: 0.0066 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0077 decode.acc_seg: 99.6196 aux.loss_ce: 0.0073 aux.acc_seg: 99.0684 +04/18 18:23:00 - mmengine - INFO - Iter(train) [104100/160000] lr: 3.9424e-03 eta: 8:34:46 time: 0.5548 data_time: 0.0074 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0071 decode.acc_seg: 99.6983 aux.loss_ce: 0.0072 aux.acc_seg: 99.3368 +04/18 18:23:28 - mmengine - INFO - Iter(train) [104150/160000] lr: 3.9393e-03 eta: 8:34:19 time: 0.5542 data_time: 0.0064 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0071 decode.acc_seg: 99.8059 aux.loss_ce: 0.0074 aux.acc_seg: 99.3739 +04/18 18:23:55 - mmengine - INFO - Iter(train) [104200/160000] lr: 3.9362e-03 eta: 8:33:51 time: 0.5543 data_time: 0.0068 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0069 decode.acc_seg: 99.7657 aux.loss_ce: 0.0069 aux.acc_seg: 99.5224 +04/18 18:24:23 - mmengine - INFO - Iter(train) [104250/160000] lr: 3.9331e-03 eta: 8:33:24 time: 0.5543 data_time: 0.0066 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0069 decode.acc_seg: 99.8130 aux.loss_ce: 0.0071 aux.acc_seg: 99.3957 +04/18 18:24:51 - mmengine - INFO - Iter(train) [104300/160000] lr: 3.9300e-03 eta: 8:32:56 time: 0.5541 data_time: 0.0063 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0069 decode.acc_seg: 99.7690 aux.loss_ce: 0.0074 aux.acc_seg: 99.3905 +04/18 18:25:19 - mmengine - INFO - Iter(train) [104350/160000] lr: 3.9269e-03 eta: 8:32:28 time: 0.5543 data_time: 0.0061 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0066 decode.acc_seg: 99.7915 aux.loss_ce: 0.0072 aux.acc_seg: 99.4453 +04/18 18:25:46 - mmengine - INFO - Iter(train) [104400/160000] lr: 3.9238e-03 eta: 8:32:01 time: 0.5546 data_time: 0.0070 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0071 decode.acc_seg: 99.6736 aux.loss_ce: 0.0076 aux.acc_seg: 99.2351 +04/18 18:26:14 - mmengine - INFO - Iter(train) [104450/160000] lr: 3.9207e-03 eta: 8:31:33 time: 0.5548 data_time: 0.0071 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0068 decode.acc_seg: 99.7692 aux.loss_ce: 0.0071 aux.acc_seg: 99.3866 +04/18 18:26:42 - mmengine - INFO - Iter(train) [104500/160000] lr: 3.9176e-03 eta: 8:31:06 time: 0.5535 data_time: 0.0067 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0073 decode.acc_seg: 99.6496 aux.loss_ce: 0.0079 aux.acc_seg: 99.0063 +04/18 18:27:10 - mmengine - INFO - Iter(train) [104550/160000] lr: 3.9145e-03 eta: 8:30:38 time: 0.5545 data_time: 0.0069 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0065 decode.acc_seg: 99.7830 aux.loss_ce: 0.0068 aux.acc_seg: 99.2777 +04/18 18:27:37 - mmengine - INFO - Iter(train) [104600/160000] lr: 3.9114e-03 eta: 8:30:11 time: 0.5547 data_time: 0.0067 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0078 decode.acc_seg: 99.5623 aux.loss_ce: 0.0078 aux.acc_seg: 98.9859 +04/18 18:28:05 - mmengine - INFO - Iter(train) [104650/160000] lr: 3.9083e-03 eta: 8:29:43 time: 0.5553 data_time: 0.0069 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0080 decode.acc_seg: 99.6039 aux.loss_ce: 0.0076 aux.acc_seg: 99.2819 +04/18 18:28:33 - mmengine - INFO - Iter(train) [104700/160000] lr: 3.9052e-03 eta: 8:29:16 time: 0.5540 data_time: 0.0068 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0074 decode.acc_seg: 99.7609 aux.loss_ce: 0.0075 aux.acc_seg: 99.2799 +04/18 18:29:01 - mmengine - INFO - Iter(train) [104750/160000] lr: 3.9021e-03 eta: 8:28:48 time: 0.5547 data_time: 0.0070 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0074 decode.acc_seg: 99.7188 aux.loss_ce: 0.0079 aux.acc_seg: 99.0936 +04/18 18:29:29 - mmengine - INFO - Iter(train) [104800/160000] lr: 3.8990e-03 eta: 8:28:21 time: 0.5535 data_time: 0.0069 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0067 decode.acc_seg: 99.7534 aux.loss_ce: 0.0069 aux.acc_seg: 99.4607 +04/18 18:29:56 - mmengine - INFO - Iter(train) [104850/160000] lr: 3.8960e-03 eta: 8:27:53 time: 0.5539 data_time: 0.0065 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0077 decode.acc_seg: 99.7664 aux.loss_ce: 0.0078 aux.acc_seg: 99.3903 +04/18 18:30:24 - mmengine - INFO - Iter(train) [104900/160000] lr: 3.8929e-03 eta: 8:27:25 time: 0.5549 data_time: 0.0072 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0082 decode.acc_seg: 99.6962 aux.loss_ce: 0.0082 aux.acc_seg: 99.1224 +04/18 18:30:52 - mmengine - INFO - Iter(train) [104950/160000] lr: 3.8898e-03 eta: 8:26:58 time: 0.5541 data_time: 0.0065 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0074 decode.acc_seg: 99.6739 aux.loss_ce: 0.0075 aux.acc_seg: 99.1560 +04/18 18:31:20 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 18:31:20 - mmengine - INFO - Iter(train) [105000/160000] lr: 3.8867e-03 eta: 8:26:30 time: 0.5538 data_time: 0.0060 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0092 decode.acc_seg: 99.6804 aux.loss_ce: 0.0081 aux.acc_seg: 99.2293 +04/18 18:31:47 - mmengine - INFO - Iter(train) [105050/160000] lr: 3.8836e-03 eta: 8:26:03 time: 0.5533 data_time: 0.0061 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0073 decode.acc_seg: 99.7019 aux.loss_ce: 0.0076 aux.acc_seg: 99.3044 +04/18 18:32:15 - mmengine - INFO - Iter(train) [105100/160000] lr: 3.8805e-03 eta: 8:25:35 time: 0.5554 data_time: 0.0066 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0065 decode.acc_seg: 99.7594 aux.loss_ce: 0.0068 aux.acc_seg: 99.5090 +04/18 18:32:43 - mmengine - INFO - Iter(train) [105150/160000] lr: 3.8774e-03 eta: 8:25:08 time: 0.5548 data_time: 0.0066 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0065 decode.acc_seg: 99.7445 aux.loss_ce: 0.0070 aux.acc_seg: 99.3206 +04/18 18:33:11 - mmengine - INFO - Iter(train) [105200/160000] lr: 3.8743e-03 eta: 8:24:40 time: 0.5550 data_time: 0.0066 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0069 decode.acc_seg: 99.7156 aux.loss_ce: 0.0074 aux.acc_seg: 99.3709 +04/18 18:33:38 - mmengine - INFO - Iter(train) [105250/160000] lr: 3.8712e-03 eta: 8:24:12 time: 0.5539 data_time: 0.0062 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0066 decode.acc_seg: 99.7749 aux.loss_ce: 0.0067 aux.acc_seg: 99.3821 +04/18 18:34:06 - mmengine - INFO - Iter(train) [105300/160000] lr: 3.8681e-03 eta: 8:23:45 time: 0.5540 data_time: 0.0065 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0072 decode.acc_seg: 99.7286 aux.loss_ce: 0.0076 aux.acc_seg: 99.3585 +04/18 18:34:34 - mmengine - INFO - Iter(train) [105350/160000] lr: 3.8650e-03 eta: 8:23:17 time: 0.5553 data_time: 0.0069 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0067 decode.acc_seg: 99.7470 aux.loss_ce: 0.0073 aux.acc_seg: 99.2317 +04/18 18:35:02 - mmengine - INFO - Iter(train) [105400/160000] lr: 3.8619e-03 eta: 8:22:50 time: 0.5548 data_time: 0.0070 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0091 decode.acc_seg: 99.6312 aux.loss_ce: 0.0079 aux.acc_seg: 99.1449 +04/18 18:35:29 - mmengine - INFO - Iter(train) [105450/160000] lr: 3.8588e-03 eta: 8:22:22 time: 0.5547 data_time: 0.0064 memory: 7635 loss: 0.0180 decode.loss_ce: 0.0096 decode.acc_seg: 99.5889 aux.loss_ce: 0.0084 aux.acc_seg: 99.1507 +04/18 18:35:57 - mmengine - INFO - Iter(train) [105500/160000] lr: 3.8557e-03 eta: 8:21:55 time: 0.5541 data_time: 0.0065 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0077 decode.acc_seg: 99.5057 aux.loss_ce: 0.0075 aux.acc_seg: 99.1350 +04/18 18:36:25 - mmengine - INFO - Iter(train) [105550/160000] lr: 3.8526e-03 eta: 8:21:27 time: 0.5560 data_time: 0.0069 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0079 decode.acc_seg: 99.6590 aux.loss_ce: 0.0080 aux.acc_seg: 98.8886 +04/18 18:36:53 - mmengine - INFO - Iter(train) [105600/160000] lr: 3.8495e-03 eta: 8:20:59 time: 0.5535 data_time: 0.0071 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0082 decode.acc_seg: 99.7232 aux.loss_ce: 0.0081 aux.acc_seg: 99.2006 +04/18 18:37:21 - mmengine - INFO - Iter(train) [105650/160000] lr: 3.8464e-03 eta: 8:20:32 time: 0.5552 data_time: 0.0070 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0077 decode.acc_seg: 99.6098 aux.loss_ce: 0.0080 aux.acc_seg: 98.9934 +04/18 18:37:48 - mmengine - INFO - Iter(train) [105700/160000] lr: 3.8433e-03 eta: 8:20:04 time: 0.5629 data_time: 0.0066 memory: 7635 loss: 0.0175 decode.loss_ce: 0.0091 decode.acc_seg: 99.6754 aux.loss_ce: 0.0084 aux.acc_seg: 99.1598 +04/18 18:38:16 - mmengine - INFO - Iter(train) [105750/160000] lr: 3.8402e-03 eta: 8:19:37 time: 0.5532 data_time: 0.0064 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0075 decode.acc_seg: 99.7292 aux.loss_ce: 0.0080 aux.acc_seg: 99.3091 +04/18 18:38:44 - mmengine - INFO - Iter(train) [105800/160000] lr: 3.8371e-03 eta: 8:19:09 time: 0.5552 data_time: 0.0061 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0071 decode.acc_seg: 99.7055 aux.loss_ce: 0.0074 aux.acc_seg: 99.2315 +04/18 18:39:12 - mmengine - INFO - Iter(train) [105850/160000] lr: 3.8339e-03 eta: 8:18:42 time: 0.5543 data_time: 0.0066 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0069 decode.acc_seg: 99.6668 aux.loss_ce: 0.0073 aux.acc_seg: 99.1922 +04/18 18:39:40 - mmengine - INFO - Iter(train) [105900/160000] lr: 3.8308e-03 eta: 8:18:14 time: 0.5559 data_time: 0.0070 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0065 decode.acc_seg: 99.8177 aux.loss_ce: 0.0070 aux.acc_seg: 99.5051 +04/18 18:40:07 - mmengine - INFO - Iter(train) [105950/160000] lr: 3.8277e-03 eta: 8:17:47 time: 0.5563 data_time: 0.0068 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0074 decode.acc_seg: 99.6799 aux.loss_ce: 0.0079 aux.acc_seg: 99.0003 +04/18 18:40:35 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 18:40:35 - mmengine - INFO - Iter(train) [106000/160000] lr: 3.8246e-03 eta: 8:17:19 time: 0.5556 data_time: 0.0063 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0067 decode.acc_seg: 99.7479 aux.loss_ce: 0.0071 aux.acc_seg: 99.2925 +04/18 18:41:03 - mmengine - INFO - Iter(train) [106050/160000] lr: 3.8215e-03 eta: 8:16:52 time: 0.5543 data_time: 0.0070 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0063 decode.acc_seg: 99.7302 aux.loss_ce: 0.0067 aux.acc_seg: 99.1390 +04/18 18:41:31 - mmengine - INFO - Iter(train) [106100/160000] lr: 3.8184e-03 eta: 8:16:24 time: 0.5549 data_time: 0.0068 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0068 decode.acc_seg: 99.7599 aux.loss_ce: 0.0074 aux.acc_seg: 99.2908 +04/18 18:41:58 - mmengine - INFO - Iter(train) [106150/160000] lr: 3.8153e-03 eta: 8:15:56 time: 0.5550 data_time: 0.0063 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0086 decode.acc_seg: 99.6630 aux.loss_ce: 0.0078 aux.acc_seg: 99.1592 +04/18 18:42:26 - mmengine - INFO - Iter(train) [106200/160000] lr: 3.8122e-03 eta: 8:15:29 time: 0.5537 data_time: 0.0066 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0081 decode.acc_seg: 99.5579 aux.loss_ce: 0.0074 aux.acc_seg: 99.1849 +04/18 18:42:54 - mmengine - INFO - Iter(train) [106250/160000] lr: 3.8091e-03 eta: 8:15:01 time: 0.5543 data_time: 0.0064 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0077 decode.acc_seg: 99.6932 aux.loss_ce: 0.0078 aux.acc_seg: 99.1446 +04/18 18:43:22 - mmengine - INFO - Iter(train) [106300/160000] lr: 3.8060e-03 eta: 8:14:34 time: 0.5545 data_time: 0.0070 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0075 decode.acc_seg: 99.6707 aux.loss_ce: 0.0074 aux.acc_seg: 99.2505 +04/18 18:43:49 - mmengine - INFO - Iter(train) [106350/160000] lr: 3.8029e-03 eta: 8:14:06 time: 0.5537 data_time: 0.0061 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0080 decode.acc_seg: 99.6784 aux.loss_ce: 0.0084 aux.acc_seg: 99.0918 +04/18 18:44:17 - mmengine - INFO - Iter(train) [106400/160000] lr: 3.7998e-03 eta: 8:13:38 time: 0.5547 data_time: 0.0072 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0070 decode.acc_seg: 99.5859 aux.loss_ce: 0.0075 aux.acc_seg: 98.9934 +04/18 18:44:45 - mmengine - INFO - Iter(train) [106450/160000] lr: 3.7967e-03 eta: 8:13:11 time: 0.5557 data_time: 0.0063 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0067 decode.acc_seg: 99.8007 aux.loss_ce: 0.0072 aux.acc_seg: 99.3562 +04/18 18:45:13 - mmengine - INFO - Iter(train) [106500/160000] lr: 3.7936e-03 eta: 8:12:43 time: 0.5548 data_time: 0.0065 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0077 decode.acc_seg: 99.6293 aux.loss_ce: 0.0076 aux.acc_seg: 99.0341 +04/18 18:45:40 - mmengine - INFO - Iter(train) [106550/160000] lr: 3.7905e-03 eta: 8:12:16 time: 0.5560 data_time: 0.0068 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0078 decode.acc_seg: 99.7142 aux.loss_ce: 0.0072 aux.acc_seg: 99.3312 +04/18 18:46:08 - mmengine - INFO - Iter(train) [106600/160000] lr: 3.7874e-03 eta: 8:11:48 time: 0.5542 data_time: 0.0066 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0088 decode.acc_seg: 99.6775 aux.loss_ce: 0.0080 aux.acc_seg: 99.2909 +04/18 18:46:36 - mmengine - INFO - Iter(train) [106650/160000] lr: 3.7843e-03 eta: 8:11:21 time: 0.5556 data_time: 0.0065 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0078 decode.acc_seg: 99.7286 aux.loss_ce: 0.0074 aux.acc_seg: 99.2894 +04/18 18:47:04 - mmengine - INFO - Iter(train) [106700/160000] lr: 3.7812e-03 eta: 8:10:53 time: 0.5543 data_time: 0.0070 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0079 decode.acc_seg: 99.7343 aux.loss_ce: 0.0080 aux.acc_seg: 99.1952 +04/18 18:47:32 - mmengine - INFO - Iter(train) [106750/160000] lr: 3.7780e-03 eta: 8:10:25 time: 0.5556 data_time: 0.0064 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0085 decode.acc_seg: 99.7152 aux.loss_ce: 0.0080 aux.acc_seg: 99.2158 +04/18 18:48:00 - mmengine - INFO - Iter(train) [106800/160000] lr: 3.7749e-03 eta: 8:09:58 time: 0.5551 data_time: 0.0060 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0082 decode.acc_seg: 99.6323 aux.loss_ce: 0.0079 aux.acc_seg: 99.2817 +04/18 18:48:27 - mmengine - INFO - Iter(train) [106850/160000] lr: 3.7718e-03 eta: 8:09:31 time: 0.5556 data_time: 0.0071 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0080 decode.acc_seg: 99.6947 aux.loss_ce: 0.0079 aux.acc_seg: 99.2083 +04/18 18:48:55 - mmengine - INFO - Iter(train) [106900/160000] lr: 3.7687e-03 eta: 8:09:03 time: 0.5559 data_time: 0.0065 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0074 decode.acc_seg: 99.6970 aux.loss_ce: 0.0075 aux.acc_seg: 99.2016 +04/18 18:49:23 - mmengine - INFO - Iter(train) [106950/160000] lr: 3.7656e-03 eta: 8:08:35 time: 0.5625 data_time: 0.0072 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0069 decode.acc_seg: 99.7646 aux.loss_ce: 0.0072 aux.acc_seg: 99.2825 +04/18 18:49:51 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 18:49:51 - mmengine - INFO - Iter(train) [107000/160000] lr: 3.7625e-03 eta: 8:08:08 time: 0.5543 data_time: 0.0064 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0071 decode.acc_seg: 99.7506 aux.loss_ce: 0.0077 aux.acc_seg: 99.2103 +04/18 18:50:18 - mmengine - INFO - Iter(train) [107050/160000] lr: 3.7594e-03 eta: 8:07:40 time: 0.5541 data_time: 0.0066 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0079 decode.acc_seg: 99.7283 aux.loss_ce: 0.0076 aux.acc_seg: 99.3507 +04/18 18:50:46 - mmengine - INFO - Iter(train) [107100/160000] lr: 3.7563e-03 eta: 8:07:13 time: 0.5571 data_time: 0.0063 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0075 decode.acc_seg: 99.6904 aux.loss_ce: 0.0075 aux.acc_seg: 99.0345 +04/18 18:51:14 - mmengine - INFO - Iter(train) [107150/160000] lr: 3.7532e-03 eta: 8:06:45 time: 0.5549 data_time: 0.0069 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0075 decode.acc_seg: 99.7509 aux.loss_ce: 0.0077 aux.acc_seg: 99.4362 +04/18 18:51:42 - mmengine - INFO - Iter(train) [107200/160000] lr: 3.7501e-03 eta: 8:06:18 time: 0.5553 data_time: 0.0075 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0067 decode.acc_seg: 99.5761 aux.loss_ce: 0.0073 aux.acc_seg: 98.9529 +04/18 18:52:10 - mmengine - INFO - Iter(train) [107250/160000] lr: 3.7470e-03 eta: 8:05:50 time: 0.5565 data_time: 0.0061 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0072 decode.acc_seg: 99.6921 aux.loss_ce: 0.0075 aux.acc_seg: 99.1709 +04/18 18:52:37 - mmengine - INFO - Iter(train) [107300/160000] lr: 3.7438e-03 eta: 8:05:22 time: 0.5551 data_time: 0.0067 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0069 decode.acc_seg: 99.7155 aux.loss_ce: 0.0074 aux.acc_seg: 99.2370 +04/18 18:53:05 - mmengine - INFO - Iter(train) [107350/160000] lr: 3.7407e-03 eta: 8:04:55 time: 0.5540 data_time: 0.0076 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0080 decode.acc_seg: 99.6866 aux.loss_ce: 0.0079 aux.acc_seg: 99.1728 +04/18 18:53:33 - mmengine - INFO - Iter(train) [107400/160000] lr: 3.7376e-03 eta: 8:04:27 time: 0.5555 data_time: 0.0066 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0068 decode.acc_seg: 99.7092 aux.loss_ce: 0.0075 aux.acc_seg: 99.1659 +04/18 18:54:01 - mmengine - INFO - Iter(train) [107450/160000] lr: 3.7345e-03 eta: 8:04:00 time: 0.5558 data_time: 0.0068 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0067 decode.acc_seg: 99.6946 aux.loss_ce: 0.0071 aux.acc_seg: 99.2869 +04/18 18:54:28 - mmengine - INFO - Iter(train) [107500/160000] lr: 3.7314e-03 eta: 8:03:32 time: 0.5542 data_time: 0.0070 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0069 decode.acc_seg: 99.7839 aux.loss_ce: 0.0071 aux.acc_seg: 99.3629 +04/18 18:54:56 - mmengine - INFO - Iter(train) [107550/160000] lr: 3.7283e-03 eta: 8:03:05 time: 0.5542 data_time: 0.0071 memory: 7635 loss: 0.0132 decode.loss_ce: 0.0062 decode.acc_seg: 99.7341 aux.loss_ce: 0.0069 aux.acc_seg: 99.3241 +04/18 18:55:24 - mmengine - INFO - Iter(train) [107600/160000] lr: 3.7252e-03 eta: 8:02:37 time: 0.5525 data_time: 0.0068 memory: 7635 loss: 0.0124 decode.loss_ce: 0.0058 decode.acc_seg: 99.7838 aux.loss_ce: 0.0065 aux.acc_seg: 99.1969 +04/18 18:55:51 - mmengine - INFO - Iter(train) [107650/160000] lr: 3.7221e-03 eta: 8:02:09 time: 0.5545 data_time: 0.0067 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0067 decode.acc_seg: 99.6625 aux.loss_ce: 0.0074 aux.acc_seg: 99.0898 +04/18 18:56:19 - mmengine - INFO - Iter(train) [107700/160000] lr: 3.7189e-03 eta: 8:01:42 time: 0.5538 data_time: 0.0065 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0069 decode.acc_seg: 99.7855 aux.loss_ce: 0.0071 aux.acc_seg: 99.3336 +04/18 18:56:47 - mmengine - INFO - Iter(train) [107750/160000] lr: 3.7158e-03 eta: 8:01:14 time: 0.5549 data_time: 0.0067 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0069 decode.acc_seg: 99.6257 aux.loss_ce: 0.0072 aux.acc_seg: 99.2513 +04/18 18:57:15 - mmengine - INFO - Iter(train) [107800/160000] lr: 3.7127e-03 eta: 8:00:47 time: 0.5551 data_time: 0.0070 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0072 decode.acc_seg: 99.7494 aux.loss_ce: 0.0077 aux.acc_seg: 99.2947 +04/18 18:57:43 - mmengine - INFO - Iter(train) [107850/160000] lr: 3.7096e-03 eta: 8:00:19 time: 0.5542 data_time: 0.0062 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0065 decode.acc_seg: 99.7247 aux.loss_ce: 0.0067 aux.acc_seg: 99.3230 +04/18 18:58:11 - mmengine - INFO - Iter(train) [107900/160000] lr: 3.7065e-03 eta: 7:59:52 time: 0.5541 data_time: 0.0063 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0072 decode.acc_seg: 99.6934 aux.loss_ce: 0.0075 aux.acc_seg: 99.1262 +04/18 18:58:38 - mmengine - INFO - Iter(train) [107950/160000] lr: 3.7034e-03 eta: 7:59:24 time: 0.5556 data_time: 0.0067 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0069 decode.acc_seg: 99.7154 aux.loss_ce: 0.0072 aux.acc_seg: 99.3361 +04/18 18:59:06 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 18:59:06 - mmengine - INFO - Iter(train) [108000/160000] lr: 3.7003e-03 eta: 7:58:56 time: 0.5529 data_time: 0.0066 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0072 decode.acc_seg: 99.5666 aux.loss_ce: 0.0079 aux.acc_seg: 98.5047 +04/18 18:59:34 - mmengine - INFO - Iter(train) [108050/160000] lr: 3.6971e-03 eta: 7:58:29 time: 0.5557 data_time: 0.0064 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0071 decode.acc_seg: 99.7347 aux.loss_ce: 0.0075 aux.acc_seg: 99.1542 +04/18 19:00:02 - mmengine - INFO - Iter(train) [108100/160000] lr: 3.6940e-03 eta: 7:58:01 time: 0.5547 data_time: 0.0069 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0067 decode.acc_seg: 99.7680 aux.loss_ce: 0.0069 aux.acc_seg: 99.2745 +04/18 19:00:29 - mmengine - INFO - Iter(train) [108150/160000] lr: 3.6909e-03 eta: 7:57:34 time: 0.5542 data_time: 0.0065 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0066 decode.acc_seg: 99.6890 aux.loss_ce: 0.0070 aux.acc_seg: 99.1372 +04/18 19:00:57 - mmengine - INFO - Iter(train) [108200/160000] lr: 3.6878e-03 eta: 7:57:06 time: 0.5549 data_time: 0.0074 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0065 decode.acc_seg: 99.7164 aux.loss_ce: 0.0070 aux.acc_seg: 99.0399 +04/18 19:01:25 - mmengine - INFO - Iter(train) [108250/160000] lr: 3.6847e-03 eta: 7:56:39 time: 0.5550 data_time: 0.0064 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0067 decode.acc_seg: 99.7497 aux.loss_ce: 0.0071 aux.acc_seg: 99.2955 +04/18 19:01:53 - mmengine - INFO - Iter(train) [108300/160000] lr: 3.6816e-03 eta: 7:56:11 time: 0.5561 data_time: 0.0063 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0066 decode.acc_seg: 99.7405 aux.loss_ce: 0.0068 aux.acc_seg: 99.2576 +04/18 19:02:20 - mmengine - INFO - Iter(train) [108350/160000] lr: 3.6784e-03 eta: 7:55:43 time: 0.5559 data_time: 0.0069 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0067 decode.acc_seg: 99.7471 aux.loss_ce: 0.0071 aux.acc_seg: 99.3069 +04/18 19:02:48 - mmengine - INFO - Iter(train) [108400/160000] lr: 3.6753e-03 eta: 7:55:16 time: 0.5539 data_time: 0.0070 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0070 decode.acc_seg: 99.7011 aux.loss_ce: 0.0071 aux.acc_seg: 99.3650 +04/18 19:03:16 - mmengine - INFO - Iter(train) [108450/160000] lr: 3.6722e-03 eta: 7:54:48 time: 0.5553 data_time: 0.0065 memory: 7635 loss: 0.0125 decode.loss_ce: 0.0061 decode.acc_seg: 99.7404 aux.loss_ce: 0.0064 aux.acc_seg: 99.4333 +04/18 19:03:44 - mmengine - INFO - Iter(train) [108500/160000] lr: 3.6691e-03 eta: 7:54:21 time: 0.5537 data_time: 0.0064 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0065 decode.acc_seg: 99.7160 aux.loss_ce: 0.0071 aux.acc_seg: 99.0922 +04/18 19:04:12 - mmengine - INFO - Iter(train) [108550/160000] lr: 3.6660e-03 eta: 7:53:53 time: 0.5558 data_time: 0.0066 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0069 decode.acc_seg: 99.6211 aux.loss_ce: 0.0075 aux.acc_seg: 99.1413 +04/18 19:04:39 - mmengine - INFO - Iter(train) [108600/160000] lr: 3.6628e-03 eta: 7:53:26 time: 0.5554 data_time: 0.0065 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0067 decode.acc_seg: 99.6576 aux.loss_ce: 0.0068 aux.acc_seg: 99.2223 +04/18 19:05:07 - mmengine - INFO - Iter(train) [108650/160000] lr: 3.6597e-03 eta: 7:52:58 time: 0.5551 data_time: 0.0065 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0064 decode.acc_seg: 99.7253 aux.loss_ce: 0.0070 aux.acc_seg: 99.1011 +04/18 19:05:35 - mmengine - INFO - Iter(train) [108700/160000] lr: 3.6566e-03 eta: 7:52:30 time: 0.5552 data_time: 0.0061 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0069 decode.acc_seg: 99.6749 aux.loss_ce: 0.0068 aux.acc_seg: 99.3142 +04/18 19:06:03 - mmengine - INFO - Iter(train) [108750/160000] lr: 3.6535e-03 eta: 7:52:03 time: 0.5544 data_time: 0.0077 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0063 decode.acc_seg: 99.6973 aux.loss_ce: 0.0067 aux.acc_seg: 99.2664 +04/18 19:06:30 - mmengine - INFO - Iter(train) [108800/160000] lr: 3.6504e-03 eta: 7:51:35 time: 0.5554 data_time: 0.0075 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0070 decode.acc_seg: 99.8243 aux.loss_ce: 0.0076 aux.acc_seg: 99.4201 +04/18 19:06:58 - mmengine - INFO - Iter(train) [108850/160000] lr: 3.6472e-03 eta: 7:51:08 time: 0.5546 data_time: 0.0074 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0068 decode.acc_seg: 99.6761 aux.loss_ce: 0.0073 aux.acc_seg: 99.2339 +04/18 19:07:26 - mmengine - INFO - Iter(train) [108900/160000] lr: 3.6441e-03 eta: 7:50:40 time: 0.5529 data_time: 0.0064 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0067 decode.acc_seg: 99.6829 aux.loss_ce: 0.0073 aux.acc_seg: 99.0892 +04/18 19:07:54 - mmengine - INFO - Iter(train) [108950/160000] lr: 3.6410e-03 eta: 7:50:13 time: 0.5547 data_time: 0.0073 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0066 decode.acc_seg: 99.7084 aux.loss_ce: 0.0075 aux.acc_seg: 99.1124 +04/18 19:08:22 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 19:08:22 - mmengine - INFO - Iter(train) [109000/160000] lr: 3.6379e-03 eta: 7:49:45 time: 0.5546 data_time: 0.0062 memory: 7635 loss: 0.0128 decode.loss_ce: 0.0061 decode.acc_seg: 99.8028 aux.loss_ce: 0.0067 aux.acc_seg: 99.3844 +04/18 19:08:49 - mmengine - INFO - Iter(train) [109050/160000] lr: 3.6348e-03 eta: 7:49:17 time: 0.5520 data_time: 0.0064 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0067 decode.acc_seg: 99.7956 aux.loss_ce: 0.0071 aux.acc_seg: 99.3993 +04/18 19:09:17 - mmengine - INFO - Iter(train) [109100/160000] lr: 3.6316e-03 eta: 7:48:50 time: 0.5635 data_time: 0.0070 memory: 7635 loss: 0.0131 decode.loss_ce: 0.0064 decode.acc_seg: 99.7819 aux.loss_ce: 0.0068 aux.acc_seg: 99.3765 +04/18 19:09:45 - mmengine - INFO - Iter(train) [109150/160000] lr: 3.6285e-03 eta: 7:48:22 time: 0.5554 data_time: 0.0070 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0073 decode.acc_seg: 99.7561 aux.loss_ce: 0.0077 aux.acc_seg: 99.4558 +04/18 19:10:13 - mmengine - INFO - Iter(train) [109200/160000] lr: 3.6254e-03 eta: 7:47:55 time: 0.5543 data_time: 0.0063 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0071 decode.acc_seg: 99.6473 aux.loss_ce: 0.0070 aux.acc_seg: 99.2697 +04/18 19:10:41 - mmengine - INFO - Iter(train) [109250/160000] lr: 3.6223e-03 eta: 7:47:27 time: 0.5551 data_time: 0.0070 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0067 decode.acc_seg: 99.7118 aux.loss_ce: 0.0068 aux.acc_seg: 99.1602 +04/18 19:11:08 - mmengine - INFO - Iter(train) [109300/160000] lr: 3.6191e-03 eta: 7:47:00 time: 0.5561 data_time: 0.0071 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0068 decode.acc_seg: 99.7101 aux.loss_ce: 0.0072 aux.acc_seg: 99.0958 +04/18 19:11:36 - mmengine - INFO - Iter(train) [109350/160000] lr: 3.6160e-03 eta: 7:46:32 time: 0.5540 data_time: 0.0068 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0079 decode.acc_seg: 99.6615 aux.loss_ce: 0.0075 aux.acc_seg: 99.2090 +04/18 19:12:04 - mmengine - INFO - Iter(train) [109400/160000] lr: 3.6129e-03 eta: 7:46:04 time: 0.5555 data_time: 0.0069 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0077 decode.acc_seg: 99.6904 aux.loss_ce: 0.0077 aux.acc_seg: 99.2303 +04/18 19:12:32 - mmengine - INFO - Iter(train) [109450/160000] lr: 3.6098e-03 eta: 7:45:37 time: 0.5556 data_time: 0.0070 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0075 decode.acc_seg: 99.7541 aux.loss_ce: 0.0078 aux.acc_seg: 99.2632 +04/18 19:12:59 - mmengine - INFO - Iter(train) [109500/160000] lr: 3.6066e-03 eta: 7:45:09 time: 0.5546 data_time: 0.0071 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0082 decode.acc_seg: 99.7097 aux.loss_ce: 0.0075 aux.acc_seg: 99.2345 +04/18 19:13:27 - mmengine - INFO - Iter(train) [109550/160000] lr: 3.6035e-03 eta: 7:44:42 time: 0.5552 data_time: 0.0069 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0080 decode.acc_seg: 99.7431 aux.loss_ce: 0.0071 aux.acc_seg: 99.4443 +04/18 19:13:55 - mmengine - INFO - Iter(train) [109600/160000] lr: 3.6004e-03 eta: 7:44:14 time: 0.5556 data_time: 0.0078 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0073 decode.acc_seg: 99.6223 aux.loss_ce: 0.0073 aux.acc_seg: 99.1431 +04/18 19:14:23 - mmengine - INFO - Iter(train) [109650/160000] lr: 3.5973e-03 eta: 7:43:47 time: 0.5559 data_time: 0.0066 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0076 decode.acc_seg: 99.6944 aux.loss_ce: 0.0072 aux.acc_seg: 99.1576 +04/18 19:14:50 - mmengine - INFO - Iter(train) [109700/160000] lr: 3.5941e-03 eta: 7:43:19 time: 0.5559 data_time: 0.0068 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0067 decode.acc_seg: 99.7169 aux.loss_ce: 0.0069 aux.acc_seg: 99.2173 +04/18 19:15:18 - mmengine - INFO - Iter(train) [109750/160000] lr: 3.5910e-03 eta: 7:42:51 time: 0.5562 data_time: 0.0062 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0069 decode.acc_seg: 99.6673 aux.loss_ce: 0.0072 aux.acc_seg: 99.1226 +04/18 19:15:46 - mmengine - INFO - Iter(train) [109800/160000] lr: 3.5879e-03 eta: 7:42:24 time: 0.5551 data_time: 0.0072 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0068 decode.acc_seg: 99.7271 aux.loss_ce: 0.0067 aux.acc_seg: 99.3597 +04/18 19:16:14 - mmengine - INFO - Iter(train) [109850/160000] lr: 3.5848e-03 eta: 7:41:56 time: 0.5554 data_time: 0.0070 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0071 decode.acc_seg: 99.6505 aux.loss_ce: 0.0075 aux.acc_seg: 99.0623 +04/18 19:16:41 - mmengine - INFO - Iter(train) [109900/160000] lr: 3.5816e-03 eta: 7:41:29 time: 0.5546 data_time: 0.0067 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0070 decode.acc_seg: 99.7892 aux.loss_ce: 0.0072 aux.acc_seg: 99.2321 +04/18 19:17:09 - mmengine - INFO - Iter(train) [109950/160000] lr: 3.5785e-03 eta: 7:41:01 time: 0.5538 data_time: 0.0065 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0082 decode.acc_seg: 99.6477 aux.loss_ce: 0.0079 aux.acc_seg: 99.0071 +04/18 19:17:37 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 19:17:37 - mmengine - INFO - Iter(train) [110000/160000] lr: 3.5754e-03 eta: 7:40:34 time: 0.5638 data_time: 0.0067 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0068 decode.acc_seg: 99.7181 aux.loss_ce: 0.0071 aux.acc_seg: 99.2119 +04/18 19:17:37 - mmengine - INFO - Saving checkpoint at 110000 iterations +04/18 19:17:41 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:07 time: 0.0464 data_time: 0.0015 memory: 1657 +04/18 19:17:43 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:04 time: 0.0466 data_time: 0.0015 memory: 1657 +04/18 19:17:46 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:02 time: 0.0466 data_time: 0.0014 memory: 1657 +04/18 19:17:48 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.0453 data_time: 0.0011 memory: 1657 +04/18 19:17:48 - mmengine - INFO - per class results: +04/18 19:17:48 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 98.93 | 99.5 | 99.46 | 99.42 | 99.5 | +| contrast | 76.86 | 86.05 | 86.91 | 87.8 | 86.05 | ++------------+-------+-------+--------+-----------+--------+ +04/18 19:17:48 - mmengine - INFO - Iter(val) [200/200] aAcc: 98.9700 mIoU: 87.8900 mAcc: 92.7800 mFscore: 93.1900 mPrecision: 93.6100 mRecall: 92.7800 data_time: 0.0015 time: 0.0464 +04/18 19:18:16 - mmengine - INFO - Iter(train) [110050/160000] lr: 3.5723e-03 eta: 7:40:06 time: 0.5549 data_time: 0.0067 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0080 decode.acc_seg: 99.7857 aux.loss_ce: 0.0078 aux.acc_seg: 99.4260 +04/18 19:18:44 - mmengine - INFO - Iter(train) [110100/160000] lr: 3.5691e-03 eta: 7:39:39 time: 0.5547 data_time: 0.0073 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0076 decode.acc_seg: 99.6800 aux.loss_ce: 0.0081 aux.acc_seg: 99.0413 +04/18 19:19:12 - mmengine - INFO - Iter(train) [110150/160000] lr: 3.5660e-03 eta: 7:39:11 time: 0.5560 data_time: 0.0070 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0077 decode.acc_seg: 99.6807 aux.loss_ce: 0.0074 aux.acc_seg: 99.0971 +04/18 19:19:39 - mmengine - INFO - Iter(train) [110200/160000] lr: 3.5629e-03 eta: 7:38:43 time: 0.5548 data_time: 0.0069 memory: 7635 loss: 0.0132 decode.loss_ce: 0.0066 decode.acc_seg: 99.7349 aux.loss_ce: 0.0066 aux.acc_seg: 99.3257 +04/18 19:20:07 - mmengine - INFO - Iter(train) [110250/160000] lr: 3.5597e-03 eta: 7:38:16 time: 0.5544 data_time: 0.0069 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0070 decode.acc_seg: 99.7406 aux.loss_ce: 0.0071 aux.acc_seg: 99.2209 +04/18 19:20:35 - mmengine - INFO - Iter(train) [110300/160000] lr: 3.5566e-03 eta: 7:37:48 time: 0.5554 data_time: 0.0068 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0068 decode.acc_seg: 99.6781 aux.loss_ce: 0.0071 aux.acc_seg: 99.2702 +04/18 19:21:03 - mmengine - INFO - Iter(train) [110350/160000] lr: 3.5535e-03 eta: 7:37:21 time: 0.5537 data_time: 0.0063 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0069 decode.acc_seg: 99.7687 aux.loss_ce: 0.0074 aux.acc_seg: 99.3010 +04/18 19:21:30 - mmengine - INFO - Iter(train) [110400/160000] lr: 3.5504e-03 eta: 7:36:53 time: 0.5538 data_time: 0.0065 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0069 decode.acc_seg: 99.7117 aux.loss_ce: 0.0075 aux.acc_seg: 99.1607 +04/18 19:21:58 - mmengine - INFO - Iter(train) [110450/160000] lr: 3.5472e-03 eta: 7:36:25 time: 0.5547 data_time: 0.0061 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0069 decode.acc_seg: 99.6261 aux.loss_ce: 0.0072 aux.acc_seg: 99.0324 +04/18 19:22:26 - mmengine - INFO - Iter(train) [110500/160000] lr: 3.5441e-03 eta: 7:35:58 time: 0.5656 data_time: 0.0066 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0065 decode.acc_seg: 99.7314 aux.loss_ce: 0.0073 aux.acc_seg: 99.1668 +04/18 19:22:54 - mmengine - INFO - Iter(train) [110550/160000] lr: 3.5410e-03 eta: 7:35:30 time: 0.5550 data_time: 0.0065 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0072 decode.acc_seg: 99.6418 aux.loss_ce: 0.0075 aux.acc_seg: 99.0856 +04/18 19:23:22 - mmengine - INFO - Iter(train) [110600/160000] lr: 3.5378e-03 eta: 7:35:03 time: 0.5529 data_time: 0.0067 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0065 decode.acc_seg: 99.7288 aux.loss_ce: 0.0073 aux.acc_seg: 99.1767 +04/18 19:23:49 - mmengine - INFO - Iter(train) [110650/160000] lr: 3.5347e-03 eta: 7:34:35 time: 0.5557 data_time: 0.0070 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0069 decode.acc_seg: 99.7032 aux.loss_ce: 0.0075 aux.acc_seg: 99.1522 +04/18 19:24:17 - mmengine - INFO - Iter(train) [110700/160000] lr: 3.5316e-03 eta: 7:34:08 time: 0.5549 data_time: 0.0069 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0069 decode.acc_seg: 99.7876 aux.loss_ce: 0.0070 aux.acc_seg: 99.5091 +04/18 19:24:45 - mmengine - INFO - Iter(train) [110750/160000] lr: 3.5284e-03 eta: 7:33:40 time: 0.5541 data_time: 0.0065 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0064 decode.acc_seg: 99.7217 aux.loss_ce: 0.0071 aux.acc_seg: 99.5057 +04/18 19:25:13 - mmengine - INFO - Iter(train) [110800/160000] lr: 3.5253e-03 eta: 7:33:12 time: 0.5557 data_time: 0.0069 memory: 7635 loss: 0.0131 decode.loss_ce: 0.0063 decode.acc_seg: 99.7141 aux.loss_ce: 0.0069 aux.acc_seg: 99.3195 +04/18 19:25:40 - mmengine - INFO - Iter(train) [110850/160000] lr: 3.5222e-03 eta: 7:32:45 time: 0.5560 data_time: 0.0073 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0068 decode.acc_seg: 99.7230 aux.loss_ce: 0.0077 aux.acc_seg: 99.1342 +04/18 19:26:08 - mmengine - INFO - Iter(train) [110900/160000] lr: 3.5190e-03 eta: 7:32:17 time: 0.5554 data_time: 0.0071 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0068 decode.acc_seg: 99.6499 aux.loss_ce: 0.0075 aux.acc_seg: 99.1427 +04/18 19:26:36 - mmengine - INFO - Iter(train) [110950/160000] lr: 3.5159e-03 eta: 7:31:50 time: 0.5551 data_time: 0.0066 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0065 decode.acc_seg: 99.7088 aux.loss_ce: 0.0068 aux.acc_seg: 99.2192 +04/18 19:27:04 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 19:27:04 - mmengine - INFO - Iter(train) [111000/160000] lr: 3.5128e-03 eta: 7:31:22 time: 0.5542 data_time: 0.0076 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0066 decode.acc_seg: 99.7270 aux.loss_ce: 0.0072 aux.acc_seg: 99.3582 +04/18 19:27:32 - mmengine - INFO - Iter(train) [111050/160000] lr: 3.5096e-03 eta: 7:30:55 time: 0.5552 data_time: 0.0062 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0069 decode.acc_seg: 99.6961 aux.loss_ce: 0.0074 aux.acc_seg: 99.1317 +04/18 19:27:59 - mmengine - INFO - Iter(train) [111100/160000] lr: 3.5065e-03 eta: 7:30:27 time: 0.5537 data_time: 0.0063 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0065 decode.acc_seg: 99.7753 aux.loss_ce: 0.0071 aux.acc_seg: 99.2970 +04/18 19:28:27 - mmengine - INFO - Iter(train) [111150/160000] lr: 3.5034e-03 eta: 7:29:59 time: 0.5545 data_time: 0.0071 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0067 decode.acc_seg: 99.7406 aux.loss_ce: 0.0072 aux.acc_seg: 99.2530 +04/18 19:28:55 - mmengine - INFO - Iter(train) [111200/160000] lr: 3.5002e-03 eta: 7:29:32 time: 0.5551 data_time: 0.0067 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0087 decode.acc_seg: 99.5930 aux.loss_ce: 0.0080 aux.acc_seg: 99.2295 +04/18 19:29:23 - mmengine - INFO - Iter(train) [111250/160000] lr: 3.4971e-03 eta: 7:29:04 time: 0.5539 data_time: 0.0067 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0072 decode.acc_seg: 99.6919 aux.loss_ce: 0.0075 aux.acc_seg: 99.1432 +04/18 19:29:51 - mmengine - INFO - Iter(train) [111300/160000] lr: 3.4940e-03 eta: 7:28:37 time: 0.5560 data_time: 0.0067 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0082 decode.acc_seg: 99.5827 aux.loss_ce: 0.0078 aux.acc_seg: 99.0034 +04/18 19:30:18 - mmengine - INFO - Iter(train) [111350/160000] lr: 3.4908e-03 eta: 7:28:09 time: 0.5544 data_time: 0.0068 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0073 decode.acc_seg: 99.7358 aux.loss_ce: 0.0078 aux.acc_seg: 99.3553 +04/18 19:30:46 - mmengine - INFO - Iter(train) [111400/160000] lr: 3.4877e-03 eta: 7:27:42 time: 0.5548 data_time: 0.0068 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0066 decode.acc_seg: 99.7693 aux.loss_ce: 0.0071 aux.acc_seg: 99.3146 +04/18 19:31:14 - mmengine - INFO - Iter(train) [111450/160000] lr: 3.4845e-03 eta: 7:27:14 time: 0.5535 data_time: 0.0067 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0069 decode.acc_seg: 99.7491 aux.loss_ce: 0.0070 aux.acc_seg: 99.2060 +04/18 19:31:42 - mmengine - INFO - Iter(train) [111500/160000] lr: 3.4814e-03 eta: 7:26:46 time: 0.5545 data_time: 0.0064 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0073 decode.acc_seg: 99.6525 aux.loss_ce: 0.0073 aux.acc_seg: 99.1815 +04/18 19:32:09 - mmengine - INFO - Iter(train) [111550/160000] lr: 3.4783e-03 eta: 7:26:19 time: 0.5545 data_time: 0.0068 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0075 decode.acc_seg: 99.7223 aux.loss_ce: 0.0074 aux.acc_seg: 99.3914 +04/18 19:32:37 - mmengine - INFO - Iter(train) [111600/160000] lr: 3.4751e-03 eta: 7:25:51 time: 0.5552 data_time: 0.0066 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0066 decode.acc_seg: 99.7206 aux.loss_ce: 0.0069 aux.acc_seg: 99.2196 +04/18 19:33:05 - mmengine - INFO - Iter(train) [111650/160000] lr: 3.4720e-03 eta: 7:25:24 time: 0.5526 data_time: 0.0067 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0065 decode.acc_seg: 99.6882 aux.loss_ce: 0.0068 aux.acc_seg: 99.2639 +04/18 19:33:33 - mmengine - INFO - Iter(train) [111700/160000] lr: 3.4689e-03 eta: 7:24:56 time: 0.5563 data_time: 0.0067 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0067 decode.acc_seg: 99.6841 aux.loss_ce: 0.0071 aux.acc_seg: 99.1851 +04/18 19:34:00 - mmengine - INFO - Iter(train) [111750/160000] lr: 3.4657e-03 eta: 7:24:28 time: 0.5552 data_time: 0.0063 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0072 decode.acc_seg: 99.7263 aux.loss_ce: 0.0074 aux.acc_seg: 99.1424 +04/18 19:34:28 - mmengine - INFO - Iter(train) [111800/160000] lr: 3.4626e-03 eta: 7:24:01 time: 0.5550 data_time: 0.0061 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0066 decode.acc_seg: 99.7540 aux.loss_ce: 0.0068 aux.acc_seg: 99.3311 +04/18 19:34:56 - mmengine - INFO - Iter(train) [111850/160000] lr: 3.4594e-03 eta: 7:23:33 time: 0.5552 data_time: 0.0074 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0067 decode.acc_seg: 99.7018 aux.loss_ce: 0.0069 aux.acc_seg: 99.1357 +04/18 19:35:24 - mmengine - INFO - Iter(train) [111900/160000] lr: 3.4563e-03 eta: 7:23:06 time: 0.5551 data_time: 0.0069 memory: 7635 loss: 0.0132 decode.loss_ce: 0.0063 decode.acc_seg: 99.7384 aux.loss_ce: 0.0068 aux.acc_seg: 99.2832 +04/18 19:35:51 - mmengine - INFO - Iter(train) [111950/160000] lr: 3.4532e-03 eta: 7:22:38 time: 0.5549 data_time: 0.0073 memory: 7635 loss: 0.0132 decode.loss_ce: 0.0063 decode.acc_seg: 99.6992 aux.loss_ce: 0.0069 aux.acc_seg: 99.2122 +04/18 19:36:19 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 19:36:19 - mmengine - INFO - Iter(train) [112000/160000] lr: 3.4500e-03 eta: 7:22:10 time: 0.5561 data_time: 0.0062 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0068 decode.acc_seg: 99.7575 aux.loss_ce: 0.0069 aux.acc_seg: 99.3261 +04/18 19:36:47 - mmengine - INFO - Iter(train) [112050/160000] lr: 3.4469e-03 eta: 7:21:43 time: 0.5567 data_time: 0.0062 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0075 decode.acc_seg: 99.6931 aux.loss_ce: 0.0077 aux.acc_seg: 99.0832 +04/18 19:37:15 - mmengine - INFO - Iter(train) [112100/160000] lr: 3.4437e-03 eta: 7:21:15 time: 0.5619 data_time: 0.0068 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0067 decode.acc_seg: 99.7355 aux.loss_ce: 0.0074 aux.acc_seg: 99.2790 +04/18 19:37:43 - mmengine - INFO - Iter(train) [112150/160000] lr: 3.4406e-03 eta: 7:20:48 time: 0.5642 data_time: 0.0068 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0074 decode.acc_seg: 99.6672 aux.loss_ce: 0.0072 aux.acc_seg: 99.2986 +04/18 19:38:11 - mmengine - INFO - Iter(train) [112200/160000] lr: 3.4374e-03 eta: 7:20:20 time: 0.5545 data_time: 0.0069 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0064 decode.acc_seg: 99.7649 aux.loss_ce: 0.0069 aux.acc_seg: 99.3267 +04/18 19:38:38 - mmengine - INFO - Iter(train) [112250/160000] lr: 3.4343e-03 eta: 7:19:53 time: 0.5560 data_time: 0.0069 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0066 decode.acc_seg: 99.6828 aux.loss_ce: 0.0074 aux.acc_seg: 99.1042 +04/18 19:39:06 - mmengine - INFO - Iter(train) [112300/160000] lr: 3.4312e-03 eta: 7:19:25 time: 0.5541 data_time: 0.0060 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0071 decode.acc_seg: 99.7375 aux.loss_ce: 0.0071 aux.acc_seg: 99.4389 +04/18 19:39:34 - mmengine - INFO - Iter(train) [112350/160000] lr: 3.4280e-03 eta: 7:18:58 time: 0.5552 data_time: 0.0074 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0067 decode.acc_seg: 99.6963 aux.loss_ce: 0.0072 aux.acc_seg: 99.0375 +04/18 19:40:02 - mmengine - INFO - Iter(train) [112400/160000] lr: 3.4249e-03 eta: 7:18:30 time: 0.5565 data_time: 0.0073 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0063 decode.acc_seg: 99.7343 aux.loss_ce: 0.0070 aux.acc_seg: 99.1905 +04/18 19:40:29 - mmengine - INFO - Iter(train) [112450/160000] lr: 3.4217e-03 eta: 7:18:02 time: 0.5547 data_time: 0.0068 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0071 decode.acc_seg: 99.6407 aux.loss_ce: 0.0075 aux.acc_seg: 99.0631 +04/18 19:40:57 - mmengine - INFO - Iter(train) [112500/160000] lr: 3.4186e-03 eta: 7:17:35 time: 0.5556 data_time: 0.0067 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0067 decode.acc_seg: 99.7557 aux.loss_ce: 0.0075 aux.acc_seg: 99.4360 +04/18 19:41:25 - mmengine - INFO - Iter(train) [112550/160000] lr: 3.4154e-03 eta: 7:17:07 time: 0.5533 data_time: 0.0064 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0085 decode.acc_seg: 99.6600 aux.loss_ce: 0.0079 aux.acc_seg: 99.1341 +04/18 19:41:53 - mmengine - INFO - Iter(train) [112600/160000] lr: 3.4123e-03 eta: 7:16:40 time: 0.5545 data_time: 0.0075 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0099 decode.acc_seg: 99.6852 aux.loss_ce: 0.0078 aux.acc_seg: 99.3675 +04/18 19:42:20 - mmengine - INFO - Iter(train) [112650/160000] lr: 3.4092e-03 eta: 7:16:12 time: 0.5636 data_time: 0.0067 memory: 7635 loss: 0.0178 decode.loss_ce: 0.0094 decode.acc_seg: 99.7015 aux.loss_ce: 0.0084 aux.acc_seg: 99.2482 +04/18 19:42:48 - mmengine - INFO - Iter(train) [112700/160000] lr: 3.4060e-03 eta: 7:15:44 time: 0.5542 data_time: 0.0061 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0070 decode.acc_seg: 99.6842 aux.loss_ce: 0.0070 aux.acc_seg: 99.1373 +04/18 19:43:16 - mmengine - INFO - Iter(train) [112750/160000] lr: 3.4029e-03 eta: 7:15:17 time: 0.5547 data_time: 0.0066 memory: 7635 loss: 0.0174 decode.loss_ce: 0.0089 decode.acc_seg: 99.6509 aux.loss_ce: 0.0085 aux.acc_seg: 99.0849 +04/18 19:43:44 - mmengine - INFO - Iter(train) [112800/160000] lr: 3.3997e-03 eta: 7:14:49 time: 0.5550 data_time: 0.0075 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0072 decode.acc_seg: 99.6431 aux.loss_ce: 0.0075 aux.acc_seg: 99.1863 +04/18 19:44:12 - mmengine - INFO - Iter(train) [112850/160000] lr: 3.3966e-03 eta: 7:14:22 time: 0.5558 data_time: 0.0062 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0078 decode.acc_seg: 99.6466 aux.loss_ce: 0.0079 aux.acc_seg: 99.2352 +04/18 19:44:39 - mmengine - INFO - Iter(train) [112900/160000] lr: 3.3934e-03 eta: 7:13:54 time: 0.5548 data_time: 0.0064 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0077 decode.acc_seg: 99.7649 aux.loss_ce: 0.0077 aux.acc_seg: 99.3188 +04/18 19:45:07 - mmengine - INFO - Iter(train) [112950/160000] lr: 3.3903e-03 eta: 7:13:26 time: 0.5555 data_time: 0.0063 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0069 decode.acc_seg: 99.7283 aux.loss_ce: 0.0073 aux.acc_seg: 99.2187 +04/18 19:45:35 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 19:45:35 - mmengine - INFO - Iter(train) [113000/160000] lr: 3.3871e-03 eta: 7:12:59 time: 0.5547 data_time: 0.0071 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0079 decode.acc_seg: 99.6073 aux.loss_ce: 0.0083 aux.acc_seg: 98.9552 +04/18 19:46:03 - mmengine - INFO - Iter(train) [113050/160000] lr: 3.3840e-03 eta: 7:12:31 time: 0.5572 data_time: 0.0074 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0071 decode.acc_seg: 99.7059 aux.loss_ce: 0.0074 aux.acc_seg: 99.1577 +04/18 19:46:30 - mmengine - INFO - Iter(train) [113100/160000] lr: 3.3808e-03 eta: 7:12:04 time: 0.5546 data_time: 0.0066 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0072 decode.acc_seg: 99.7546 aux.loss_ce: 0.0078 aux.acc_seg: 99.2054 +04/18 19:46:58 - mmengine - INFO - Iter(train) [113150/160000] lr: 3.3777e-03 eta: 7:11:36 time: 0.5567 data_time: 0.0069 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0075 decode.acc_seg: 99.6556 aux.loss_ce: 0.0077 aux.acc_seg: 99.0924 +04/18 19:47:26 - mmengine - INFO - Iter(train) [113200/160000] lr: 3.3745e-03 eta: 7:11:09 time: 0.5619 data_time: 0.0062 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0071 decode.acc_seg: 99.6857 aux.loss_ce: 0.0074 aux.acc_seg: 98.9956 +04/18 19:47:54 - mmengine - INFO - Iter(train) [113250/160000] lr: 3.3714e-03 eta: 7:10:41 time: 0.5630 data_time: 0.0066 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0070 decode.acc_seg: 99.6474 aux.loss_ce: 0.0072 aux.acc_seg: 99.1713 +04/18 19:48:22 - mmengine - INFO - Iter(train) [113300/160000] lr: 3.3682e-03 eta: 7:10:13 time: 0.5558 data_time: 0.0066 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0072 decode.acc_seg: 99.7250 aux.loss_ce: 0.0073 aux.acc_seg: 99.3504 +04/18 19:48:49 - mmengine - INFO - Iter(train) [113350/160000] lr: 3.3651e-03 eta: 7:09:46 time: 0.5532 data_time: 0.0066 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0069 decode.acc_seg: 99.7818 aux.loss_ce: 0.0071 aux.acc_seg: 99.4512 +04/18 19:49:17 - mmengine - INFO - Iter(train) [113400/160000] lr: 3.3619e-03 eta: 7:09:18 time: 0.5560 data_time: 0.0072 memory: 7635 loss: 0.0157 decode.loss_ce: 0.0078 decode.acc_seg: 99.7405 aux.loss_ce: 0.0079 aux.acc_seg: 99.1410 +04/18 19:49:45 - mmengine - INFO - Iter(train) [113450/160000] lr: 3.3588e-03 eta: 7:08:51 time: 0.5569 data_time: 0.0075 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0068 decode.acc_seg: 99.6954 aux.loss_ce: 0.0075 aux.acc_seg: 99.2711 +04/18 19:50:13 - mmengine - INFO - Iter(train) [113500/160000] lr: 3.3556e-03 eta: 7:08:23 time: 0.5547 data_time: 0.0064 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0068 decode.acc_seg: 99.6919 aux.loss_ce: 0.0072 aux.acc_seg: 99.0643 +04/18 19:50:40 - mmengine - INFO - Iter(train) [113550/160000] lr: 3.3525e-03 eta: 7:07:55 time: 0.5543 data_time: 0.0065 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0068 decode.acc_seg: 99.7746 aux.loss_ce: 0.0073 aux.acc_seg: 99.2249 +04/18 19:51:08 - mmengine - INFO - Iter(train) [113600/160000] lr: 3.3493e-03 eta: 7:07:28 time: 0.5530 data_time: 0.0073 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0071 decode.acc_seg: 99.6819 aux.loss_ce: 0.0076 aux.acc_seg: 99.1260 +04/18 19:51:36 - mmengine - INFO - Iter(train) [113650/160000] lr: 3.3462e-03 eta: 7:07:00 time: 0.5534 data_time: 0.0077 memory: 7635 loss: 0.0131 decode.loss_ce: 0.0063 decode.acc_seg: 99.7600 aux.loss_ce: 0.0068 aux.acc_seg: 99.3909 +04/18 19:52:04 - mmengine - INFO - Iter(train) [113700/160000] lr: 3.3430e-03 eta: 7:06:33 time: 0.5542 data_time: 0.0060 memory: 7635 loss: 0.0170 decode.loss_ce: 0.0088 decode.acc_seg: 99.6358 aux.loss_ce: 0.0082 aux.acc_seg: 99.0391 +04/18 19:52:32 - mmengine - INFO - Iter(train) [113750/160000] lr: 3.3399e-03 eta: 7:06:05 time: 0.5547 data_time: 0.0062 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0080 decode.acc_seg: 99.6324 aux.loss_ce: 0.0081 aux.acc_seg: 99.2223 +04/18 19:52:59 - mmengine - INFO - Iter(train) [113800/160000] lr: 3.3367e-03 eta: 7:05:37 time: 0.5555 data_time: 0.0073 memory: 7635 loss: 0.0203 decode.loss_ce: 0.0109 decode.acc_seg: 99.4678 aux.loss_ce: 0.0094 aux.acc_seg: 98.9538 +04/18 19:53:27 - mmengine - INFO - Iter(train) [113850/160000] lr: 3.3336e-03 eta: 7:05:10 time: 0.5554 data_time: 0.0066 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0072 decode.acc_seg: 99.6481 aux.loss_ce: 0.0069 aux.acc_seg: 99.2805 +04/18 19:53:55 - mmengine - INFO - Iter(train) [113900/160000] lr: 3.3304e-03 eta: 7:04:42 time: 0.5567 data_time: 0.0067 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0077 decode.acc_seg: 99.7268 aux.loss_ce: 0.0076 aux.acc_seg: 99.1993 +04/18 19:54:23 - mmengine - INFO - Iter(train) [113950/160000] lr: 3.3273e-03 eta: 7:04:15 time: 0.5543 data_time: 0.0065 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0078 decode.acc_seg: 99.7354 aux.loss_ce: 0.0074 aux.acc_seg: 99.4023 +04/18 19:54:50 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 19:54:50 - mmengine - INFO - Iter(train) [114000/160000] lr: 3.3241e-03 eta: 7:03:47 time: 0.5536 data_time: 0.0064 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0075 decode.acc_seg: 99.6449 aux.loss_ce: 0.0076 aux.acc_seg: 99.0225 +04/18 19:55:18 - mmengine - INFO - Iter(train) [114050/160000] lr: 3.3210e-03 eta: 7:03:19 time: 0.5533 data_time: 0.0062 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0073 decode.acc_seg: 99.6897 aux.loss_ce: 0.0076 aux.acc_seg: 99.1895 +04/18 19:55:46 - mmengine - INFO - Iter(train) [114100/160000] lr: 3.3178e-03 eta: 7:02:52 time: 0.5558 data_time: 0.0070 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0072 decode.acc_seg: 99.6418 aux.loss_ce: 0.0074 aux.acc_seg: 99.0686 +04/18 19:56:14 - mmengine - INFO - Iter(train) [114150/160000] lr: 3.3147e-03 eta: 7:02:24 time: 0.5549 data_time: 0.0068 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0077 decode.acc_seg: 99.7098 aux.loss_ce: 0.0077 aux.acc_seg: 99.3317 +04/18 19:56:41 - mmengine - INFO - Iter(train) [114200/160000] lr: 3.3115e-03 eta: 7:01:57 time: 0.5560 data_time: 0.0065 memory: 7635 loss: 0.0177 decode.loss_ce: 0.0095 decode.acc_seg: 99.6345 aux.loss_ce: 0.0081 aux.acc_seg: 98.8980 +04/18 19:57:09 - mmengine - INFO - Iter(train) [114250/160000] lr: 3.3083e-03 eta: 7:01:29 time: 0.5647 data_time: 0.0065 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0073 decode.acc_seg: 99.7429 aux.loss_ce: 0.0075 aux.acc_seg: 99.4370 +04/18 19:57:37 - mmengine - INFO - Iter(train) [114300/160000] lr: 3.3052e-03 eta: 7:01:02 time: 0.5725 data_time: 0.0070 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0074 decode.acc_seg: 99.6933 aux.loss_ce: 0.0079 aux.acc_seg: 99.0668 +04/18 19:58:05 - mmengine - INFO - Iter(train) [114350/160000] lr: 3.3020e-03 eta: 7:00:34 time: 0.5539 data_time: 0.0066 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0069 decode.acc_seg: 99.7735 aux.loss_ce: 0.0069 aux.acc_seg: 99.4714 +04/18 19:58:33 - mmengine - INFO - Iter(train) [114400/160000] lr: 3.2989e-03 eta: 7:00:06 time: 0.5551 data_time: 0.0071 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0068 decode.acc_seg: 99.7492 aux.loss_ce: 0.0071 aux.acc_seg: 99.3660 +04/18 19:59:00 - mmengine - INFO - Iter(train) [114450/160000] lr: 3.2957e-03 eta: 6:59:39 time: 0.5549 data_time: 0.0065 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0064 decode.acc_seg: 99.7345 aux.loss_ce: 0.0073 aux.acc_seg: 99.1737 +04/18 19:59:28 - mmengine - INFO - Iter(train) [114500/160000] lr: 3.2926e-03 eta: 6:59:11 time: 0.5561 data_time: 0.0056 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0069 decode.acc_seg: 99.7748 aux.loss_ce: 0.0071 aux.acc_seg: 99.1717 +04/18 19:59:56 - mmengine - INFO - Iter(train) [114550/160000] lr: 3.2894e-03 eta: 6:58:44 time: 0.5544 data_time: 0.0068 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0064 decode.acc_seg: 99.7562 aux.loss_ce: 0.0073 aux.acc_seg: 99.1133 +04/18 20:00:24 - mmengine - INFO - Iter(train) [114600/160000] lr: 3.2863e-03 eta: 6:58:16 time: 0.5542 data_time: 0.0066 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0071 decode.acc_seg: 99.6882 aux.loss_ce: 0.0076 aux.acc_seg: 99.2651 +04/18 20:00:51 - mmengine - INFO - Iter(train) [114650/160000] lr: 3.2831e-03 eta: 6:57:48 time: 0.5543 data_time: 0.0062 memory: 7635 loss: 0.0131 decode.loss_ce: 0.0063 decode.acc_seg: 99.8008 aux.loss_ce: 0.0068 aux.acc_seg: 99.3563 +04/18 20:01:19 - mmengine - INFO - Iter(train) [114700/160000] lr: 3.2799e-03 eta: 6:57:21 time: 0.5543 data_time: 0.0067 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0068 decode.acc_seg: 99.7588 aux.loss_ce: 0.0070 aux.acc_seg: 99.4811 +04/18 20:01:47 - mmengine - INFO - Iter(train) [114750/160000] lr: 3.2768e-03 eta: 6:56:53 time: 0.5556 data_time: 0.0065 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0067 decode.acc_seg: 99.7975 aux.loss_ce: 0.0071 aux.acc_seg: 99.3069 +04/18 20:02:15 - mmengine - INFO - Iter(train) [114800/160000] lr: 3.2736e-03 eta: 6:56:26 time: 0.5538 data_time: 0.0068 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0067 decode.acc_seg: 99.7615 aux.loss_ce: 0.0073 aux.acc_seg: 99.4371 +04/18 20:02:43 - mmengine - INFO - Iter(train) [114850/160000] lr: 3.2705e-03 eta: 6:55:58 time: 0.5564 data_time: 0.0064 memory: 7635 loss: 0.0129 decode.loss_ce: 0.0061 decode.acc_seg: 99.7408 aux.loss_ce: 0.0067 aux.acc_seg: 99.3417 +04/18 20:03:10 - mmengine - INFO - Iter(train) [114900/160000] lr: 3.2673e-03 eta: 6:55:30 time: 0.5568 data_time: 0.0072 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0071 decode.acc_seg: 99.6960 aux.loss_ce: 0.0075 aux.acc_seg: 99.2154 +04/18 20:03:38 - mmengine - INFO - Iter(train) [114950/160000] lr: 3.2641e-03 eta: 6:55:03 time: 0.5526 data_time: 0.0066 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0067 decode.acc_seg: 99.6946 aux.loss_ce: 0.0070 aux.acc_seg: 99.2776 +04/18 20:04:06 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 20:04:06 - mmengine - INFO - Iter(train) [115000/160000] lr: 3.2610e-03 eta: 6:54:35 time: 0.5558 data_time: 0.0075 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0065 decode.acc_seg: 99.7103 aux.loss_ce: 0.0072 aux.acc_seg: 99.0646 +04/18 20:04:34 - mmengine - INFO - Iter(train) [115050/160000] lr: 3.2578e-03 eta: 6:54:08 time: 0.5538 data_time: 0.0065 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0068 decode.acc_seg: 99.6477 aux.loss_ce: 0.0072 aux.acc_seg: 99.2191 +04/18 20:05:01 - mmengine - INFO - Iter(train) [115100/160000] lr: 3.2547e-03 eta: 6:53:40 time: 0.5549 data_time: 0.0066 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0067 decode.acc_seg: 99.7582 aux.loss_ce: 0.0073 aux.acc_seg: 99.3569 +04/18 20:05:29 - mmengine - INFO - Iter(train) [115150/160000] lr: 3.2515e-03 eta: 6:53:13 time: 0.5551 data_time: 0.0068 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0068 decode.acc_seg: 99.7608 aux.loss_ce: 0.0072 aux.acc_seg: 99.2516 +04/18 20:05:57 - mmengine - INFO - Iter(train) [115200/160000] lr: 3.2483e-03 eta: 6:52:45 time: 0.5542 data_time: 0.0071 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0070 decode.acc_seg: 99.6777 aux.loss_ce: 0.0076 aux.acc_seg: 99.1147 +04/18 20:06:25 - mmengine - INFO - Iter(train) [115250/160000] lr: 3.2452e-03 eta: 6:52:17 time: 0.5561 data_time: 0.0068 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0068 decode.acc_seg: 99.7757 aux.loss_ce: 0.0074 aux.acc_seg: 99.3195 +04/18 20:06:52 - mmengine - INFO - Iter(train) [115300/160000] lr: 3.2420e-03 eta: 6:51:50 time: 0.5568 data_time: 0.0077 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0069 decode.acc_seg: 99.7368 aux.loss_ce: 0.0073 aux.acc_seg: 99.2317 +04/18 20:07:20 - mmengine - INFO - Iter(train) [115350/160000] lr: 3.2388e-03 eta: 6:51:22 time: 0.5553 data_time: 0.0068 memory: 7635 loss: 0.0132 decode.loss_ce: 0.0064 decode.acc_seg: 99.7233 aux.loss_ce: 0.0068 aux.acc_seg: 99.2118 +04/18 20:07:48 - mmengine - INFO - Iter(train) [115400/160000] lr: 3.2357e-03 eta: 6:50:55 time: 0.5643 data_time: 0.0069 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0068 decode.acc_seg: 99.7232 aux.loss_ce: 0.0071 aux.acc_seg: 99.2002 +04/18 20:08:16 - mmengine - INFO - Iter(train) [115450/160000] lr: 3.2325e-03 eta: 6:50:27 time: 0.5546 data_time: 0.0077 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0063 decode.acc_seg: 99.7201 aux.loss_ce: 0.0071 aux.acc_seg: 99.2654 +04/18 20:08:44 - mmengine - INFO - Iter(train) [115500/160000] lr: 3.2293e-03 eta: 6:49:59 time: 0.5553 data_time: 0.0068 memory: 7635 loss: 0.0131 decode.loss_ce: 0.0064 decode.acc_seg: 99.7291 aux.loss_ce: 0.0067 aux.acc_seg: 99.1709 +04/18 20:09:12 - mmengine - INFO - Iter(train) [115550/160000] lr: 3.2262e-03 eta: 6:49:32 time: 0.5546 data_time: 0.0062 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0077 decode.acc_seg: 99.6208 aux.loss_ce: 0.0071 aux.acc_seg: 99.1677 +04/18 20:09:39 - mmengine - INFO - Iter(train) [115600/160000] lr: 3.2230e-03 eta: 6:49:04 time: 0.5543 data_time: 0.0065 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0070 decode.acc_seg: 99.6173 aux.loss_ce: 0.0076 aux.acc_seg: 99.1159 +04/18 20:10:07 - mmengine - INFO - Iter(train) [115650/160000] lr: 3.2199e-03 eta: 6:48:37 time: 0.5543 data_time: 0.0065 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0066 decode.acc_seg: 99.7155 aux.loss_ce: 0.0069 aux.acc_seg: 99.2580 +04/18 20:10:35 - mmengine - INFO - Iter(train) [115700/160000] lr: 3.2167e-03 eta: 6:48:09 time: 0.5548 data_time: 0.0064 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0065 decode.acc_seg: 99.7417 aux.loss_ce: 0.0069 aux.acc_seg: 99.3217 +04/18 20:11:03 - mmengine - INFO - Iter(train) [115750/160000] lr: 3.2135e-03 eta: 6:47:41 time: 0.5552 data_time: 0.0071 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0067 decode.acc_seg: 99.7369 aux.loss_ce: 0.0072 aux.acc_seg: 99.1931 +04/18 20:11:30 - mmengine - INFO - Iter(train) [115800/160000] lr: 3.2104e-03 eta: 6:47:14 time: 0.5556 data_time: 0.0068 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0065 decode.acc_seg: 99.7627 aux.loss_ce: 0.0070 aux.acc_seg: 99.3275 +04/18 20:11:58 - mmengine - INFO - Iter(train) [115850/160000] lr: 3.2072e-03 eta: 6:46:46 time: 0.5549 data_time: 0.0072 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0067 decode.acc_seg: 99.7222 aux.loss_ce: 0.0071 aux.acc_seg: 99.2514 +04/18 20:12:26 - mmengine - INFO - Iter(train) [115900/160000] lr: 3.2040e-03 eta: 6:46:19 time: 0.5558 data_time: 0.0067 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0066 decode.acc_seg: 99.7405 aux.loss_ce: 0.0072 aux.acc_seg: 99.2482 +04/18 20:12:54 - mmengine - INFO - Iter(train) [115950/160000] lr: 3.2009e-03 eta: 6:45:51 time: 0.5549 data_time: 0.0065 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0071 decode.acc_seg: 99.7407 aux.loss_ce: 0.0076 aux.acc_seg: 99.2984 +04/18 20:13:22 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 20:13:22 - mmengine - INFO - Iter(train) [116000/160000] lr: 3.1977e-03 eta: 6:45:24 time: 0.5552 data_time: 0.0065 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0067 decode.acc_seg: 99.6977 aux.loss_ce: 0.0072 aux.acc_seg: 99.2315 +04/18 20:13:49 - mmengine - INFO - Iter(train) [116050/160000] lr: 3.1945e-03 eta: 6:44:56 time: 0.5551 data_time: 0.0061 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0065 decode.acc_seg: 99.8324 aux.loss_ce: 0.0070 aux.acc_seg: 99.5077 +04/18 20:14:17 - mmengine - INFO - Iter(train) [116100/160000] lr: 3.1913e-03 eta: 6:44:28 time: 0.5574 data_time: 0.0070 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0065 decode.acc_seg: 99.7281 aux.loss_ce: 0.0071 aux.acc_seg: 99.3584 +04/18 20:14:45 - mmengine - INFO - Iter(train) [116150/160000] lr: 3.1882e-03 eta: 6:44:01 time: 0.5553 data_time: 0.0066 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0072 decode.acc_seg: 99.7532 aux.loss_ce: 0.0073 aux.acc_seg: 99.2749 +04/18 20:15:13 - mmengine - INFO - Iter(train) [116200/160000] lr: 3.1850e-03 eta: 6:43:33 time: 0.5565 data_time: 0.0071 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0063 decode.acc_seg: 99.7695 aux.loss_ce: 0.0070 aux.acc_seg: 99.1457 +04/18 20:15:41 - mmengine - INFO - Iter(train) [116250/160000] lr: 3.1818e-03 eta: 6:43:06 time: 0.5553 data_time: 0.0076 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0069 decode.acc_seg: 99.7055 aux.loss_ce: 0.0077 aux.acc_seg: 99.0957 +04/18 20:16:08 - mmengine - INFO - Iter(train) [116300/160000] lr: 3.1787e-03 eta: 6:42:38 time: 0.5552 data_time: 0.0060 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0067 decode.acc_seg: 99.7753 aux.loss_ce: 0.0076 aux.acc_seg: 99.3449 +04/18 20:16:36 - mmengine - INFO - Iter(train) [116350/160000] lr: 3.1755e-03 eta: 6:42:10 time: 0.5556 data_time: 0.0073 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0071 decode.acc_seg: 99.7267 aux.loss_ce: 0.0076 aux.acc_seg: 99.2901 +04/18 20:17:04 - mmengine - INFO - Iter(train) [116400/160000] lr: 3.1723e-03 eta: 6:41:43 time: 0.5553 data_time: 0.0069 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0069 decode.acc_seg: 99.7842 aux.loss_ce: 0.0074 aux.acc_seg: 99.4181 +04/18 20:17:32 - mmengine - INFO - Iter(train) [116450/160000] lr: 3.1692e-03 eta: 6:41:15 time: 0.5569 data_time: 0.0067 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0062 decode.acc_seg: 99.7127 aux.loss_ce: 0.0068 aux.acc_seg: 99.2112 +04/18 20:18:00 - mmengine - INFO - Iter(train) [116500/160000] lr: 3.1660e-03 eta: 6:40:48 time: 0.5567 data_time: 0.0070 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0064 decode.acc_seg: 99.7738 aux.loss_ce: 0.0070 aux.acc_seg: 99.2536 +04/18 20:18:28 - mmengine - INFO - Iter(train) [116550/160000] lr: 3.1628e-03 eta: 6:40:20 time: 0.5561 data_time: 0.0078 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0068 decode.acc_seg: 99.7598 aux.loss_ce: 0.0074 aux.acc_seg: 99.3213 +04/18 20:18:55 - mmengine - INFO - Iter(train) [116600/160000] lr: 3.1596e-03 eta: 6:39:53 time: 0.5547 data_time: 0.0065 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0063 decode.acc_seg: 99.7788 aux.loss_ce: 0.0071 aux.acc_seg: 99.2798 +04/18 20:19:23 - mmengine - INFO - Iter(train) [116650/160000] lr: 3.1565e-03 eta: 6:39:25 time: 0.5531 data_time: 0.0065 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0068 decode.acc_seg: 99.7314 aux.loss_ce: 0.0073 aux.acc_seg: 99.2580 +04/18 20:19:51 - mmengine - INFO - Iter(train) [116700/160000] lr: 3.1533e-03 eta: 6:38:57 time: 0.5554 data_time: 0.0070 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0066 decode.acc_seg: 99.7270 aux.loss_ce: 0.0072 aux.acc_seg: 99.2430 +04/18 20:20:19 - mmengine - INFO - Iter(train) [116750/160000] lr: 3.1501e-03 eta: 6:38:30 time: 0.5544 data_time: 0.0074 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0071 decode.acc_seg: 99.7726 aux.loss_ce: 0.0073 aux.acc_seg: 99.3933 +04/18 20:20:46 - mmengine - INFO - Iter(train) [116800/160000] lr: 3.1469e-03 eta: 6:38:02 time: 0.5545 data_time: 0.0069 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0063 decode.acc_seg: 99.7207 aux.loss_ce: 0.0071 aux.acc_seg: 99.1790 +04/18 20:21:14 - mmengine - INFO - Iter(train) [116850/160000] lr: 3.1438e-03 eta: 6:37:35 time: 0.5552 data_time: 0.0065 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0065 decode.acc_seg: 99.6984 aux.loss_ce: 0.0071 aux.acc_seg: 99.2356 +04/18 20:21:42 - mmengine - INFO - Iter(train) [116900/160000] lr: 3.1406e-03 eta: 6:37:07 time: 0.5557 data_time: 0.0065 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0065 decode.acc_seg: 99.7245 aux.loss_ce: 0.0071 aux.acc_seg: 99.2350 +04/18 20:22:10 - mmengine - INFO - Iter(train) [116950/160000] lr: 3.1374e-03 eta: 6:36:39 time: 0.5650 data_time: 0.0066 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0066 decode.acc_seg: 99.7678 aux.loss_ce: 0.0074 aux.acc_seg: 99.3468 +04/18 20:22:38 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 20:22:38 - mmengine - INFO - Iter(train) [117000/160000] lr: 3.1342e-03 eta: 6:36:12 time: 0.5557 data_time: 0.0061 memory: 7635 loss: 0.0128 decode.loss_ce: 0.0060 decode.acc_seg: 99.7696 aux.loss_ce: 0.0068 aux.acc_seg: 99.4011 +04/18 20:23:05 - mmengine - INFO - Iter(train) [117050/160000] lr: 3.1311e-03 eta: 6:35:44 time: 0.5540 data_time: 0.0070 memory: 7635 loss: 0.0123 decode.loss_ce: 0.0059 decode.acc_seg: 99.8018 aux.loss_ce: 0.0064 aux.acc_seg: 99.4440 +04/18 20:23:33 - mmengine - INFO - Iter(train) [117100/160000] lr: 3.1279e-03 eta: 6:35:17 time: 0.5555 data_time: 0.0082 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0066 decode.acc_seg: 99.7345 aux.loss_ce: 0.0074 aux.acc_seg: 99.0630 +04/18 20:24:01 - mmengine - INFO - Iter(train) [117150/160000] lr: 3.1247e-03 eta: 6:34:49 time: 0.5568 data_time: 0.0067 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0067 decode.acc_seg: 99.7351 aux.loss_ce: 0.0072 aux.acc_seg: 99.3247 +04/18 20:24:29 - mmengine - INFO - Iter(train) [117200/160000] lr: 3.1215e-03 eta: 6:34:21 time: 0.5551 data_time: 0.0070 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0066 decode.acc_seg: 99.7007 aux.loss_ce: 0.0073 aux.acc_seg: 99.0025 +04/18 20:24:56 - mmengine - INFO - Iter(train) [117250/160000] lr: 3.1184e-03 eta: 6:33:54 time: 0.5574 data_time: 0.0077 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0064 decode.acc_seg: 99.6803 aux.loss_ce: 0.0066 aux.acc_seg: 99.1085 +04/18 20:25:24 - mmengine - INFO - Iter(train) [117300/160000] lr: 3.1152e-03 eta: 6:33:26 time: 0.5548 data_time: 0.0070 memory: 7635 loss: 0.0128 decode.loss_ce: 0.0060 decode.acc_seg: 99.7799 aux.loss_ce: 0.0068 aux.acc_seg: 99.2558 +04/18 20:25:52 - mmengine - INFO - Iter(train) [117350/160000] lr: 3.1120e-03 eta: 6:32:59 time: 0.5543 data_time: 0.0066 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0065 decode.acc_seg: 99.6716 aux.loss_ce: 0.0072 aux.acc_seg: 99.1094 +04/18 20:26:20 - mmengine - INFO - Iter(train) [117400/160000] lr: 3.1088e-03 eta: 6:32:31 time: 0.5554 data_time: 0.0065 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0072 decode.acc_seg: 99.7077 aux.loss_ce: 0.0075 aux.acc_seg: 99.0832 +04/18 20:26:48 - mmengine - INFO - Iter(train) [117450/160000] lr: 3.1057e-03 eta: 6:32:03 time: 0.5553 data_time: 0.0079 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0065 decode.acc_seg: 99.7317 aux.loss_ce: 0.0075 aux.acc_seg: 99.3379 +04/18 20:27:15 - mmengine - INFO - Iter(train) [117500/160000] lr: 3.1025e-03 eta: 6:31:36 time: 0.5551 data_time: 0.0071 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0066 decode.acc_seg: 99.7749 aux.loss_ce: 0.0070 aux.acc_seg: 99.2544 +04/18 20:27:43 - mmengine - INFO - Iter(train) [117550/160000] lr: 3.0993e-03 eta: 6:31:08 time: 0.5541 data_time: 0.0067 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0069 decode.acc_seg: 99.7262 aux.loss_ce: 0.0073 aux.acc_seg: 99.3453 +04/18 20:28:11 - mmengine - INFO - Iter(train) [117600/160000] lr: 3.0961e-03 eta: 6:30:41 time: 0.5570 data_time: 0.0070 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0065 decode.acc_seg: 99.7079 aux.loss_ce: 0.0073 aux.acc_seg: 99.0181 +04/18 20:28:39 - mmengine - INFO - Iter(train) [117650/160000] lr: 3.0929e-03 eta: 6:30:13 time: 0.5541 data_time: 0.0063 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0066 decode.acc_seg: 99.7430 aux.loss_ce: 0.0072 aux.acc_seg: 99.2660 +04/18 20:29:07 - mmengine - INFO - Iter(train) [117700/160000] lr: 3.0898e-03 eta: 6:29:45 time: 0.5548 data_time: 0.0072 memory: 7635 loss: 0.0127 decode.loss_ce: 0.0059 decode.acc_seg: 99.7495 aux.loss_ce: 0.0068 aux.acc_seg: 99.3814 +04/18 20:29:34 - mmengine - INFO - Iter(train) [117750/160000] lr: 3.0866e-03 eta: 6:29:18 time: 0.5551 data_time: 0.0067 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0065 decode.acc_seg: 99.7581 aux.loss_ce: 0.0074 aux.acc_seg: 99.3503 +04/18 20:30:02 - mmengine - INFO - Iter(train) [117800/160000] lr: 3.0834e-03 eta: 6:28:50 time: 0.5548 data_time: 0.0065 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0061 decode.acc_seg: 99.7300 aux.loss_ce: 0.0069 aux.acc_seg: 99.1374 +04/18 20:30:30 - mmengine - INFO - Iter(train) [117850/160000] lr: 3.0802e-03 eta: 6:28:23 time: 0.5566 data_time: 0.0060 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0066 decode.acc_seg: 99.7112 aux.loss_ce: 0.0074 aux.acc_seg: 99.1974 +04/18 20:30:58 - mmengine - INFO - Iter(train) [117900/160000] lr: 3.0770e-03 eta: 6:27:55 time: 0.5561 data_time: 0.0070 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0072 decode.acc_seg: 99.7927 aux.loss_ce: 0.0077 aux.acc_seg: 99.3387 +04/18 20:31:26 - mmengine - INFO - Iter(train) [117950/160000] lr: 3.0738e-03 eta: 6:27:27 time: 0.5552 data_time: 0.0071 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0067 decode.acc_seg: 99.7211 aux.loss_ce: 0.0076 aux.acc_seg: 99.2286 +04/18 20:31:53 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 20:31:53 - mmengine - INFO - Iter(train) [118000/160000] lr: 3.0707e-03 eta: 6:27:00 time: 0.5555 data_time: 0.0066 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0064 decode.acc_seg: 99.7520 aux.loss_ce: 0.0070 aux.acc_seg: 99.3025 +04/18 20:32:21 - mmengine - INFO - Iter(train) [118050/160000] lr: 3.0675e-03 eta: 6:26:32 time: 0.5552 data_time: 0.0065 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0062 decode.acc_seg: 99.8034 aux.loss_ce: 0.0069 aux.acc_seg: 99.4008 +04/18 20:32:49 - mmengine - INFO - Iter(train) [118100/160000] lr: 3.0643e-03 eta: 6:26:05 time: 0.5573 data_time: 0.0070 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0070 decode.acc_seg: 99.7596 aux.loss_ce: 0.0077 aux.acc_seg: 99.3524 +04/18 20:33:17 - mmengine - INFO - Iter(train) [118150/160000] lr: 3.0611e-03 eta: 6:25:37 time: 0.5552 data_time: 0.0068 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0079 decode.acc_seg: 99.5859 aux.loss_ce: 0.0077 aux.acc_seg: 99.0843 +04/18 20:33:45 - mmengine - INFO - Iter(train) [118200/160000] lr: 3.0579e-03 eta: 6:25:10 time: 0.5553 data_time: 0.0061 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0079 decode.acc_seg: 99.7565 aux.loss_ce: 0.0080 aux.acc_seg: 99.3854 +04/18 20:34:12 - mmengine - INFO - Iter(train) [118250/160000] lr: 3.0547e-03 eta: 6:24:42 time: 0.5561 data_time: 0.0070 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0077 decode.acc_seg: 99.7158 aux.loss_ce: 0.0076 aux.acc_seg: 99.0129 +04/18 20:34:40 - mmengine - INFO - Iter(train) [118300/160000] lr: 3.0516e-03 eta: 6:24:14 time: 0.5558 data_time: 0.0067 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0068 decode.acc_seg: 99.6723 aux.loss_ce: 0.0074 aux.acc_seg: 99.1504 +04/18 20:35:08 - mmengine - INFO - Iter(train) [118350/160000] lr: 3.0484e-03 eta: 6:23:47 time: 0.5561 data_time: 0.0065 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0078 decode.acc_seg: 99.7306 aux.loss_ce: 0.0075 aux.acc_seg: 99.3378 +04/18 20:35:36 - mmengine - INFO - Iter(train) [118400/160000] lr: 3.0452e-03 eta: 6:23:19 time: 0.5549 data_time: 0.0064 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0075 decode.acc_seg: 99.6934 aux.loss_ce: 0.0072 aux.acc_seg: 99.1377 +04/18 20:36:04 - mmengine - INFO - Iter(train) [118450/160000] lr: 3.0420e-03 eta: 6:22:52 time: 0.5552 data_time: 0.0066 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0076 decode.acc_seg: 99.6614 aux.loss_ce: 0.0078 aux.acc_seg: 99.1085 +04/18 20:36:31 - mmengine - INFO - Iter(train) [118500/160000] lr: 3.0388e-03 eta: 6:22:24 time: 0.5557 data_time: 0.0067 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0065 decode.acc_seg: 99.7893 aux.loss_ce: 0.0070 aux.acc_seg: 99.4375 +04/18 20:36:59 - mmengine - INFO - Iter(train) [118550/160000] lr: 3.0356e-03 eta: 6:21:56 time: 0.5642 data_time: 0.0069 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0066 decode.acc_seg: 99.7399 aux.loss_ce: 0.0069 aux.acc_seg: 99.2958 +04/18 20:37:27 - mmengine - INFO - Iter(train) [118600/160000] lr: 3.0324e-03 eta: 6:21:29 time: 0.5631 data_time: 0.0070 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0064 decode.acc_seg: 99.7432 aux.loss_ce: 0.0072 aux.acc_seg: 99.3347 +04/18 20:37:55 - mmengine - INFO - Iter(train) [118650/160000] lr: 3.0293e-03 eta: 6:21:01 time: 0.5561 data_time: 0.0064 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0068 decode.acc_seg: 99.6925 aux.loss_ce: 0.0068 aux.acc_seg: 99.2470 +04/18 20:38:23 - mmengine - INFO - Iter(train) [118700/160000] lr: 3.0261e-03 eta: 6:20:34 time: 0.5544 data_time: 0.0065 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0102 decode.acc_seg: 99.7601 aux.loss_ce: 0.0081 aux.acc_seg: 99.2406 +04/18 20:38:50 - mmengine - INFO - Iter(train) [118750/160000] lr: 3.0229e-03 eta: 6:20:06 time: 0.5557 data_time: 0.0069 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0084 decode.acc_seg: 99.6594 aux.loss_ce: 0.0077 aux.acc_seg: 99.0275 +04/18 20:39:18 - mmengine - INFO - Iter(train) [118800/160000] lr: 3.0197e-03 eta: 6:19:38 time: 0.5539 data_time: 0.0068 memory: 7635 loss: 0.0158 decode.loss_ce: 0.0081 decode.acc_seg: 99.6863 aux.loss_ce: 0.0078 aux.acc_seg: 99.2388 +04/18 20:39:46 - mmengine - INFO - Iter(train) [118850/160000] lr: 3.0165e-03 eta: 6:19:11 time: 0.5559 data_time: 0.0065 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0071 decode.acc_seg: 99.7490 aux.loss_ce: 0.0073 aux.acc_seg: 99.2550 +04/18 20:40:14 - mmengine - INFO - Iter(train) [118900/160000] lr: 3.0133e-03 eta: 6:18:43 time: 0.5569 data_time: 0.0073 memory: 7635 loss: 0.0165 decode.loss_ce: 0.0084 decode.acc_seg: 99.6354 aux.loss_ce: 0.0081 aux.acc_seg: 99.0667 +04/18 20:40:42 - mmengine - INFO - Iter(train) [118950/160000] lr: 3.0101e-03 eta: 6:18:16 time: 0.5560 data_time: 0.0078 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0079 decode.acc_seg: 99.6362 aux.loss_ce: 0.0073 aux.acc_seg: 99.2110 +04/18 20:41:09 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 20:41:09 - mmengine - INFO - Iter(train) [119000/160000] lr: 3.0069e-03 eta: 6:17:48 time: 0.5538 data_time: 0.0066 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0077 decode.acc_seg: 99.7345 aux.loss_ce: 0.0071 aux.acc_seg: 99.3922 +04/18 20:41:37 - mmengine - INFO - Iter(train) [119050/160000] lr: 3.0037e-03 eta: 6:17:20 time: 0.5547 data_time: 0.0065 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0066 decode.acc_seg: 99.7228 aux.loss_ce: 0.0074 aux.acc_seg: 99.0550 +04/18 20:42:05 - mmengine - INFO - Iter(train) [119100/160000] lr: 3.0006e-03 eta: 6:16:53 time: 0.5646 data_time: 0.0069 memory: 7635 loss: 0.0211 decode.loss_ce: 0.0119 decode.acc_seg: 99.7412 aux.loss_ce: 0.0092 aux.acc_seg: 99.2513 +04/18 20:42:33 - mmengine - INFO - Iter(train) [119150/160000] lr: 2.9974e-03 eta: 6:16:25 time: 0.5574 data_time: 0.0068 memory: 7635 loss: 0.0184 decode.loss_ce: 0.0097 decode.acc_seg: 99.7432 aux.loss_ce: 0.0087 aux.acc_seg: 99.3627 +04/18 20:43:01 - mmengine - INFO - Iter(train) [119200/160000] lr: 2.9942e-03 eta: 6:15:58 time: 0.5562 data_time: 0.0071 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0071 decode.acc_seg: 99.6814 aux.loss_ce: 0.0071 aux.acc_seg: 99.1776 +04/18 20:43:28 - mmengine - INFO - Iter(train) [119250/160000] lr: 2.9910e-03 eta: 6:15:30 time: 0.5549 data_time: 0.0064 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0072 decode.acc_seg: 99.6810 aux.loss_ce: 0.0072 aux.acc_seg: 99.3422 +04/18 20:43:56 - mmengine - INFO - Iter(train) [119300/160000] lr: 2.9878e-03 eta: 6:15:02 time: 0.5546 data_time: 0.0070 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0074 decode.acc_seg: 99.6921 aux.loss_ce: 0.0077 aux.acc_seg: 99.2339 +04/18 20:44:24 - mmengine - INFO - Iter(train) [119350/160000] lr: 2.9846e-03 eta: 6:14:35 time: 0.5541 data_time: 0.0062 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0070 decode.acc_seg: 99.6759 aux.loss_ce: 0.0076 aux.acc_seg: 99.1552 +04/18 20:44:52 - mmengine - INFO - Iter(train) [119400/160000] lr: 2.9814e-03 eta: 6:14:07 time: 0.5553 data_time: 0.0065 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0078 decode.acc_seg: 99.7362 aux.loss_ce: 0.0082 aux.acc_seg: 99.2253 +04/18 20:45:20 - mmengine - INFO - Iter(train) [119450/160000] lr: 2.9782e-03 eta: 6:13:40 time: 0.5554 data_time: 0.0071 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0068 decode.acc_seg: 99.6972 aux.loss_ce: 0.0071 aux.acc_seg: 99.1246 +04/18 20:45:47 - mmengine - INFO - Iter(train) [119500/160000] lr: 2.9750e-03 eta: 6:13:12 time: 0.5548 data_time: 0.0079 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0069 decode.acc_seg: 99.7215 aux.loss_ce: 0.0071 aux.acc_seg: 99.3957 +04/18 20:46:15 - mmengine - INFO - Iter(train) [119550/160000] lr: 2.9718e-03 eta: 6:12:44 time: 0.5565 data_time: 0.0076 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0077 decode.acc_seg: 99.6366 aux.loss_ce: 0.0073 aux.acc_seg: 99.0820 +04/18 20:46:43 - mmengine - INFO - Iter(train) [119600/160000] lr: 2.9686e-03 eta: 6:12:17 time: 0.5543 data_time: 0.0066 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0068 decode.acc_seg: 99.7691 aux.loss_ce: 0.0069 aux.acc_seg: 99.2675 +04/18 20:47:11 - mmengine - INFO - Iter(train) [119650/160000] lr: 2.9654e-03 eta: 6:11:49 time: 0.5635 data_time: 0.0061 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0073 decode.acc_seg: 99.6838 aux.loss_ce: 0.0072 aux.acc_seg: 99.0096 +04/18 20:47:39 - mmengine - INFO - Iter(train) [119700/160000] lr: 2.9622e-03 eta: 6:11:22 time: 0.5639 data_time: 0.0070 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0069 decode.acc_seg: 99.7218 aux.loss_ce: 0.0077 aux.acc_seg: 99.3238 +04/18 20:48:07 - mmengine - INFO - Iter(train) [119750/160000] lr: 2.9590e-03 eta: 6:10:54 time: 0.5542 data_time: 0.0069 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0070 decode.acc_seg: 99.6610 aux.loss_ce: 0.0076 aux.acc_seg: 99.1394 +04/18 20:48:34 - mmengine - INFO - Iter(train) [119800/160000] lr: 2.9558e-03 eta: 6:10:26 time: 0.5548 data_time: 0.0075 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0069 decode.acc_seg: 99.6144 aux.loss_ce: 0.0074 aux.acc_seg: 98.9707 +04/18 20:49:02 - mmengine - INFO - Iter(train) [119850/160000] lr: 2.9526e-03 eta: 6:09:59 time: 0.5546 data_time: 0.0062 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0076 decode.acc_seg: 99.6242 aux.loss_ce: 0.0080 aux.acc_seg: 98.6984 +04/18 20:49:30 - mmengine - INFO - Iter(train) [119900/160000] lr: 2.9494e-03 eta: 6:09:31 time: 0.5547 data_time: 0.0069 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0069 decode.acc_seg: 99.7492 aux.loss_ce: 0.0072 aux.acc_seg: 99.2558 +04/18 20:49:58 - mmengine - INFO - Iter(train) [119950/160000] lr: 2.9462e-03 eta: 6:09:04 time: 0.5544 data_time: 0.0065 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0065 decode.acc_seg: 99.6436 aux.loss_ce: 0.0071 aux.acc_seg: 99.0393 +04/18 20:50:25 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 20:50:25 - mmengine - INFO - Iter(train) [120000/160000] lr: 2.9430e-03 eta: 6:08:36 time: 0.5553 data_time: 0.0068 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0072 decode.acc_seg: 99.7734 aux.loss_ce: 0.0074 aux.acc_seg: 99.3732 +04/18 20:50:25 - mmengine - INFO - Saving checkpoint at 120000 iterations +04/18 20:50:29 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:07 time: 0.0463 data_time: 0.0014 memory: 1657 +04/18 20:50:32 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:04 time: 0.0465 data_time: 0.0016 memory: 1657 +04/18 20:50:34 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:02 time: 0.0462 data_time: 0.0014 memory: 1657 +04/18 20:50:36 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.0457 data_time: 0.0013 memory: 1657 +04/18 20:50:37 - mmengine - INFO - per class results: +04/18 20:50:37 - mmengine - INFO - ++------------+-------+------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+------+--------+-----------+--------+ +| background | 99.09 | 99.5 | 99.54 | 99.58 | 99.5 | +| contrast | 80.39 | 90.0 | 89.13 | 88.28 | 90.0 | ++------------+-------+------+--------+-----------+--------+ +04/18 20:50:37 - mmengine - INFO - Iter(val) [200/200] aAcc: 99.1200 mIoU: 89.7400 mAcc: 94.7500 mFscore: 94.3400 mPrecision: 93.9300 mRecall: 94.7500 data_time: 0.0015 time: 0.0466 +04/18 20:51:04 - mmengine - INFO - Iter(train) [120050/160000] lr: 2.9398e-03 eta: 6:08:08 time: 0.5548 data_time: 0.0065 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0073 decode.acc_seg: 99.7140 aux.loss_ce: 0.0076 aux.acc_seg: 99.1138 +04/18 20:51:32 - mmengine - INFO - Iter(train) [120100/160000] lr: 2.9366e-03 eta: 6:07:41 time: 0.5565 data_time: 0.0075 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0067 decode.acc_seg: 99.6864 aux.loss_ce: 0.0074 aux.acc_seg: 99.1917 +04/18 20:52:00 - mmengine - INFO - Iter(train) [120150/160000] lr: 2.9334e-03 eta: 6:07:13 time: 0.5557 data_time: 0.0067 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0067 decode.acc_seg: 99.7345 aux.loss_ce: 0.0074 aux.acc_seg: 99.0086 +04/18 20:52:28 - mmengine - INFO - Iter(train) [120200/160000] lr: 2.9302e-03 eta: 6:06:46 time: 0.5563 data_time: 0.0072 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0064 decode.acc_seg: 99.6479 aux.loss_ce: 0.0073 aux.acc_seg: 98.8611 +04/18 20:52:55 - mmengine - INFO - Iter(train) [120250/160000] lr: 2.9270e-03 eta: 6:06:18 time: 0.5556 data_time: 0.0075 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0067 decode.acc_seg: 99.7431 aux.loss_ce: 0.0073 aux.acc_seg: 99.3074 +04/18 20:53:23 - mmengine - INFO - Iter(train) [120300/160000] lr: 2.9238e-03 eta: 6:05:50 time: 0.5546 data_time: 0.0063 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0068 decode.acc_seg: 99.7306 aux.loss_ce: 0.0072 aux.acc_seg: 99.1041 +04/18 20:53:51 - mmengine - INFO - Iter(train) [120350/160000] lr: 2.9206e-03 eta: 6:05:23 time: 0.5562 data_time: 0.0065 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0069 decode.acc_seg: 99.7080 aux.loss_ce: 0.0072 aux.acc_seg: 99.1747 +04/18 20:54:19 - mmengine - INFO - Iter(train) [120400/160000] lr: 2.9174e-03 eta: 6:04:55 time: 0.5540 data_time: 0.0061 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0064 decode.acc_seg: 99.7389 aux.loss_ce: 0.0071 aux.acc_seg: 99.1768 +04/18 20:54:46 - mmengine - INFO - Iter(train) [120450/160000] lr: 2.9142e-03 eta: 6:04:28 time: 0.5533 data_time: 0.0069 memory: 7635 loss: 0.0131 decode.loss_ce: 0.0062 decode.acc_seg: 99.7629 aux.loss_ce: 0.0068 aux.acc_seg: 99.4340 +04/18 20:55:14 - mmengine - INFO - Iter(train) [120500/160000] lr: 2.9110e-03 eta: 6:04:00 time: 0.5652 data_time: 0.0068 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0065 decode.acc_seg: 99.7408 aux.loss_ce: 0.0072 aux.acc_seg: 99.2261 +04/18 20:55:42 - mmengine - INFO - Iter(train) [120550/160000] lr: 2.9078e-03 eta: 6:03:32 time: 0.5540 data_time: 0.0067 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0073 decode.acc_seg: 99.7856 aux.loss_ce: 0.0071 aux.acc_seg: 99.4425 +04/18 20:56:10 - mmengine - INFO - Iter(train) [120600/160000] lr: 2.9046e-03 eta: 6:03:05 time: 0.5525 data_time: 0.0062 memory: 7635 loss: 0.0129 decode.loss_ce: 0.0062 decode.acc_seg: 99.6795 aux.loss_ce: 0.0068 aux.acc_seg: 98.9835 +04/18 20:56:38 - mmengine - INFO - Iter(train) [120650/160000] lr: 2.9014e-03 eta: 6:02:37 time: 0.5562 data_time: 0.0064 memory: 7635 loss: 0.0127 decode.loss_ce: 0.0060 decode.acc_seg: 99.7681 aux.loss_ce: 0.0067 aux.acc_seg: 99.3935 +04/18 20:57:06 - mmengine - INFO - Iter(train) [120700/160000] lr: 2.8982e-03 eta: 6:02:10 time: 0.5553 data_time: 0.0064 memory: 7635 loss: 0.0129 decode.loss_ce: 0.0061 decode.acc_seg: 99.7891 aux.loss_ce: 0.0068 aux.acc_seg: 99.2044 +04/18 20:57:33 - mmengine - INFO - Iter(train) [120750/160000] lr: 2.8950e-03 eta: 6:01:42 time: 0.5548 data_time: 0.0070 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0072 decode.acc_seg: 99.7631 aux.loss_ce: 0.0078 aux.acc_seg: 99.3780 +04/18 20:58:01 - mmengine - INFO - Iter(train) [120800/160000] lr: 2.8918e-03 eta: 6:01:14 time: 0.5540 data_time: 0.0077 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0070 decode.acc_seg: 99.7164 aux.loss_ce: 0.0070 aux.acc_seg: 99.3695 +04/18 20:58:29 - mmengine - INFO - Iter(train) [120850/160000] lr: 2.8886e-03 eta: 6:00:47 time: 0.5545 data_time: 0.0072 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0072 decode.acc_seg: 99.6573 aux.loss_ce: 0.0076 aux.acc_seg: 98.8897 +04/18 20:58:57 - mmengine - INFO - Iter(train) [120900/160000] lr: 2.8854e-03 eta: 6:00:19 time: 0.5555 data_time: 0.0079 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0071 decode.acc_seg: 99.7353 aux.loss_ce: 0.0075 aux.acc_seg: 99.3001 +04/18 20:59:25 - mmengine - INFO - Iter(train) [120950/160000] lr: 2.8822e-03 eta: 5:59:52 time: 0.5547 data_time: 0.0068 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0065 decode.acc_seg: 99.7063 aux.loss_ce: 0.0071 aux.acc_seg: 99.1960 +04/18 20:59:52 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 20:59:52 - mmengine - INFO - Iter(train) [121000/160000] lr: 2.8790e-03 eta: 5:59:24 time: 0.5553 data_time: 0.0070 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0067 decode.acc_seg: 99.7291 aux.loss_ce: 0.0071 aux.acc_seg: 99.1344 +04/18 21:00:20 - mmengine - INFO - Iter(train) [121050/160000] lr: 2.8758e-03 eta: 5:58:56 time: 0.5549 data_time: 0.0067 memory: 7635 loss: 0.0128 decode.loss_ce: 0.0062 decode.acc_seg: 99.6985 aux.loss_ce: 0.0066 aux.acc_seg: 99.1734 +04/18 21:00:48 - mmengine - INFO - Iter(train) [121100/160000] lr: 2.8726e-03 eta: 5:58:29 time: 0.5558 data_time: 0.0070 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0068 decode.acc_seg: 99.6827 aux.loss_ce: 0.0072 aux.acc_seg: 99.1752 +04/18 21:01:16 - mmengine - INFO - Iter(train) [121150/160000] lr: 2.8694e-03 eta: 5:58:01 time: 0.5560 data_time: 0.0067 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0071 decode.acc_seg: 99.7724 aux.loss_ce: 0.0074 aux.acc_seg: 99.2701 +04/18 21:01:43 - mmengine - INFO - Iter(train) [121200/160000] lr: 2.8662e-03 eta: 5:57:34 time: 0.5554 data_time: 0.0068 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0063 decode.acc_seg: 99.7206 aux.loss_ce: 0.0071 aux.acc_seg: 99.1982 +04/18 21:02:11 - mmengine - INFO - Iter(train) [121250/160000] lr: 2.8630e-03 eta: 5:57:06 time: 0.5551 data_time: 0.0069 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0068 decode.acc_seg: 99.6980 aux.loss_ce: 0.0070 aux.acc_seg: 99.3225 +04/18 21:02:39 - mmengine - INFO - Iter(train) [121300/160000] lr: 2.8597e-03 eta: 5:56:38 time: 0.5537 data_time: 0.0067 memory: 7635 loss: 0.0213 decode.loss_ce: 0.0119 decode.acc_seg: 99.4873 aux.loss_ce: 0.0094 aux.acc_seg: 99.1308 +04/18 21:03:07 - mmengine - INFO - Iter(train) [121350/160000] lr: 2.8565e-03 eta: 5:56:11 time: 0.5537 data_time: 0.0061 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0086 decode.acc_seg: 99.5501 aux.loss_ce: 0.0082 aux.acc_seg: 98.8340 +04/18 21:03:34 - mmengine - INFO - Iter(train) [121400/160000] lr: 2.8533e-03 eta: 5:55:43 time: 0.5541 data_time: 0.0063 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0070 decode.acc_seg: 99.7892 aux.loss_ce: 0.0073 aux.acc_seg: 99.3712 +04/18 21:04:02 - mmengine - INFO - Iter(train) [121450/160000] lr: 2.8501e-03 eta: 5:55:15 time: 0.5543 data_time: 0.0067 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0081 decode.acc_seg: 99.6509 aux.loss_ce: 0.0081 aux.acc_seg: 98.9856 +04/18 21:04:30 - mmengine - INFO - Iter(train) [121500/160000] lr: 2.8469e-03 eta: 5:54:48 time: 0.5564 data_time: 0.0072 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0080 decode.acc_seg: 99.7168 aux.loss_ce: 0.0080 aux.acc_seg: 99.2508 +04/18 21:04:58 - mmengine - INFO - Iter(train) [121550/160000] lr: 2.8437e-03 eta: 5:54:20 time: 0.5551 data_time: 0.0081 memory: 7635 loss: 0.0168 decode.loss_ce: 0.0087 decode.acc_seg: 99.6490 aux.loss_ce: 0.0081 aux.acc_seg: 99.1926 +04/18 21:05:26 - mmengine - INFO - Iter(train) [121600/160000] lr: 2.8405e-03 eta: 5:53:53 time: 0.5552 data_time: 0.0062 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0076 decode.acc_seg: 99.6541 aux.loss_ce: 0.0080 aux.acc_seg: 99.0889 +04/18 21:05:53 - mmengine - INFO - Iter(train) [121650/160000] lr: 2.8373e-03 eta: 5:53:25 time: 0.5557 data_time: 0.0071 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0084 decode.acc_seg: 99.6855 aux.loss_ce: 0.0078 aux.acc_seg: 99.1094 +04/18 21:06:21 - mmengine - INFO - Iter(train) [121700/160000] lr: 2.8341e-03 eta: 5:52:57 time: 0.5566 data_time: 0.0066 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0068 decode.acc_seg: 99.7393 aux.loss_ce: 0.0074 aux.acc_seg: 99.1827 +04/18 21:06:49 - mmengine - INFO - Iter(train) [121750/160000] lr: 2.8309e-03 eta: 5:52:30 time: 0.5551 data_time: 0.0075 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0068 decode.acc_seg: 99.7639 aux.loss_ce: 0.0071 aux.acc_seg: 99.2971 +04/18 21:07:17 - mmengine - INFO - Iter(train) [121800/160000] lr: 2.8276e-03 eta: 5:52:02 time: 0.5650 data_time: 0.0069 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0069 decode.acc_seg: 99.7086 aux.loss_ce: 0.0070 aux.acc_seg: 99.1890 +04/18 21:07:45 - mmengine - INFO - Iter(train) [121850/160000] lr: 2.8244e-03 eta: 5:51:35 time: 0.5546 data_time: 0.0065 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0073 decode.acc_seg: 99.7795 aux.loss_ce: 0.0077 aux.acc_seg: 99.3040 +04/18 21:08:12 - mmengine - INFO - Iter(train) [121900/160000] lr: 2.8212e-03 eta: 5:51:07 time: 0.5546 data_time: 0.0071 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0074 decode.acc_seg: 99.7078 aux.loss_ce: 0.0073 aux.acc_seg: 99.0706 +04/18 21:08:40 - mmengine - INFO - Iter(train) [121950/160000] lr: 2.8180e-03 eta: 5:50:39 time: 0.5542 data_time: 0.0067 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0073 decode.acc_seg: 99.7528 aux.loss_ce: 0.0077 aux.acc_seg: 99.2546 +04/18 21:09:08 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 21:09:08 - mmengine - INFO - Iter(train) [122000/160000] lr: 2.8148e-03 eta: 5:50:12 time: 0.5533 data_time: 0.0069 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0067 decode.acc_seg: 99.7618 aux.loss_ce: 0.0071 aux.acc_seg: 99.5121 +04/18 21:09:36 - mmengine - INFO - Iter(train) [122050/160000] lr: 2.8116e-03 eta: 5:49:44 time: 0.5540 data_time: 0.0065 memory: 7635 loss: 0.0129 decode.loss_ce: 0.0061 decode.acc_seg: 99.7833 aux.loss_ce: 0.0068 aux.acc_seg: 99.3933 +04/18 21:10:03 - mmengine - INFO - Iter(train) [122100/160000] lr: 2.8084e-03 eta: 5:49:17 time: 0.5534 data_time: 0.0064 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0079 decode.acc_seg: 99.6613 aux.loss_ce: 0.0075 aux.acc_seg: 99.2094 +04/18 21:10:31 - mmengine - INFO - Iter(train) [122150/160000] lr: 2.8051e-03 eta: 5:48:49 time: 0.5537 data_time: 0.0071 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0072 decode.acc_seg: 99.7586 aux.loss_ce: 0.0072 aux.acc_seg: 99.2103 +04/18 21:10:59 - mmengine - INFO - Iter(train) [122200/160000] lr: 2.8019e-03 eta: 5:48:21 time: 0.5521 data_time: 0.0064 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0070 decode.acc_seg: 99.6831 aux.loss_ce: 0.0074 aux.acc_seg: 99.1498 +04/18 21:11:27 - mmengine - INFO - Iter(train) [122250/160000] lr: 2.7987e-03 eta: 5:47:54 time: 0.5538 data_time: 0.0067 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0064 decode.acc_seg: 99.7026 aux.loss_ce: 0.0071 aux.acc_seg: 99.1042 +04/18 21:11:54 - mmengine - INFO - Iter(train) [122300/160000] lr: 2.7955e-03 eta: 5:47:26 time: 0.5542 data_time: 0.0062 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0066 decode.acc_seg: 99.7649 aux.loss_ce: 0.0071 aux.acc_seg: 99.4239 +04/18 21:12:22 - mmengine - INFO - Iter(train) [122350/160000] lr: 2.7923e-03 eta: 5:46:58 time: 0.5562 data_time: 0.0068 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0069 decode.acc_seg: 99.7177 aux.loss_ce: 0.0075 aux.acc_seg: 99.2822 +04/18 21:12:50 - mmengine - INFO - Iter(train) [122400/160000] lr: 2.7890e-03 eta: 5:46:31 time: 0.5555 data_time: 0.0076 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0068 decode.acc_seg: 99.7313 aux.loss_ce: 0.0076 aux.acc_seg: 99.1595 +04/18 21:13:18 - mmengine - INFO - Iter(train) [122450/160000] lr: 2.7858e-03 eta: 5:46:03 time: 0.5556 data_time: 0.0072 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0064 decode.acc_seg: 99.7914 aux.loss_ce: 0.0073 aux.acc_seg: 99.1714 +04/18 21:13:45 - mmengine - INFO - Iter(train) [122500/160000] lr: 2.7826e-03 eta: 5:45:36 time: 0.5556 data_time: 0.0078 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0067 decode.acc_seg: 99.8101 aux.loss_ce: 0.0072 aux.acc_seg: 99.4071 +04/18 21:14:13 - mmengine - INFO - Iter(train) [122550/160000] lr: 2.7794e-03 eta: 5:45:08 time: 0.5539 data_time: 0.0069 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0073 decode.acc_seg: 99.6625 aux.loss_ce: 0.0078 aux.acc_seg: 99.1623 +04/18 21:14:41 - mmengine - INFO - Iter(train) [122600/160000] lr: 2.7762e-03 eta: 5:44:40 time: 0.5530 data_time: 0.0062 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0064 decode.acc_seg: 99.6551 aux.loss_ce: 0.0073 aux.acc_seg: 99.1289 +04/18 21:15:09 - mmengine - INFO - Iter(train) [122650/160000] lr: 2.7730e-03 eta: 5:44:13 time: 0.5644 data_time: 0.0063 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0065 decode.acc_seg: 99.7834 aux.loss_ce: 0.0071 aux.acc_seg: 99.4628 +04/18 21:15:36 - mmengine - INFO - Iter(train) [122700/160000] lr: 2.7697e-03 eta: 5:43:45 time: 0.5541 data_time: 0.0070 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0065 decode.acc_seg: 99.7046 aux.loss_ce: 0.0074 aux.acc_seg: 98.9109 +04/18 21:16:04 - mmengine - INFO - Iter(train) [122750/160000] lr: 2.7665e-03 eta: 5:43:18 time: 0.5557 data_time: 0.0070 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0067 decode.acc_seg: 99.7723 aux.loss_ce: 0.0074 aux.acc_seg: 99.4095 +04/18 21:16:32 - mmengine - INFO - Iter(train) [122800/160000] lr: 2.7633e-03 eta: 5:42:50 time: 0.5546 data_time: 0.0072 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0069 decode.acc_seg: 99.7140 aux.loss_ce: 0.0074 aux.acc_seg: 99.2131 +04/18 21:17:00 - mmengine - INFO - Iter(train) [122850/160000] lr: 2.7601e-03 eta: 5:42:22 time: 0.5548 data_time: 0.0071 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0066 decode.acc_seg: 99.7643 aux.loss_ce: 0.0071 aux.acc_seg: 99.3224 +04/18 21:17:28 - mmengine - INFO - Iter(train) [122900/160000] lr: 2.7568e-03 eta: 5:41:55 time: 0.5552 data_time: 0.0072 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0066 decode.acc_seg: 99.6893 aux.loss_ce: 0.0070 aux.acc_seg: 99.1234 +04/18 21:17:55 - mmengine - INFO - Iter(train) [122950/160000] lr: 2.7536e-03 eta: 5:41:27 time: 0.5542 data_time: 0.0069 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0068 decode.acc_seg: 99.7011 aux.loss_ce: 0.0073 aux.acc_seg: 99.1089 +04/18 21:18:23 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 21:18:23 - mmengine - INFO - Iter(train) [123000/160000] lr: 2.7504e-03 eta: 5:41:00 time: 0.5541 data_time: 0.0066 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0065 decode.acc_seg: 99.7240 aux.loss_ce: 0.0072 aux.acc_seg: 99.1662 +04/18 21:18:51 - mmengine - INFO - Iter(train) [123050/160000] lr: 2.7472e-03 eta: 5:40:32 time: 0.5555 data_time: 0.0069 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0068 decode.acc_seg: 99.7085 aux.loss_ce: 0.0071 aux.acc_seg: 99.1843 +04/18 21:19:19 - mmengine - INFO - Iter(train) [123100/160000] lr: 2.7440e-03 eta: 5:40:04 time: 0.5564 data_time: 0.0061 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0066 decode.acc_seg: 99.7354 aux.loss_ce: 0.0072 aux.acc_seg: 99.2057 +04/18 21:19:47 - mmengine - INFO - Iter(train) [123150/160000] lr: 2.7407e-03 eta: 5:39:37 time: 0.5559 data_time: 0.0064 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0063 decode.acc_seg: 99.7125 aux.loss_ce: 0.0070 aux.acc_seg: 99.1646 +04/18 21:20:14 - mmengine - INFO - Iter(train) [123200/160000] lr: 2.7375e-03 eta: 5:39:09 time: 0.5541 data_time: 0.0064 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0063 decode.acc_seg: 99.7466 aux.loss_ce: 0.0067 aux.acc_seg: 99.2326 +04/18 21:20:42 - mmengine - INFO - Iter(train) [123250/160000] lr: 2.7343e-03 eta: 5:38:41 time: 0.5550 data_time: 0.0070 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0069 decode.acc_seg: 99.7112 aux.loss_ce: 0.0074 aux.acc_seg: 99.1482 +04/18 21:21:10 - mmengine - INFO - Iter(train) [123300/160000] lr: 2.7311e-03 eta: 5:38:14 time: 0.5544 data_time: 0.0065 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0067 decode.acc_seg: 99.7950 aux.loss_ce: 0.0073 aux.acc_seg: 99.2402 +04/18 21:21:38 - mmengine - INFO - Iter(train) [123350/160000] lr: 2.7278e-03 eta: 5:37:46 time: 0.5556 data_time: 0.0072 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0068 decode.acc_seg: 99.6663 aux.loss_ce: 0.0073 aux.acc_seg: 99.0669 +04/18 21:22:05 - mmengine - INFO - Iter(train) [123400/160000] lr: 2.7246e-03 eta: 5:37:19 time: 0.5565 data_time: 0.0069 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0071 decode.acc_seg: 99.6578 aux.loss_ce: 0.0079 aux.acc_seg: 98.9572 +04/18 21:22:33 - mmengine - INFO - Iter(train) [123450/160000] lr: 2.7214e-03 eta: 5:36:51 time: 0.5557 data_time: 0.0062 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0068 decode.acc_seg: 99.7147 aux.loss_ce: 0.0071 aux.acc_seg: 99.3171 +04/18 21:23:01 - mmengine - INFO - Iter(train) [123500/160000] lr: 2.7181e-03 eta: 5:36:23 time: 0.5556 data_time: 0.0063 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0064 decode.acc_seg: 99.8658 aux.loss_ce: 0.0067 aux.acc_seg: 99.5777 +04/18 21:23:29 - mmengine - INFO - Iter(train) [123550/160000] lr: 2.7149e-03 eta: 5:35:56 time: 0.5555 data_time: 0.0072 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0066 decode.acc_seg: 99.7674 aux.loss_ce: 0.0070 aux.acc_seg: 99.3142 +04/18 21:23:57 - mmengine - INFO - Iter(train) [123600/160000] lr: 2.7117e-03 eta: 5:35:28 time: 0.5556 data_time: 0.0074 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0072 decode.acc_seg: 99.6979 aux.loss_ce: 0.0075 aux.acc_seg: 98.9636 +04/18 21:24:24 - mmengine - INFO - Iter(train) [123650/160000] lr: 2.7085e-03 eta: 5:35:01 time: 0.5562 data_time: 0.0073 memory: 7635 loss: 0.0152 decode.loss_ce: 0.0072 decode.acc_seg: 99.6576 aux.loss_ce: 0.0080 aux.acc_seg: 98.9332 +04/18 21:24:52 - mmengine - INFO - Iter(train) [123700/160000] lr: 2.7052e-03 eta: 5:34:33 time: 0.5555 data_time: 0.0062 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0069 decode.acc_seg: 99.7257 aux.loss_ce: 0.0072 aux.acc_seg: 99.1588 +04/18 21:25:20 - mmengine - INFO - Iter(train) [123750/160000] lr: 2.7020e-03 eta: 5:34:05 time: 0.5546 data_time: 0.0066 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0064 decode.acc_seg: 99.8016 aux.loss_ce: 0.0070 aux.acc_seg: 99.3775 +04/18 21:25:48 - mmengine - INFO - Iter(train) [123800/160000] lr: 2.6988e-03 eta: 5:33:38 time: 0.5548 data_time: 0.0064 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0066 decode.acc_seg: 99.7156 aux.loss_ce: 0.0069 aux.acc_seg: 99.3465 +04/18 21:26:15 - mmengine - INFO - Iter(train) [123850/160000] lr: 2.6955e-03 eta: 5:33:10 time: 0.5547 data_time: 0.0069 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0073 decode.acc_seg: 99.7180 aux.loss_ce: 0.0077 aux.acc_seg: 99.0951 +04/18 21:26:43 - mmengine - INFO - Iter(train) [123900/160000] lr: 2.6923e-03 eta: 5:32:42 time: 0.5556 data_time: 0.0069 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0073 decode.acc_seg: 99.7747 aux.loss_ce: 0.0077 aux.acc_seg: 99.3084 +04/18 21:27:11 - mmengine - INFO - Iter(train) [123950/160000] lr: 2.6891e-03 eta: 5:32:15 time: 0.5559 data_time: 0.0072 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0073 decode.acc_seg: 99.7608 aux.loss_ce: 0.0073 aux.acc_seg: 99.3449 +04/18 21:27:39 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 21:27:39 - mmengine - INFO - Iter(train) [124000/160000] lr: 2.6858e-03 eta: 5:31:47 time: 0.5536 data_time: 0.0061 memory: 7635 loss: 0.0173 decode.loss_ce: 0.0090 decode.acc_seg: 99.6581 aux.loss_ce: 0.0083 aux.acc_seg: 99.2547 +04/18 21:28:07 - mmengine - INFO - Iter(train) [124050/160000] lr: 2.6826e-03 eta: 5:31:20 time: 0.5546 data_time: 0.0066 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0078 decode.acc_seg: 99.6129 aux.loss_ce: 0.0075 aux.acc_seg: 99.0871 +04/18 21:28:34 - mmengine - INFO - Iter(train) [124100/160000] lr: 2.6794e-03 eta: 5:30:52 time: 0.5558 data_time: 0.0064 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0073 decode.acc_seg: 99.6805 aux.loss_ce: 0.0072 aux.acc_seg: 99.3385 +04/18 21:29:02 - mmengine - INFO - Iter(train) [124150/160000] lr: 2.6761e-03 eta: 5:30:24 time: 0.5540 data_time: 0.0066 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0070 decode.acc_seg: 99.7404 aux.loss_ce: 0.0071 aux.acc_seg: 99.1956 +04/18 21:29:30 - mmengine - INFO - Iter(train) [124200/160000] lr: 2.6729e-03 eta: 5:29:57 time: 0.5558 data_time: 0.0069 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0074 decode.acc_seg: 99.6528 aux.loss_ce: 0.0074 aux.acc_seg: 99.0789 +04/18 21:29:58 - mmengine - INFO - Iter(train) [124250/160000] lr: 2.6697e-03 eta: 5:29:29 time: 0.5556 data_time: 0.0071 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0069 decode.acc_seg: 99.7506 aux.loss_ce: 0.0070 aux.acc_seg: 99.4198 +04/18 21:30:26 - mmengine - INFO - Iter(train) [124300/160000] lr: 2.6664e-03 eta: 5:29:02 time: 0.5549 data_time: 0.0071 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0066 decode.acc_seg: 99.6850 aux.loss_ce: 0.0072 aux.acc_seg: 99.1245 +04/18 21:30:53 - mmengine - INFO - Iter(train) [124350/160000] lr: 2.6632e-03 eta: 5:28:34 time: 0.5536 data_time: 0.0061 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0070 decode.acc_seg: 99.7113 aux.loss_ce: 0.0076 aux.acc_seg: 99.2085 +04/18 21:31:21 - mmengine - INFO - Iter(train) [124400/160000] lr: 2.6600e-03 eta: 5:28:06 time: 0.5555 data_time: 0.0072 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0069 decode.acc_seg: 99.7376 aux.loss_ce: 0.0074 aux.acc_seg: 99.2411 +04/18 21:31:49 - mmengine - INFO - Iter(train) [124450/160000] lr: 2.6567e-03 eta: 5:27:39 time: 0.5545 data_time: 0.0069 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0068 decode.acc_seg: 99.7454 aux.loss_ce: 0.0073 aux.acc_seg: 99.1824 +04/18 21:32:17 - mmengine - INFO - Iter(train) [124500/160000] lr: 2.6535e-03 eta: 5:27:11 time: 0.5542 data_time: 0.0066 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0069 decode.acc_seg: 99.7325 aux.loss_ce: 0.0075 aux.acc_seg: 99.2898 +04/18 21:32:44 - mmengine - INFO - Iter(train) [124550/160000] lr: 2.6503e-03 eta: 5:26:43 time: 0.5555 data_time: 0.0073 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0061 decode.acc_seg: 99.7578 aux.loss_ce: 0.0069 aux.acc_seg: 99.2954 +04/18 21:33:12 - mmengine - INFO - Iter(train) [124600/160000] lr: 2.6470e-03 eta: 5:26:16 time: 0.5555 data_time: 0.0070 memory: 7635 loss: 0.0127 decode.loss_ce: 0.0062 decode.acc_seg: 99.7704 aux.loss_ce: 0.0065 aux.acc_seg: 99.3500 +04/18 21:33:40 - mmengine - INFO - Iter(train) [124650/160000] lr: 2.6438e-03 eta: 5:25:48 time: 0.5545 data_time: 0.0066 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0072 decode.acc_seg: 99.6255 aux.loss_ce: 0.0075 aux.acc_seg: 99.0353 +04/18 21:34:08 - mmengine - INFO - Iter(train) [124700/160000] lr: 2.6405e-03 eta: 5:25:21 time: 0.5537 data_time: 0.0061 memory: 7635 loss: 0.0129 decode.loss_ce: 0.0061 decode.acc_seg: 99.8405 aux.loss_ce: 0.0068 aux.acc_seg: 99.5062 +04/18 21:34:35 - mmengine - INFO - Iter(train) [124750/160000] lr: 2.6373e-03 eta: 5:24:53 time: 0.5538 data_time: 0.0064 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0076 decode.acc_seg: 99.7898 aux.loss_ce: 0.0078 aux.acc_seg: 99.4054 +04/18 21:35:03 - mmengine - INFO - Iter(train) [124800/160000] lr: 2.6341e-03 eta: 5:24:25 time: 0.5544 data_time: 0.0065 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0068 decode.acc_seg: 99.7371 aux.loss_ce: 0.0073 aux.acc_seg: 99.2552 +04/18 21:35:31 - mmengine - INFO - Iter(train) [124850/160000] lr: 2.6308e-03 eta: 5:23:58 time: 0.5554 data_time: 0.0077 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0066 decode.acc_seg: 99.7819 aux.loss_ce: 0.0073 aux.acc_seg: 99.3029 +04/18 21:35:59 - mmengine - INFO - Iter(train) [124900/160000] lr: 2.6276e-03 eta: 5:23:30 time: 0.5551 data_time: 0.0070 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0070 decode.acc_seg: 99.7383 aux.loss_ce: 0.0074 aux.acc_seg: 99.2082 +04/18 21:36:27 - mmengine - INFO - Iter(train) [124950/160000] lr: 2.6243e-03 eta: 5:23:02 time: 0.5545 data_time: 0.0071 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0066 decode.acc_seg: 99.7116 aux.loss_ce: 0.0076 aux.acc_seg: 99.3921 +04/18 21:36:55 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 21:36:55 - mmengine - INFO - Iter(train) [125000/160000] lr: 2.6211e-03 eta: 5:22:35 time: 0.5733 data_time: 0.0073 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0067 decode.acc_seg: 99.6754 aux.loss_ce: 0.0075 aux.acc_seg: 98.9037 +04/18 21:37:22 - mmengine - INFO - Iter(train) [125050/160000] lr: 2.6179e-03 eta: 5:22:07 time: 0.5554 data_time: 0.0070 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0064 decode.acc_seg: 99.7577 aux.loss_ce: 0.0072 aux.acc_seg: 99.2687 +04/18 21:37:50 - mmengine - INFO - Iter(train) [125100/160000] lr: 2.6146e-03 eta: 5:21:40 time: 0.5546 data_time: 0.0066 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0068 decode.acc_seg: 99.6872 aux.loss_ce: 0.0075 aux.acc_seg: 99.3011 +04/18 21:38:18 - mmengine - INFO - Iter(train) [125150/160000] lr: 2.6114e-03 eta: 5:21:12 time: 0.5561 data_time: 0.0066 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0069 decode.acc_seg: 99.7501 aux.loss_ce: 0.0075 aux.acc_seg: 99.2669 +04/18 21:38:46 - mmengine - INFO - Iter(train) [125200/160000] lr: 2.6081e-03 eta: 5:20:44 time: 0.5539 data_time: 0.0064 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0067 decode.acc_seg: 99.7517 aux.loss_ce: 0.0072 aux.acc_seg: 99.2521 +04/18 21:39:14 - mmengine - INFO - Iter(train) [125250/160000] lr: 2.6049e-03 eta: 5:20:17 time: 0.5546 data_time: 0.0069 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0071 decode.acc_seg: 99.7034 aux.loss_ce: 0.0074 aux.acc_seg: 99.3413 +04/18 21:39:41 - mmengine - INFO - Iter(train) [125300/160000] lr: 2.6016e-03 eta: 5:19:49 time: 0.5565 data_time: 0.0078 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0067 decode.acc_seg: 99.7685 aux.loss_ce: 0.0068 aux.acc_seg: 99.3257 +04/18 21:40:09 - mmengine - INFO - Iter(train) [125350/160000] lr: 2.5984e-03 eta: 5:19:22 time: 0.5566 data_time: 0.0069 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0069 decode.acc_seg: 99.7415 aux.loss_ce: 0.0075 aux.acc_seg: 99.2738 +04/18 21:40:37 - mmengine - INFO - Iter(train) [125400/160000] lr: 2.5952e-03 eta: 5:18:54 time: 0.5551 data_time: 0.0067 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0071 decode.acc_seg: 99.7213 aux.loss_ce: 0.0076 aux.acc_seg: 99.2002 +04/18 21:41:05 - mmengine - INFO - Iter(train) [125450/160000] lr: 2.5919e-03 eta: 5:18:26 time: 0.5559 data_time: 0.0069 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0065 decode.acc_seg: 99.7778 aux.loss_ce: 0.0072 aux.acc_seg: 99.1993 +04/18 21:41:32 - mmengine - INFO - Iter(train) [125500/160000] lr: 2.5887e-03 eta: 5:17:59 time: 0.5523 data_time: 0.0062 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0065 decode.acc_seg: 99.7118 aux.loss_ce: 0.0072 aux.acc_seg: 99.2544 +04/18 21:42:00 - mmengine - INFO - Iter(train) [125550/160000] lr: 2.5854e-03 eta: 5:17:31 time: 0.5545 data_time: 0.0067 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0064 decode.acc_seg: 99.7169 aux.loss_ce: 0.0072 aux.acc_seg: 99.1631 +04/18 21:42:28 - mmengine - INFO - Iter(train) [125600/160000] lr: 2.5822e-03 eta: 5:17:03 time: 0.5552 data_time: 0.0074 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0068 decode.acc_seg: 99.6750 aux.loss_ce: 0.0071 aux.acc_seg: 99.3027 +04/18 21:42:56 - mmengine - INFO - Iter(train) [125650/160000] lr: 2.5789e-03 eta: 5:16:36 time: 0.5554 data_time: 0.0065 memory: 7635 loss: 0.0127 decode.loss_ce: 0.0062 decode.acc_seg: 99.7605 aux.loss_ce: 0.0066 aux.acc_seg: 99.3504 +04/18 21:43:23 - mmengine - INFO - Iter(train) [125700/160000] lr: 2.5757e-03 eta: 5:16:08 time: 0.5550 data_time: 0.0066 memory: 7635 loss: 0.0128 decode.loss_ce: 0.0061 decode.acc_seg: 99.7562 aux.loss_ce: 0.0067 aux.acc_seg: 99.3752 +04/18 21:43:51 - mmengine - INFO - Iter(train) [125750/160000] lr: 2.5724e-03 eta: 5:15:41 time: 0.5557 data_time: 0.0062 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0064 decode.acc_seg: 99.7888 aux.loss_ce: 0.0071 aux.acc_seg: 99.4412 +04/18 21:44:19 - mmengine - INFO - Iter(train) [125800/160000] lr: 2.5692e-03 eta: 5:15:13 time: 0.5564 data_time: 0.0068 memory: 7635 loss: 0.0132 decode.loss_ce: 0.0062 decode.acc_seg: 99.7596 aux.loss_ce: 0.0069 aux.acc_seg: 99.4021 +04/18 21:44:47 - mmengine - INFO - Iter(train) [125850/160000] lr: 2.5659e-03 eta: 5:14:45 time: 0.5560 data_time: 0.0066 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0064 decode.acc_seg: 99.7498 aux.loss_ce: 0.0070 aux.acc_seg: 99.2921 +04/18 21:45:15 - mmengine - INFO - Iter(train) [125900/160000] lr: 2.5627e-03 eta: 5:14:18 time: 0.5558 data_time: 0.0064 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0063 decode.acc_seg: 99.8037 aux.loss_ce: 0.0071 aux.acc_seg: 99.4242 +04/18 21:45:42 - mmengine - INFO - Iter(train) [125950/160000] lr: 2.5594e-03 eta: 5:13:50 time: 0.5553 data_time: 0.0062 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0069 decode.acc_seg: 99.7516 aux.loss_ce: 0.0070 aux.acc_seg: 99.4208 +04/18 21:46:10 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 21:46:10 - mmengine - INFO - Iter(train) [126000/160000] lr: 2.5562e-03 eta: 5:13:22 time: 0.5554 data_time: 0.0070 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0067 decode.acc_seg: 99.7361 aux.loss_ce: 0.0078 aux.acc_seg: 99.3279 +04/18 21:46:38 - mmengine - INFO - Iter(train) [126050/160000] lr: 2.5529e-03 eta: 5:12:55 time: 0.5562 data_time: 0.0071 memory: 7635 loss: 0.0131 decode.loss_ce: 0.0061 decode.acc_seg: 99.6084 aux.loss_ce: 0.0070 aux.acc_seg: 98.8375 +04/18 21:47:06 - mmengine - INFO - Iter(train) [126100/160000] lr: 2.5497e-03 eta: 5:12:27 time: 0.5555 data_time: 0.0063 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0060 decode.acc_seg: 99.6986 aux.loss_ce: 0.0069 aux.acc_seg: 99.1886 +04/18 21:47:34 - mmengine - INFO - Iter(train) [126150/160000] lr: 2.5464e-03 eta: 5:12:00 time: 0.5556 data_time: 0.0072 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0066 decode.acc_seg: 99.7086 aux.loss_ce: 0.0069 aux.acc_seg: 99.1688 +04/18 21:48:01 - mmengine - INFO - Iter(train) [126200/160000] lr: 2.5432e-03 eta: 5:11:32 time: 0.5551 data_time: 0.0073 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0073 decode.acc_seg: 99.6912 aux.loss_ce: 0.0077 aux.acc_seg: 99.2783 +04/18 21:48:29 - mmengine - INFO - Iter(train) [126250/160000] lr: 2.5399e-03 eta: 5:11:04 time: 0.5544 data_time: 0.0064 memory: 7635 loss: 0.0127 decode.loss_ce: 0.0060 decode.acc_seg: 99.7904 aux.loss_ce: 0.0067 aux.acc_seg: 99.2697 +04/18 21:48:57 - mmengine - INFO - Iter(train) [126300/160000] lr: 2.5367e-03 eta: 5:10:37 time: 0.5549 data_time: 0.0072 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0066 decode.acc_seg: 99.7656 aux.loss_ce: 0.0072 aux.acc_seg: 99.1049 +04/18 21:49:25 - mmengine - INFO - Iter(train) [126350/160000] lr: 2.5334e-03 eta: 5:10:09 time: 0.5549 data_time: 0.0069 memory: 7635 loss: 0.0132 decode.loss_ce: 0.0063 decode.acc_seg: 99.7588 aux.loss_ce: 0.0069 aux.acc_seg: 99.3914 +04/18 21:49:53 - mmengine - INFO - Iter(train) [126400/160000] lr: 2.5302e-03 eta: 5:09:42 time: 0.5555 data_time: 0.0066 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0070 decode.acc_seg: 99.7691 aux.loss_ce: 0.0076 aux.acc_seg: 99.2879 +04/18 21:50:20 - mmengine - INFO - Iter(train) [126450/160000] lr: 2.5269e-03 eta: 5:09:14 time: 0.5558 data_time: 0.0066 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0068 decode.acc_seg: 99.7457 aux.loss_ce: 0.0072 aux.acc_seg: 99.3004 +04/18 21:50:48 - mmengine - INFO - Iter(train) [126500/160000] lr: 2.5237e-03 eta: 5:08:46 time: 0.5540 data_time: 0.0065 memory: 7635 loss: 0.0128 decode.loss_ce: 0.0061 decode.acc_seg: 99.6911 aux.loss_ce: 0.0068 aux.acc_seg: 99.1219 +04/18 21:51:16 - mmengine - INFO - Iter(train) [126550/160000] lr: 2.5204e-03 eta: 5:08:19 time: 0.5541 data_time: 0.0068 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0062 decode.acc_seg: 99.7453 aux.loss_ce: 0.0067 aux.acc_seg: 99.2411 +04/18 21:51:44 - mmengine - INFO - Iter(train) [126600/160000] lr: 2.5171e-03 eta: 5:07:51 time: 0.5542 data_time: 0.0069 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0070 decode.acc_seg: 99.6019 aux.loss_ce: 0.0074 aux.acc_seg: 99.0595 +04/18 21:52:11 - mmengine - INFO - Iter(train) [126650/160000] lr: 2.5139e-03 eta: 5:07:23 time: 0.5549 data_time: 0.0072 memory: 7635 loss: 0.0129 decode.loss_ce: 0.0063 decode.acc_seg: 99.7804 aux.loss_ce: 0.0067 aux.acc_seg: 99.3526 +04/18 21:52:39 - mmengine - INFO - Iter(train) [126700/160000] lr: 2.5106e-03 eta: 5:06:56 time: 0.5556 data_time: 0.0064 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0064 decode.acc_seg: 99.7977 aux.loss_ce: 0.0071 aux.acc_seg: 99.4468 +04/18 21:53:07 - mmengine - INFO - Iter(train) [126750/160000] lr: 2.5074e-03 eta: 5:06:28 time: 0.5566 data_time: 0.0073 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0066 decode.acc_seg: 99.7871 aux.loss_ce: 0.0075 aux.acc_seg: 99.3344 +04/18 21:53:35 - mmengine - INFO - Iter(train) [126800/160000] lr: 2.5041e-03 eta: 5:06:01 time: 0.5546 data_time: 0.0066 memory: 7635 loss: 0.0131 decode.loss_ce: 0.0062 decode.acc_seg: 99.7893 aux.loss_ce: 0.0070 aux.acc_seg: 99.4941 +04/18 21:54:03 - mmengine - INFO - Iter(train) [126850/160000] lr: 2.5008e-03 eta: 5:05:33 time: 0.5545 data_time: 0.0065 memory: 7635 loss: 0.0132 decode.loss_ce: 0.0062 decode.acc_seg: 99.7021 aux.loss_ce: 0.0071 aux.acc_seg: 98.9094 +04/18 21:54:30 - mmengine - INFO - Iter(train) [126900/160000] lr: 2.4976e-03 eta: 5:05:05 time: 0.5569 data_time: 0.0064 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0066 decode.acc_seg: 99.7677 aux.loss_ce: 0.0074 aux.acc_seg: 99.2421 +04/18 21:54:58 - mmengine - INFO - Iter(train) [126950/160000] lr: 2.4943e-03 eta: 5:04:38 time: 0.5640 data_time: 0.0070 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0070 decode.acc_seg: 99.7974 aux.loss_ce: 0.0074 aux.acc_seg: 99.5192 +04/18 21:55:26 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 21:55:26 - mmengine - INFO - Iter(train) [127000/160000] lr: 2.4911e-03 eta: 5:04:10 time: 0.5559 data_time: 0.0068 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0067 decode.acc_seg: 99.7417 aux.loss_ce: 0.0075 aux.acc_seg: 99.4516 +04/18 21:55:54 - mmengine - INFO - Iter(train) [127050/160000] lr: 2.4878e-03 eta: 5:03:43 time: 0.5547 data_time: 0.0064 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0075 decode.acc_seg: 99.7052 aux.loss_ce: 0.0075 aux.acc_seg: 99.0387 +04/18 21:56:22 - mmengine - INFO - Iter(train) [127100/160000] lr: 2.4845e-03 eta: 5:03:15 time: 0.5544 data_time: 0.0063 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0064 decode.acc_seg: 99.8090 aux.loss_ce: 0.0074 aux.acc_seg: 99.4081 +04/18 21:56:50 - mmengine - INFO - Iter(train) [127150/160000] lr: 2.4813e-03 eta: 5:02:47 time: 0.5738 data_time: 0.0064 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0063 decode.acc_seg: 99.8383 aux.loss_ce: 0.0071 aux.acc_seg: 99.4491 +04/18 21:57:17 - mmengine - INFO - Iter(train) [127200/160000] lr: 2.4780e-03 eta: 5:02:20 time: 0.5639 data_time: 0.0068 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0071 decode.acc_seg: 99.7498 aux.loss_ce: 0.0078 aux.acc_seg: 99.1600 +04/18 21:57:45 - mmengine - INFO - Iter(train) [127250/160000] lr: 2.4748e-03 eta: 5:01:52 time: 0.5543 data_time: 0.0070 memory: 7635 loss: 0.0132 decode.loss_ce: 0.0063 decode.acc_seg: 99.7391 aux.loss_ce: 0.0069 aux.acc_seg: 99.3620 +04/18 21:58:13 - mmengine - INFO - Iter(train) [127300/160000] lr: 2.4715e-03 eta: 5:01:24 time: 0.5555 data_time: 0.0068 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0074 decode.acc_seg: 99.6553 aux.loss_ce: 0.0080 aux.acc_seg: 98.9337 +04/18 21:58:41 - mmengine - INFO - Iter(train) [127350/160000] lr: 2.4682e-03 eta: 5:00:57 time: 0.5537 data_time: 0.0063 memory: 7635 loss: 0.0163 decode.loss_ce: 0.0080 decode.acc_seg: 99.6821 aux.loss_ce: 0.0083 aux.acc_seg: 99.1216 +04/18 21:59:09 - mmengine - INFO - Iter(train) [127400/160000] lr: 2.4650e-03 eta: 5:00:29 time: 0.5552 data_time: 0.0069 memory: 7635 loss: 0.0126 decode.loss_ce: 0.0060 decode.acc_seg: 99.7319 aux.loss_ce: 0.0067 aux.acc_seg: 99.2224 +04/18 21:59:36 - mmengine - INFO - Iter(train) [127450/160000] lr: 2.4617e-03 eta: 5:00:02 time: 0.5560 data_time: 0.0068 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0071 decode.acc_seg: 99.7175 aux.loss_ce: 0.0076 aux.acc_seg: 99.2351 +04/18 22:00:04 - mmengine - INFO - Iter(train) [127500/160000] lr: 2.4584e-03 eta: 4:59:34 time: 0.5551 data_time: 0.0071 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0066 decode.acc_seg: 99.7566 aux.loss_ce: 0.0070 aux.acc_seg: 99.3694 +04/18 22:00:32 - mmengine - INFO - Iter(train) [127550/160000] lr: 2.4552e-03 eta: 4:59:06 time: 0.5554 data_time: 0.0069 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0071 decode.acc_seg: 99.6647 aux.loss_ce: 0.0079 aux.acc_seg: 98.8403 +04/18 22:01:00 - mmengine - INFO - Iter(train) [127600/160000] lr: 2.4519e-03 eta: 4:58:39 time: 0.5561 data_time: 0.0074 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0061 decode.acc_seg: 99.7548 aux.loss_ce: 0.0069 aux.acc_seg: 99.2050 +04/18 22:01:27 - mmengine - INFO - Iter(train) [127650/160000] lr: 2.4486e-03 eta: 4:58:11 time: 0.5558 data_time: 0.0064 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0068 decode.acc_seg: 99.7610 aux.loss_ce: 0.0079 aux.acc_seg: 99.3730 +04/18 22:01:55 - mmengine - INFO - Iter(train) [127700/160000] lr: 2.4454e-03 eta: 4:57:44 time: 0.5555 data_time: 0.0071 memory: 7635 loss: 0.0122 decode.loss_ce: 0.0060 decode.acc_seg: 99.8035 aux.loss_ce: 0.0062 aux.acc_seg: 99.6104 +04/18 22:02:23 - mmengine - INFO - Iter(train) [127750/160000] lr: 2.4421e-03 eta: 4:57:16 time: 0.5550 data_time: 0.0071 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0070 decode.acc_seg: 99.7979 aux.loss_ce: 0.0074 aux.acc_seg: 99.4246 +04/18 22:02:51 - mmengine - INFO - Iter(train) [127800/160000] lr: 2.4388e-03 eta: 4:56:48 time: 0.5549 data_time: 0.0062 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0071 decode.acc_seg: 99.7128 aux.loss_ce: 0.0078 aux.acc_seg: 99.1030 +04/18 22:03:19 - mmengine - INFO - Iter(train) [127850/160000] lr: 2.4356e-03 eta: 4:56:21 time: 0.5565 data_time: 0.0076 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0073 decode.acc_seg: 99.6639 aux.loss_ce: 0.0075 aux.acc_seg: 99.2200 +04/18 22:03:47 - mmengine - INFO - Iter(train) [127900/160000] lr: 2.4323e-03 eta: 4:55:53 time: 0.5556 data_time: 0.0066 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0069 decode.acc_seg: 99.5625 aux.loss_ce: 0.0073 aux.acc_seg: 99.0203 +04/18 22:04:14 - mmengine - INFO - Iter(train) [127950/160000] lr: 2.4290e-03 eta: 4:55:25 time: 0.5552 data_time: 0.0066 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0065 decode.acc_seg: 99.7223 aux.loss_ce: 0.0072 aux.acc_seg: 99.1844 +04/18 22:04:42 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 22:04:42 - mmengine - INFO - Iter(train) [128000/160000] lr: 2.4258e-03 eta: 4:54:58 time: 0.5557 data_time: 0.0070 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0064 decode.acc_seg: 99.7183 aux.loss_ce: 0.0071 aux.acc_seg: 99.1471 +04/18 22:05:10 - mmengine - INFO - Iter(train) [128050/160000] lr: 2.4225e-03 eta: 4:54:30 time: 0.5547 data_time: 0.0065 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0071 decode.acc_seg: 99.6617 aux.loss_ce: 0.0078 aux.acc_seg: 99.0107 +04/18 22:05:38 - mmengine - INFO - Iter(train) [128100/160000] lr: 2.4192e-03 eta: 4:54:03 time: 0.5559 data_time: 0.0068 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0070 decode.acc_seg: 99.6387 aux.loss_ce: 0.0075 aux.acc_seg: 99.1412 +04/18 22:06:06 - mmengine - INFO - Iter(train) [128150/160000] lr: 2.4159e-03 eta: 4:53:35 time: 0.5531 data_time: 0.0068 memory: 7635 loss: 0.0160 decode.loss_ce: 0.0076 decode.acc_seg: 99.5480 aux.loss_ce: 0.0083 aux.acc_seg: 98.7090 +04/18 22:06:33 - mmengine - INFO - Iter(train) [128200/160000] lr: 2.4127e-03 eta: 4:53:07 time: 0.5544 data_time: 0.0070 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0071 decode.acc_seg: 99.6588 aux.loss_ce: 0.0074 aux.acc_seg: 99.1114 +04/18 22:07:01 - mmengine - INFO - Iter(train) [128250/160000] lr: 2.4094e-03 eta: 4:52:40 time: 0.5566 data_time: 0.0068 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0068 decode.acc_seg: 99.7232 aux.loss_ce: 0.0073 aux.acc_seg: 99.2478 +04/18 22:07:29 - mmengine - INFO - Iter(train) [128300/160000] lr: 2.4061e-03 eta: 4:52:12 time: 0.5548 data_time: 0.0070 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0067 decode.acc_seg: 99.7443 aux.loss_ce: 0.0073 aux.acc_seg: 99.2952 +04/18 22:07:57 - mmengine - INFO - Iter(train) [128350/160000] lr: 2.4029e-03 eta: 4:51:44 time: 0.5537 data_time: 0.0069 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0072 decode.acc_seg: 99.7016 aux.loss_ce: 0.0082 aux.acc_seg: 99.1817 +04/18 22:08:25 - mmengine - INFO - Iter(train) [128400/160000] lr: 2.3996e-03 eta: 4:51:17 time: 0.5556 data_time: 0.0071 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0064 decode.acc_seg: 99.7628 aux.loss_ce: 0.0071 aux.acc_seg: 99.3367 +04/18 22:08:52 - mmengine - INFO - Iter(train) [128450/160000] lr: 2.3963e-03 eta: 4:50:49 time: 0.5557 data_time: 0.0069 memory: 7635 loss: 0.0131 decode.loss_ce: 0.0063 decode.acc_seg: 99.8194 aux.loss_ce: 0.0069 aux.acc_seg: 99.4799 +04/18 22:09:20 - mmengine - INFO - Iter(train) [128500/160000] lr: 2.3930e-03 eta: 4:50:22 time: 0.5542 data_time: 0.0076 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0066 decode.acc_seg: 99.7598 aux.loss_ce: 0.0075 aux.acc_seg: 99.1301 +04/18 22:09:48 - mmengine - INFO - Iter(train) [128550/160000] lr: 2.3898e-03 eta: 4:49:54 time: 0.5542 data_time: 0.0066 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0070 decode.acc_seg: 99.6736 aux.loss_ce: 0.0076 aux.acc_seg: 99.2837 +04/18 22:10:16 - mmengine - INFO - Iter(train) [128600/160000] lr: 2.3865e-03 eta: 4:49:26 time: 0.5555 data_time: 0.0073 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0073 decode.acc_seg: 99.7605 aux.loss_ce: 0.0079 aux.acc_seg: 99.3143 +04/18 22:10:44 - mmengine - INFO - Iter(train) [128650/160000] lr: 2.3832e-03 eta: 4:48:59 time: 0.5564 data_time: 0.0064 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0070 decode.acc_seg: 99.6891 aux.loss_ce: 0.0074 aux.acc_seg: 99.3534 +04/18 22:11:11 - mmengine - INFO - Iter(train) [128700/160000] lr: 2.3799e-03 eta: 4:48:31 time: 0.5561 data_time: 0.0069 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0063 decode.acc_seg: 99.7433 aux.loss_ce: 0.0070 aux.acc_seg: 99.2893 +04/18 22:11:39 - mmengine - INFO - Iter(train) [128750/160000] lr: 2.3766e-03 eta: 4:48:03 time: 0.5550 data_time: 0.0071 memory: 7635 loss: 0.0124 decode.loss_ce: 0.0059 decode.acc_seg: 99.7640 aux.loss_ce: 0.0065 aux.acc_seg: 99.3477 +04/18 22:12:07 - mmengine - INFO - Iter(train) [128800/160000] lr: 2.3734e-03 eta: 4:47:36 time: 0.5568 data_time: 0.0074 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0069 decode.acc_seg: 99.7740 aux.loss_ce: 0.0082 aux.acc_seg: 99.2670 +04/18 22:12:35 - mmengine - INFO - Iter(train) [128850/160000] lr: 2.3701e-03 eta: 4:47:08 time: 0.5557 data_time: 0.0067 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0069 decode.acc_seg: 99.6268 aux.loss_ce: 0.0079 aux.acc_seg: 98.8998 +04/18 22:13:03 - mmengine - INFO - Iter(train) [128900/160000] lr: 2.3668e-03 eta: 4:46:41 time: 0.5543 data_time: 0.0066 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0082 decode.acc_seg: 99.6543 aux.loss_ce: 0.0081 aux.acc_seg: 99.3200 +04/18 22:13:30 - mmengine - INFO - Iter(train) [128950/160000] lr: 2.3635e-03 eta: 4:46:13 time: 0.5560 data_time: 0.0066 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0078 decode.acc_seg: 99.7135 aux.loss_ce: 0.0077 aux.acc_seg: 99.2747 +04/18 22:13:58 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 22:13:58 - mmengine - INFO - Iter(train) [129000/160000] lr: 2.3602e-03 eta: 4:45:45 time: 0.5563 data_time: 0.0069 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0068 decode.acc_seg: 99.6754 aux.loss_ce: 0.0071 aux.acc_seg: 98.9764 +04/18 22:14:26 - mmengine - INFO - Iter(train) [129050/160000] lr: 2.3570e-03 eta: 4:45:18 time: 0.5567 data_time: 0.0071 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0075 decode.acc_seg: 99.6849 aux.loss_ce: 0.0080 aux.acc_seg: 99.0967 +04/18 22:14:54 - mmengine - INFO - Iter(train) [129100/160000] lr: 2.3537e-03 eta: 4:44:50 time: 0.5653 data_time: 0.0065 memory: 7635 loss: 0.0131 decode.loss_ce: 0.0062 decode.acc_seg: 99.7360 aux.loss_ce: 0.0068 aux.acc_seg: 99.1823 +04/18 22:15:22 - mmengine - INFO - Iter(train) [129150/160000] lr: 2.3504e-03 eta: 4:44:23 time: 0.5553 data_time: 0.0071 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0062 decode.acc_seg: 99.7160 aux.loss_ce: 0.0068 aux.acc_seg: 99.1836 +04/18 22:15:49 - mmengine - INFO - Iter(train) [129200/160000] lr: 2.3471e-03 eta: 4:43:55 time: 0.5556 data_time: 0.0067 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0066 decode.acc_seg: 99.6051 aux.loss_ce: 0.0075 aux.acc_seg: 98.8124 +04/18 22:16:17 - mmengine - INFO - Iter(train) [129250/160000] lr: 2.3438e-03 eta: 4:43:27 time: 0.5571 data_time: 0.0067 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0065 decode.acc_seg: 99.7875 aux.loss_ce: 0.0073 aux.acc_seg: 99.3378 +04/18 22:16:45 - mmengine - INFO - Iter(train) [129300/160000] lr: 2.3405e-03 eta: 4:43:00 time: 0.5565 data_time: 0.0066 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0066 decode.acc_seg: 99.7327 aux.loss_ce: 0.0073 aux.acc_seg: 99.0170 +04/18 22:17:13 - mmengine - INFO - Iter(train) [129350/160000] lr: 2.3373e-03 eta: 4:42:32 time: 0.5650 data_time: 0.0064 memory: 7635 loss: 0.0132 decode.loss_ce: 0.0064 decode.acc_seg: 99.7454 aux.loss_ce: 0.0068 aux.acc_seg: 99.3069 +04/18 22:17:41 - mmengine - INFO - Iter(train) [129400/160000] lr: 2.3340e-03 eta: 4:42:04 time: 0.5561 data_time: 0.0074 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0081 decode.acc_seg: 99.6684 aux.loss_ce: 0.0080 aux.acc_seg: 99.0279 +04/18 22:18:09 - mmengine - INFO - Iter(train) [129450/160000] lr: 2.3307e-03 eta: 4:41:37 time: 0.5534 data_time: 0.0064 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0077 decode.acc_seg: 99.6218 aux.loss_ce: 0.0073 aux.acc_seg: 99.3181 +04/18 22:18:36 - mmengine - INFO - Iter(train) [129500/160000] lr: 2.3274e-03 eta: 4:41:09 time: 0.5562 data_time: 0.0064 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0073 decode.acc_seg: 99.7198 aux.loss_ce: 0.0076 aux.acc_seg: 99.3068 +04/18 22:19:04 - mmengine - INFO - Iter(train) [129550/160000] lr: 2.3241e-03 eta: 4:40:42 time: 0.5559 data_time: 0.0069 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0073 decode.acc_seg: 99.6206 aux.loss_ce: 0.0072 aux.acc_seg: 99.1624 +04/18 22:19:32 - mmengine - INFO - Iter(train) [129600/160000] lr: 2.3208e-03 eta: 4:40:14 time: 0.5543 data_time: 0.0062 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0065 decode.acc_seg: 99.7700 aux.loss_ce: 0.0071 aux.acc_seg: 99.3279 +04/18 22:20:00 - mmengine - INFO - Iter(train) [129650/160000] lr: 2.3175e-03 eta: 4:39:46 time: 0.5557 data_time: 0.0073 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0068 decode.acc_seg: 99.6285 aux.loss_ce: 0.0073 aux.acc_seg: 99.0627 +04/18 22:20:28 - mmengine - INFO - Iter(train) [129700/160000] lr: 2.3143e-03 eta: 4:39:19 time: 0.5566 data_time: 0.0062 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0067 decode.acc_seg: 99.6714 aux.loss_ce: 0.0072 aux.acc_seg: 99.1223 +04/18 22:20:55 - mmengine - INFO - Iter(train) [129750/160000] lr: 2.3110e-03 eta: 4:38:51 time: 0.5543 data_time: 0.0065 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0065 decode.acc_seg: 99.8057 aux.loss_ce: 0.0070 aux.acc_seg: 99.3267 +04/18 22:21:23 - mmengine - INFO - Iter(train) [129800/160000] lr: 2.3077e-03 eta: 4:38:23 time: 0.5568 data_time: 0.0067 memory: 7635 loss: 0.0131 decode.loss_ce: 0.0063 decode.acc_seg: 99.7604 aux.loss_ce: 0.0068 aux.acc_seg: 99.4051 +04/18 22:21:51 - mmengine - INFO - Iter(train) [129850/160000] lr: 2.3044e-03 eta: 4:37:56 time: 0.5544 data_time: 0.0063 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0069 decode.acc_seg: 99.7201 aux.loss_ce: 0.0074 aux.acc_seg: 99.1375 +04/18 22:22:19 - mmengine - INFO - Iter(train) [129900/160000] lr: 2.3011e-03 eta: 4:37:28 time: 0.5556 data_time: 0.0067 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0066 decode.acc_seg: 99.7195 aux.loss_ce: 0.0071 aux.acc_seg: 99.2496 +04/18 22:22:46 - mmengine - INFO - Iter(train) [129950/160000] lr: 2.2978e-03 eta: 4:37:01 time: 0.5555 data_time: 0.0069 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0064 decode.acc_seg: 99.7942 aux.loss_ce: 0.0073 aux.acc_seg: 99.4062 +04/18 22:23:14 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 22:23:14 - mmengine - INFO - Iter(train) [130000/160000] lr: 2.2945e-03 eta: 4:36:33 time: 0.5549 data_time: 0.0064 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0074 decode.acc_seg: 99.7582 aux.loss_ce: 0.0075 aux.acc_seg: 99.2439 +04/18 22:23:14 - mmengine - INFO - Saving checkpoint at 130000 iterations +04/18 22:23:18 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:07 time: 0.0461 data_time: 0.0014 memory: 1657 +04/18 22:23:21 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:04 time: 0.0466 data_time: 0.0017 memory: 1657 +04/18 22:23:23 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:02 time: 0.0466 data_time: 0.0015 memory: 1657 +04/18 22:23:25 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.0454 data_time: 0.0011 memory: 1657 +04/18 22:23:25 - mmengine - INFO - per class results: +04/18 22:23:25 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 99.08 | 99.52 | 99.54 | 99.56 | 99.52 | +| contrast | 80.15 | 89.42 | 88.98 | 88.55 | 89.42 | ++------------+-------+-------+--------+-----------+--------+ +04/18 22:23:25 - mmengine - INFO - Iter(val) [200/200] aAcc: 99.1200 mIoU: 89.6200 mAcc: 94.4700 mFscore: 94.2600 mPrecision: 94.0500 mRecall: 94.4700 data_time: 0.0016 time: 0.0466 +04/18 22:23:53 - mmengine - INFO - Iter(train) [130050/160000] lr: 2.2912e-03 eta: 4:36:05 time: 0.5559 data_time: 0.0062 memory: 7635 loss: 0.0132 decode.loss_ce: 0.0064 decode.acc_seg: 99.6917 aux.loss_ce: 0.0068 aux.acc_seg: 99.1994 +04/18 22:24:21 - mmengine - INFO - Iter(train) [130100/160000] lr: 2.2879e-03 eta: 4:35:38 time: 0.5559 data_time: 0.0083 memory: 7635 loss: 0.0162 decode.loss_ce: 0.0080 decode.acc_seg: 99.7116 aux.loss_ce: 0.0082 aux.acc_seg: 99.2403 +04/18 22:24:49 - mmengine - INFO - Iter(train) [130150/160000] lr: 2.2846e-03 eta: 4:35:10 time: 0.5563 data_time: 0.0074 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0073 decode.acc_seg: 99.6255 aux.loss_ce: 0.0075 aux.acc_seg: 99.1024 +04/18 22:25:17 - mmengine - INFO - Iter(train) [130200/160000] lr: 2.2813e-03 eta: 4:34:42 time: 0.5563 data_time: 0.0071 memory: 7635 loss: 0.0164 decode.loss_ce: 0.0083 decode.acc_seg: 99.6431 aux.loss_ce: 0.0081 aux.acc_seg: 99.2937 +04/18 22:25:44 - mmengine - INFO - Iter(train) [130250/160000] lr: 2.2781e-03 eta: 4:34:15 time: 0.5559 data_time: 0.0067 memory: 7635 loss: 0.0167 decode.loss_ce: 0.0087 decode.acc_seg: 99.6962 aux.loss_ce: 0.0080 aux.acc_seg: 99.3298 +04/18 22:26:12 - mmengine - INFO - Iter(train) [130300/160000] lr: 2.2748e-03 eta: 4:33:47 time: 0.5547 data_time: 0.0062 memory: 7635 loss: 0.0156 decode.loss_ce: 0.0079 decode.acc_seg: 99.7221 aux.loss_ce: 0.0078 aux.acc_seg: 99.2074 +04/18 22:26:40 - mmengine - INFO - Iter(train) [130350/160000] lr: 2.2715e-03 eta: 4:33:20 time: 0.5566 data_time: 0.0071 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0071 decode.acc_seg: 99.7002 aux.loss_ce: 0.0074 aux.acc_seg: 99.1498 +04/18 22:27:08 - mmengine - INFO - Iter(train) [130400/160000] lr: 2.2682e-03 eta: 4:32:52 time: 0.5557 data_time: 0.0063 memory: 7635 loss: 0.0154 decode.loss_ce: 0.0076 decode.acc_seg: 99.6872 aux.loss_ce: 0.0078 aux.acc_seg: 99.0938 +04/18 22:27:36 - mmengine - INFO - Iter(train) [130450/160000] lr: 2.2649e-03 eta: 4:32:24 time: 0.5570 data_time: 0.0070 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0066 decode.acc_seg: 99.7517 aux.loss_ce: 0.0070 aux.acc_seg: 99.2586 +04/18 22:28:04 - mmengine - INFO - Iter(train) [130500/160000] lr: 2.2616e-03 eta: 4:31:57 time: 0.5558 data_time: 0.0067 memory: 7635 loss: 0.0153 decode.loss_ce: 0.0074 decode.acc_seg: 99.6672 aux.loss_ce: 0.0079 aux.acc_seg: 99.0973 +04/18 22:28:31 - mmengine - INFO - Iter(train) [130550/160000] lr: 2.2583e-03 eta: 4:31:29 time: 0.5561 data_time: 0.0065 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0067 decode.acc_seg: 99.6420 aux.loss_ce: 0.0077 aux.acc_seg: 99.1776 +04/18 22:28:59 - mmengine - INFO - Iter(train) [130600/160000] lr: 2.2550e-03 eta: 4:31:01 time: 0.5545 data_time: 0.0068 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0067 decode.acc_seg: 99.6999 aux.loss_ce: 0.0074 aux.acc_seg: 99.2559 +04/18 22:29:27 - mmengine - INFO - Iter(train) [130650/160000] lr: 2.2517e-03 eta: 4:30:34 time: 0.5567 data_time: 0.0067 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0066 decode.acc_seg: 99.7292 aux.loss_ce: 0.0073 aux.acc_seg: 99.2261 +04/18 22:29:55 - mmengine - INFO - Iter(train) [130700/160000] lr: 2.2484e-03 eta: 4:30:06 time: 0.5557 data_time: 0.0072 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0067 decode.acc_seg: 99.6858 aux.loss_ce: 0.0074 aux.acc_seg: 99.2261 +04/18 22:30:22 - mmengine - INFO - Iter(train) [130750/160000] lr: 2.2451e-03 eta: 4:29:39 time: 0.5547 data_time: 0.0068 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0069 decode.acc_seg: 99.7329 aux.loss_ce: 0.0073 aux.acc_seg: 99.4963 +04/18 22:30:50 - mmengine - INFO - Iter(train) [130800/160000] lr: 2.2418e-03 eta: 4:29:11 time: 0.5563 data_time: 0.0070 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0067 decode.acc_seg: 99.6983 aux.loss_ce: 0.0075 aux.acc_seg: 99.2035 +04/18 22:31:18 - mmengine - INFO - Iter(train) [130850/160000] lr: 2.2385e-03 eta: 4:28:43 time: 0.5548 data_time: 0.0061 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0067 decode.acc_seg: 99.7300 aux.loss_ce: 0.0070 aux.acc_seg: 99.1772 +04/18 22:31:46 - mmengine - INFO - Iter(train) [130900/160000] lr: 2.2352e-03 eta: 4:28:16 time: 0.5576 data_time: 0.0063 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0067 decode.acc_seg: 99.8247 aux.loss_ce: 0.0072 aux.acc_seg: 99.4732 +04/18 22:32:14 - mmengine - INFO - Iter(train) [130950/160000] lr: 2.2319e-03 eta: 4:27:48 time: 0.5563 data_time: 0.0066 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0062 decode.acc_seg: 99.7023 aux.loss_ce: 0.0068 aux.acc_seg: 99.1722 +04/18 22:32:42 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 22:32:42 - mmengine - INFO - Iter(train) [131000/160000] lr: 2.2286e-03 eta: 4:27:20 time: 0.5551 data_time: 0.0063 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0068 decode.acc_seg: 99.7419 aux.loss_ce: 0.0077 aux.acc_seg: 99.1072 +04/18 22:33:09 - mmengine - INFO - Iter(train) [131050/160000] lr: 2.2253e-03 eta: 4:26:53 time: 0.5554 data_time: 0.0062 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0066 decode.acc_seg: 99.6530 aux.loss_ce: 0.0072 aux.acc_seg: 98.9323 +04/18 22:33:37 - mmengine - INFO - Iter(train) [131100/160000] lr: 2.2220e-03 eta: 4:26:25 time: 0.5562 data_time: 0.0073 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0064 decode.acc_seg: 99.6785 aux.loss_ce: 0.0070 aux.acc_seg: 99.1549 +04/18 22:34:05 - mmengine - INFO - Iter(train) [131150/160000] lr: 2.2187e-03 eta: 4:25:58 time: 0.5554 data_time: 0.0068 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0066 decode.acc_seg: 99.7717 aux.loss_ce: 0.0073 aux.acc_seg: 99.3612 +04/18 22:34:33 - mmengine - INFO - Iter(train) [131200/160000] lr: 2.2154e-03 eta: 4:25:30 time: 0.5556 data_time: 0.0068 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0065 decode.acc_seg: 99.8078 aux.loss_ce: 0.0073 aux.acc_seg: 99.1734 +04/18 22:35:00 - mmengine - INFO - Iter(train) [131250/160000] lr: 2.2120e-03 eta: 4:25:02 time: 0.5584 data_time: 0.0064 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0068 decode.acc_seg: 99.7023 aux.loss_ce: 0.0077 aux.acc_seg: 99.0439 +04/18 22:35:28 - mmengine - INFO - Iter(train) [131300/160000] lr: 2.2087e-03 eta: 4:24:35 time: 0.5549 data_time: 0.0073 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0062 decode.acc_seg: 99.7127 aux.loss_ce: 0.0068 aux.acc_seg: 99.1218 +04/18 22:35:56 - mmengine - INFO - Iter(train) [131350/160000] lr: 2.2054e-03 eta: 4:24:07 time: 0.5555 data_time: 0.0070 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0066 decode.acc_seg: 99.7238 aux.loss_ce: 0.0077 aux.acc_seg: 99.0662 +04/18 22:36:24 - mmengine - INFO - Iter(train) [131400/160000] lr: 2.2021e-03 eta: 4:23:39 time: 0.5559 data_time: 0.0065 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0066 decode.acc_seg: 99.7224 aux.loss_ce: 0.0074 aux.acc_seg: 99.1974 +04/18 22:36:52 - mmengine - INFO - Iter(train) [131450/160000] lr: 2.1988e-03 eta: 4:23:12 time: 0.5570 data_time: 0.0071 memory: 7635 loss: 0.0131 decode.loss_ce: 0.0062 decode.acc_seg: 99.7651 aux.loss_ce: 0.0069 aux.acc_seg: 99.3143 +04/18 22:37:20 - mmengine - INFO - Iter(train) [131500/160000] lr: 2.1955e-03 eta: 4:22:44 time: 0.5560 data_time: 0.0073 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0064 decode.acc_seg: 99.7942 aux.loss_ce: 0.0071 aux.acc_seg: 99.2578 +04/18 22:37:48 - mmengine - INFO - Iter(train) [131550/160000] lr: 2.1922e-03 eta: 4:22:17 time: 0.5547 data_time: 0.0069 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0062 decode.acc_seg: 99.8041 aux.loss_ce: 0.0068 aux.acc_seg: 99.4134 +04/18 22:38:15 - mmengine - INFO - Iter(train) [131600/160000] lr: 2.1889e-03 eta: 4:21:49 time: 0.5573 data_time: 0.0070 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0066 decode.acc_seg: 99.7626 aux.loss_ce: 0.0073 aux.acc_seg: 99.3815 +04/18 22:38:43 - mmengine - INFO - Iter(train) [131650/160000] lr: 2.1856e-03 eta: 4:21:21 time: 0.5554 data_time: 0.0072 memory: 7635 loss: 0.0131 decode.loss_ce: 0.0061 decode.acc_seg: 99.8192 aux.loss_ce: 0.0069 aux.acc_seg: 99.4833 +04/18 22:39:11 - mmengine - INFO - Iter(train) [131700/160000] lr: 2.1823e-03 eta: 4:20:54 time: 0.5558 data_time: 0.0061 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0064 decode.acc_seg: 99.7510 aux.loss_ce: 0.0077 aux.acc_seg: 99.0620 +04/18 22:39:39 - mmengine - INFO - Iter(train) [131750/160000] lr: 2.1790e-03 eta: 4:20:26 time: 0.5547 data_time: 0.0062 memory: 7635 loss: 0.0132 decode.loss_ce: 0.0061 decode.acc_seg: 99.7917 aux.loss_ce: 0.0071 aux.acc_seg: 99.4679 +04/18 22:40:07 - mmengine - INFO - Iter(train) [131800/160000] lr: 2.1756e-03 eta: 4:19:58 time: 0.5557 data_time: 0.0064 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0065 decode.acc_seg: 99.7558 aux.loss_ce: 0.0072 aux.acc_seg: 99.3421 +04/18 22:40:34 - mmengine - INFO - Iter(train) [131850/160000] lr: 2.1723e-03 eta: 4:19:31 time: 0.5559 data_time: 0.0067 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0065 decode.acc_seg: 99.7120 aux.loss_ce: 0.0071 aux.acc_seg: 99.1834 +04/18 22:41:02 - mmengine - INFO - Iter(train) [131900/160000] lr: 2.1690e-03 eta: 4:19:03 time: 0.5555 data_time: 0.0069 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0067 decode.acc_seg: 99.6838 aux.loss_ce: 0.0073 aux.acc_seg: 98.8683 +04/18 22:41:30 - mmengine - INFO - Iter(train) [131950/160000] lr: 2.1657e-03 eta: 4:18:35 time: 0.5565 data_time: 0.0063 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0063 decode.acc_seg: 99.7451 aux.loss_ce: 0.0071 aux.acc_seg: 99.2819 +04/18 22:41:58 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 22:41:58 - mmengine - INFO - Iter(train) [132000/160000] lr: 2.1624e-03 eta: 4:18:08 time: 0.5566 data_time: 0.0067 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0065 decode.acc_seg: 99.7520 aux.loss_ce: 0.0074 aux.acc_seg: 99.1823 +04/18 22:42:25 - mmengine - INFO - Iter(train) [132050/160000] lr: 2.1591e-03 eta: 4:17:40 time: 0.5552 data_time: 0.0066 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0063 decode.acc_seg: 99.7133 aux.loss_ce: 0.0071 aux.acc_seg: 99.2467 +04/18 22:42:53 - mmengine - INFO - Iter(train) [132100/160000] lr: 2.1558e-03 eta: 4:17:13 time: 0.5542 data_time: 0.0066 memory: 7635 loss: 0.0126 decode.loss_ce: 0.0059 decode.acc_seg: 99.7382 aux.loss_ce: 0.0067 aux.acc_seg: 99.0999 +04/18 22:43:21 - mmengine - INFO - Iter(train) [132150/160000] lr: 2.1524e-03 eta: 4:16:45 time: 0.5556 data_time: 0.0065 memory: 7635 loss: 0.0129 decode.loss_ce: 0.0059 decode.acc_seg: 99.7906 aux.loss_ce: 0.0070 aux.acc_seg: 99.3847 +04/18 22:43:49 - mmengine - INFO - Iter(train) [132200/160000] lr: 2.1491e-03 eta: 4:16:17 time: 0.5571 data_time: 0.0072 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0066 decode.acc_seg: 99.6938 aux.loss_ce: 0.0074 aux.acc_seg: 99.0777 +04/18 22:44:17 - mmengine - INFO - Iter(train) [132250/160000] lr: 2.1458e-03 eta: 4:15:50 time: 0.5570 data_time: 0.0067 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0071 decode.acc_seg: 99.7658 aux.loss_ce: 0.0080 aux.acc_seg: 99.4457 +04/18 22:44:44 - mmengine - INFO - Iter(train) [132300/160000] lr: 2.1425e-03 eta: 4:15:22 time: 0.5560 data_time: 0.0071 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0065 decode.acc_seg: 99.7627 aux.loss_ce: 0.0073 aux.acc_seg: 99.3998 +04/18 22:45:12 - mmengine - INFO - Iter(train) [132350/160000] lr: 2.1392e-03 eta: 4:14:54 time: 0.5563 data_time: 0.0077 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0066 decode.acc_seg: 99.7610 aux.loss_ce: 0.0075 aux.acc_seg: 99.2947 +04/18 22:45:40 - mmengine - INFO - Iter(train) [132400/160000] lr: 2.1359e-03 eta: 4:14:27 time: 0.5549 data_time: 0.0065 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0063 decode.acc_seg: 99.7573 aux.loss_ce: 0.0071 aux.acc_seg: 99.2528 +04/18 22:46:08 - mmengine - INFO - Iter(train) [132450/160000] lr: 2.1325e-03 eta: 4:13:59 time: 0.5559 data_time: 0.0075 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0066 decode.acc_seg: 99.8019 aux.loss_ce: 0.0072 aux.acc_seg: 99.3565 +04/18 22:46:36 - mmengine - INFO - Iter(train) [132500/160000] lr: 2.1292e-03 eta: 4:13:32 time: 0.5549 data_time: 0.0079 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0068 decode.acc_seg: 99.6260 aux.loss_ce: 0.0075 aux.acc_seg: 99.0879 +04/18 22:47:03 - mmengine - INFO - Iter(train) [132550/160000] lr: 2.1259e-03 eta: 4:13:04 time: 0.5545 data_time: 0.0066 memory: 7635 loss: 0.0128 decode.loss_ce: 0.0060 decode.acc_seg: 99.7648 aux.loss_ce: 0.0068 aux.acc_seg: 99.2874 +04/18 22:47:31 - mmengine - INFO - Iter(train) [132600/160000] lr: 2.1226e-03 eta: 4:12:36 time: 0.5560 data_time: 0.0077 memory: 7635 loss: 0.0148 decode.loss_ce: 0.0069 decode.acc_seg: 99.7509 aux.loss_ce: 0.0079 aux.acc_seg: 99.0533 +04/18 22:47:59 - mmengine - INFO - Iter(train) [132650/160000] lr: 2.1193e-03 eta: 4:12:09 time: 0.5552 data_time: 0.0063 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0061 decode.acc_seg: 99.5701 aux.loss_ce: 0.0072 aux.acc_seg: 98.5176 +04/18 22:48:27 - mmengine - INFO - Iter(train) [132700/160000] lr: 2.1159e-03 eta: 4:11:41 time: 0.5540 data_time: 0.0068 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0063 decode.acc_seg: 99.7020 aux.loss_ce: 0.0071 aux.acc_seg: 99.2132 +04/18 22:48:54 - mmengine - INFO - Iter(train) [132750/160000] lr: 2.1126e-03 eta: 4:11:13 time: 0.5557 data_time: 0.0076 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0063 decode.acc_seg: 99.7324 aux.loss_ce: 0.0074 aux.acc_seg: 99.2095 +04/18 22:49:22 - mmengine - INFO - Iter(train) [132800/160000] lr: 2.1093e-03 eta: 4:10:46 time: 0.5550 data_time: 0.0072 memory: 7635 loss: 0.0129 decode.loss_ce: 0.0060 decode.acc_seg: 99.7230 aux.loss_ce: 0.0069 aux.acc_seg: 99.1921 +04/18 22:49:50 - mmengine - INFO - Iter(train) [132850/160000] lr: 2.1060e-03 eta: 4:10:18 time: 0.5532 data_time: 0.0070 memory: 7635 loss: 0.0125 decode.loss_ce: 0.0061 decode.acc_seg: 99.7101 aux.loss_ce: 0.0064 aux.acc_seg: 99.2487 +04/18 22:50:18 - mmengine - INFO - Iter(train) [132900/160000] lr: 2.1026e-03 eta: 4:09:50 time: 0.5528 data_time: 0.0074 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0062 decode.acc_seg: 99.6808 aux.loss_ce: 0.0071 aux.acc_seg: 99.2229 +04/18 22:50:45 - mmengine - INFO - Iter(train) [132950/160000] lr: 2.0993e-03 eta: 4:09:23 time: 0.5546 data_time: 0.0073 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0065 decode.acc_seg: 99.7187 aux.loss_ce: 0.0074 aux.acc_seg: 99.2344 +04/18 22:51:13 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 22:51:13 - mmengine - INFO - Iter(train) [133000/160000] lr: 2.0960e-03 eta: 4:08:55 time: 0.5545 data_time: 0.0064 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0075 decode.acc_seg: 99.7322 aux.loss_ce: 0.0076 aux.acc_seg: 99.3469 +04/18 22:51:41 - mmengine - INFO - Iter(train) [133050/160000] lr: 2.0927e-03 eta: 4:08:27 time: 0.5533 data_time: 0.0069 memory: 7635 loss: 0.0161 decode.loss_ce: 0.0082 decode.acc_seg: 99.7278 aux.loss_ce: 0.0079 aux.acc_seg: 99.2898 +04/18 22:52:08 - mmengine - INFO - Iter(train) [133100/160000] lr: 2.0893e-03 eta: 4:08:00 time: 0.5540 data_time: 0.0075 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0072 decode.acc_seg: 99.7031 aux.loss_ce: 0.0072 aux.acc_seg: 99.2740 +04/18 22:52:36 - mmengine - INFO - Iter(train) [133150/160000] lr: 2.0860e-03 eta: 4:07:32 time: 0.5536 data_time: 0.0074 memory: 7635 loss: 0.0159 decode.loss_ce: 0.0079 decode.acc_seg: 99.7276 aux.loss_ce: 0.0080 aux.acc_seg: 99.3108 +04/18 22:53:04 - mmengine - INFO - Iter(train) [133200/160000] lr: 2.0827e-03 eta: 4:07:04 time: 0.5529 data_time: 0.0068 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0067 decode.acc_seg: 99.6853 aux.loss_ce: 0.0074 aux.acc_seg: 99.2971 +04/18 22:53:32 - mmengine - INFO - Iter(train) [133250/160000] lr: 2.0793e-03 eta: 4:06:37 time: 0.5532 data_time: 0.0064 memory: 7635 loss: 0.0172 decode.loss_ce: 0.0085 decode.acc_seg: 99.5847 aux.loss_ce: 0.0087 aux.acc_seg: 99.1069 +04/18 22:53:59 - mmengine - INFO - Iter(train) [133300/160000] lr: 2.0760e-03 eta: 4:06:09 time: 0.5533 data_time: 0.0071 memory: 7635 loss: 0.0151 decode.loss_ce: 0.0076 decode.acc_seg: 99.6610 aux.loss_ce: 0.0075 aux.acc_seg: 99.1081 +04/18 22:54:27 - mmengine - INFO - Iter(train) [133350/160000] lr: 2.0727e-03 eta: 4:05:42 time: 0.5559 data_time: 0.0066 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0071 decode.acc_seg: 99.7018 aux.loss_ce: 0.0074 aux.acc_seg: 99.1064 +04/18 22:54:55 - mmengine - INFO - Iter(train) [133400/160000] lr: 2.0694e-03 eta: 4:05:14 time: 0.5542 data_time: 0.0078 memory: 7635 loss: 0.0155 decode.loss_ce: 0.0074 decode.acc_seg: 99.6956 aux.loss_ce: 0.0081 aux.acc_seg: 99.1029 +04/18 22:55:22 - mmengine - INFO - Iter(train) [133450/160000] lr: 2.0660e-03 eta: 4:04:46 time: 0.5521 data_time: 0.0069 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0065 decode.acc_seg: 99.7554 aux.loss_ce: 0.0075 aux.acc_seg: 99.3796 +04/18 22:55:50 - mmengine - INFO - Iter(train) [133500/160000] lr: 2.0627e-03 eta: 4:04:19 time: 0.5527 data_time: 0.0062 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0070 decode.acc_seg: 99.6744 aux.loss_ce: 0.0075 aux.acc_seg: 99.0935 +04/18 22:56:18 - mmengine - INFO - Iter(train) [133550/160000] lr: 2.0594e-03 eta: 4:03:51 time: 0.5545 data_time: 0.0065 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0064 decode.acc_seg: 99.7576 aux.loss_ce: 0.0072 aux.acc_seg: 99.2503 +04/18 22:56:46 - mmengine - INFO - Iter(train) [133600/160000] lr: 2.0560e-03 eta: 4:03:23 time: 0.5538 data_time: 0.0062 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0072 decode.acc_seg: 99.6710 aux.loss_ce: 0.0078 aux.acc_seg: 99.2356 +04/18 22:57:13 - mmengine - INFO - Iter(train) [133650/160000] lr: 2.0527e-03 eta: 4:02:56 time: 0.5622 data_time: 0.0067 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0069 decode.acc_seg: 99.7643 aux.loss_ce: 0.0077 aux.acc_seg: 99.2272 +04/18 22:57:41 - mmengine - INFO - Iter(train) [133700/160000] lr: 2.0494e-03 eta: 4:02:28 time: 0.5530 data_time: 0.0063 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0065 decode.acc_seg: 99.6926 aux.loss_ce: 0.0069 aux.acc_seg: 99.0005 +04/18 22:58:09 - mmengine - INFO - Iter(train) [133750/160000] lr: 2.0460e-03 eta: 4:02:00 time: 0.5539 data_time: 0.0067 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0070 decode.acc_seg: 99.7770 aux.loss_ce: 0.0071 aux.acc_seg: 99.2735 +04/18 22:58:36 - mmengine - INFO - Iter(train) [133800/160000] lr: 2.0427e-03 eta: 4:01:33 time: 0.5540 data_time: 0.0068 memory: 7635 loss: 0.0145 decode.loss_ce: 0.0073 decode.acc_seg: 99.6379 aux.loss_ce: 0.0073 aux.acc_seg: 99.2218 +04/18 22:59:04 - mmengine - INFO - Iter(train) [133850/160000] lr: 2.0393e-03 eta: 4:01:05 time: 0.5537 data_time: 0.0075 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0068 decode.acc_seg: 99.7842 aux.loss_ce: 0.0073 aux.acc_seg: 99.4088 +04/18 22:59:32 - mmengine - INFO - Iter(train) [133900/160000] lr: 2.0360e-03 eta: 4:00:37 time: 0.5528 data_time: 0.0065 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0067 decode.acc_seg: 99.7386 aux.loss_ce: 0.0072 aux.acc_seg: 99.2549 +04/18 22:59:59 - mmengine - INFO - Iter(train) [133950/160000] lr: 2.0327e-03 eta: 4:00:10 time: 0.5541 data_time: 0.0071 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0068 decode.acc_seg: 99.7540 aux.loss_ce: 0.0072 aux.acc_seg: 99.4673 +04/18 23:00:27 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 23:00:27 - mmengine - INFO - Iter(train) [134000/160000] lr: 2.0293e-03 eta: 3:59:42 time: 0.5534 data_time: 0.0068 memory: 7635 loss: 0.0147 decode.loss_ce: 0.0072 decode.acc_seg: 99.7256 aux.loss_ce: 0.0075 aux.acc_seg: 99.2776 +04/18 23:00:55 - mmengine - INFO - Iter(train) [134050/160000] lr: 2.0260e-03 eta: 3:59:14 time: 0.5534 data_time: 0.0066 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0068 decode.acc_seg: 99.7190 aux.loss_ce: 0.0075 aux.acc_seg: 99.4332 +04/18 23:01:23 - mmengine - INFO - Iter(train) [134100/160000] lr: 2.0226e-03 eta: 3:58:47 time: 0.5530 data_time: 0.0071 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0068 decode.acc_seg: 99.7502 aux.loss_ce: 0.0070 aux.acc_seg: 99.2365 +04/18 23:01:50 - mmengine - INFO - Iter(train) [134150/160000] lr: 2.0193e-03 eta: 3:58:19 time: 0.5527 data_time: 0.0074 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0066 decode.acc_seg: 99.7362 aux.loss_ce: 0.0071 aux.acc_seg: 99.1766 +04/18 23:02:18 - mmengine - INFO - Iter(train) [134200/160000] lr: 2.0160e-03 eta: 3:57:51 time: 0.5536 data_time: 0.0064 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0065 decode.acc_seg: 99.7310 aux.loss_ce: 0.0071 aux.acc_seg: 99.1065 +04/18 23:02:46 - mmengine - INFO - Iter(train) [134250/160000] lr: 2.0126e-03 eta: 3:57:24 time: 0.5535 data_time: 0.0072 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0066 decode.acc_seg: 99.8188 aux.loss_ce: 0.0075 aux.acc_seg: 99.2989 +04/18 23:03:13 - mmengine - INFO - Iter(train) [134300/160000] lr: 2.0093e-03 eta: 3:56:56 time: 0.5540 data_time: 0.0072 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0067 decode.acc_seg: 99.7257 aux.loss_ce: 0.0070 aux.acc_seg: 99.3756 +04/18 23:03:41 - mmengine - INFO - Iter(train) [134350/160000] lr: 2.0059e-03 eta: 3:56:28 time: 0.5522 data_time: 0.0067 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0066 decode.acc_seg: 99.7658 aux.loss_ce: 0.0071 aux.acc_seg: 99.3269 +04/18 23:04:09 - mmengine - INFO - Iter(train) [134400/160000] lr: 2.0026e-03 eta: 3:56:01 time: 0.5527 data_time: 0.0071 memory: 7635 loss: 0.0131 decode.loss_ce: 0.0064 decode.acc_seg: 99.7603 aux.loss_ce: 0.0068 aux.acc_seg: 99.5281 +04/18 23:04:36 - mmengine - INFO - Iter(train) [134450/160000] lr: 1.9992e-03 eta: 3:55:33 time: 0.5536 data_time: 0.0074 memory: 7635 loss: 0.0131 decode.loss_ce: 0.0063 decode.acc_seg: 99.8034 aux.loss_ce: 0.0068 aux.acc_seg: 99.4308 +04/18 23:05:04 - mmengine - INFO - Iter(train) [134500/160000] lr: 1.9959e-03 eta: 3:55:05 time: 0.5532 data_time: 0.0074 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0072 decode.acc_seg: 99.7137 aux.loss_ce: 0.0077 aux.acc_seg: 99.3555 +04/18 23:05:32 - mmengine - INFO - Iter(train) [134550/160000] lr: 1.9926e-03 eta: 3:54:38 time: 0.5527 data_time: 0.0071 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0066 decode.acc_seg: 99.7665 aux.loss_ce: 0.0075 aux.acc_seg: 99.2996 +04/18 23:05:59 - mmengine - INFO - Iter(train) [134600/160000] lr: 1.9892e-03 eta: 3:54:10 time: 0.5518 data_time: 0.0063 memory: 7635 loss: 0.0146 decode.loss_ce: 0.0070 decode.acc_seg: 99.7805 aux.loss_ce: 0.0076 aux.acc_seg: 99.3905 +04/18 23:06:27 - mmengine - INFO - Iter(train) [134650/160000] lr: 1.9859e-03 eta: 3:53:43 time: 0.5613 data_time: 0.0071 memory: 7635 loss: 0.0127 decode.loss_ce: 0.0061 decode.acc_seg: 99.8212 aux.loss_ce: 0.0066 aux.acc_seg: 99.4777 +04/18 23:06:55 - mmengine - INFO - Iter(train) [134700/160000] lr: 1.9825e-03 eta: 3:53:15 time: 0.5635 data_time: 0.0066 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0065 decode.acc_seg: 99.6467 aux.loss_ce: 0.0073 aux.acc_seg: 98.8048 +04/18 23:07:22 - mmengine - INFO - Iter(train) [134750/160000] lr: 1.9792e-03 eta: 3:52:47 time: 0.5521 data_time: 0.0066 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0064 decode.acc_seg: 99.6914 aux.loss_ce: 0.0071 aux.acc_seg: 99.1985 +04/18 23:07:50 - mmengine - INFO - Iter(train) [134800/160000] lr: 1.9758e-03 eta: 3:52:20 time: 0.5546 data_time: 0.0069 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0066 decode.acc_seg: 99.6202 aux.loss_ce: 0.0072 aux.acc_seg: 99.2083 +04/18 23:08:18 - mmengine - INFO - Iter(train) [134850/160000] lr: 1.9725e-03 eta: 3:51:52 time: 0.5513 data_time: 0.0070 memory: 7635 loss: 0.0128 decode.loss_ce: 0.0060 decode.acc_seg: 99.7522 aux.loss_ce: 0.0068 aux.acc_seg: 99.2096 +04/18 23:08:45 - mmengine - INFO - Iter(train) [134900/160000] lr: 1.9691e-03 eta: 3:51:24 time: 0.5533 data_time: 0.0063 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0065 decode.acc_seg: 99.7433 aux.loss_ce: 0.0072 aux.acc_seg: 99.1336 +04/18 23:09:13 - mmengine - INFO - Iter(train) [134950/160000] lr: 1.9658e-03 eta: 3:50:57 time: 0.5510 data_time: 0.0064 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0064 decode.acc_seg: 99.7322 aux.loss_ce: 0.0073 aux.acc_seg: 99.1809 +04/18 23:09:41 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 23:09:41 - mmengine - INFO - Iter(train) [135000/160000] lr: 1.9624e-03 eta: 3:50:29 time: 0.5533 data_time: 0.0067 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0064 decode.acc_seg: 99.7094 aux.loss_ce: 0.0071 aux.acc_seg: 99.1610 +04/18 23:10:08 - mmengine - INFO - Iter(train) [135050/160000] lr: 1.9591e-03 eta: 3:50:01 time: 0.5515 data_time: 0.0068 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0065 decode.acc_seg: 99.7212 aux.loss_ce: 0.0069 aux.acc_seg: 99.3560 +04/18 23:10:36 - mmengine - INFO - Iter(train) [135100/160000] lr: 1.9557e-03 eta: 3:49:34 time: 0.5521 data_time: 0.0062 memory: 7635 loss: 0.0128 decode.loss_ce: 0.0060 decode.acc_seg: 99.6983 aux.loss_ce: 0.0068 aux.acc_seg: 99.0344 +04/18 23:11:04 - mmengine - INFO - Iter(train) [135150/160000] lr: 1.9524e-03 eta: 3:49:06 time: 0.5527 data_time: 0.0064 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0064 decode.acc_seg: 99.7859 aux.loss_ce: 0.0072 aux.acc_seg: 99.3858 +04/18 23:11:31 - mmengine - INFO - Iter(train) [135200/160000] lr: 1.9490e-03 eta: 3:48:38 time: 0.5522 data_time: 0.0065 memory: 7635 loss: 0.0149 decode.loss_ce: 0.0072 decode.acc_seg: 99.6347 aux.loss_ce: 0.0077 aux.acc_seg: 99.1165 +04/18 23:11:59 - mmengine - INFO - Iter(train) [135250/160000] lr: 1.9456e-03 eta: 3:48:11 time: 0.5526 data_time: 0.0061 memory: 7635 loss: 0.0132 decode.loss_ce: 0.0064 decode.acc_seg: 99.7595 aux.loss_ce: 0.0068 aux.acc_seg: 99.3470 +04/18 23:12:27 - mmengine - INFO - Iter(train) [135300/160000] lr: 1.9423e-03 eta: 3:47:43 time: 0.5524 data_time: 0.0076 memory: 7635 loss: 0.0128 decode.loss_ce: 0.0061 decode.acc_seg: 99.7675 aux.loss_ce: 0.0067 aux.acc_seg: 99.2603 +04/18 23:12:54 - mmengine - INFO - Iter(train) [135350/160000] lr: 1.9389e-03 eta: 3:47:15 time: 0.5531 data_time: 0.0077 memory: 7635 loss: 0.0126 decode.loss_ce: 0.0058 decode.acc_seg: 99.8325 aux.loss_ce: 0.0067 aux.acc_seg: 99.3667 +04/18 23:13:22 - mmengine - INFO - Iter(train) [135400/160000] lr: 1.9356e-03 eta: 3:46:48 time: 0.5520 data_time: 0.0060 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0068 decode.acc_seg: 99.7214 aux.loss_ce: 0.0073 aux.acc_seg: 99.1510 +04/18 23:13:49 - mmengine - INFO - Iter(train) [135450/160000] lr: 1.9322e-03 eta: 3:46:20 time: 0.5524 data_time: 0.0064 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0062 decode.acc_seg: 99.7041 aux.loss_ce: 0.0070 aux.acc_seg: 99.0205 +04/18 23:14:17 - mmengine - INFO - Iter(train) [135500/160000] lr: 1.9289e-03 eta: 3:45:52 time: 0.5527 data_time: 0.0069 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0060 decode.acc_seg: 99.8290 aux.loss_ce: 0.0070 aux.acc_seg: 99.3267 +04/18 23:14:45 - mmengine - INFO - Iter(train) [135550/160000] lr: 1.9255e-03 eta: 3:45:25 time: 0.5524 data_time: 0.0071 memory: 7635 loss: 0.0126 decode.loss_ce: 0.0058 decode.acc_seg: 99.7816 aux.loss_ce: 0.0068 aux.acc_seg: 99.3285 +04/18 23:15:12 - mmengine - INFO - Iter(train) [135600/160000] lr: 1.9221e-03 eta: 3:44:57 time: 0.5514 data_time: 0.0064 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0065 decode.acc_seg: 99.7851 aux.loss_ce: 0.0073 aux.acc_seg: 99.2583 +04/18 23:15:40 - mmengine - INFO - Iter(train) [135650/160000] lr: 1.9188e-03 eta: 3:44:29 time: 0.5526 data_time: 0.0063 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0069 decode.acc_seg: 99.7872 aux.loss_ce: 0.0070 aux.acc_seg: 99.4615 +04/18 23:16:08 - mmengine - INFO - Iter(train) [135700/160000] lr: 1.9154e-03 eta: 3:44:02 time: 0.5614 data_time: 0.0072 memory: 7635 loss: 0.0132 decode.loss_ce: 0.0063 decode.acc_seg: 99.7512 aux.loss_ce: 0.0069 aux.acc_seg: 99.3599 +04/18 23:16:35 - mmengine - INFO - Iter(train) [135750/160000] lr: 1.9121e-03 eta: 3:43:34 time: 0.5516 data_time: 0.0064 memory: 7635 loss: 0.0131 decode.loss_ce: 0.0060 decode.acc_seg: 99.7575 aux.loss_ce: 0.0071 aux.acc_seg: 99.1708 +04/18 23:17:03 - mmengine - INFO - Iter(train) [135800/160000] lr: 1.9087e-03 eta: 3:43:06 time: 0.5609 data_time: 0.0068 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0068 decode.acc_seg: 99.7773 aux.loss_ce: 0.0072 aux.acc_seg: 99.3564 +04/18 23:17:31 - mmengine - INFO - Iter(train) [135850/160000] lr: 1.9053e-03 eta: 3:42:39 time: 0.5517 data_time: 0.0066 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0066 decode.acc_seg: 99.6757 aux.loss_ce: 0.0074 aux.acc_seg: 98.9448 +04/18 23:17:58 - mmengine - INFO - Iter(train) [135900/160000] lr: 1.9020e-03 eta: 3:42:11 time: 0.5522 data_time: 0.0070 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0065 decode.acc_seg: 99.8109 aux.loss_ce: 0.0072 aux.acc_seg: 99.4719 +04/18 23:18:26 - mmengine - INFO - Iter(train) [135950/160000] lr: 1.8986e-03 eta: 3:41:43 time: 0.5509 data_time: 0.0069 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0061 decode.acc_seg: 99.7066 aux.loss_ce: 0.0069 aux.acc_seg: 99.1639 +04/18 23:18:53 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 23:18:53 - mmengine - INFO - Iter(train) [136000/160000] lr: 1.8952e-03 eta: 3:41:16 time: 0.5521 data_time: 0.0069 memory: 7635 loss: 0.0133 decode.loss_ce: 0.0063 decode.acc_seg: 99.7808 aux.loss_ce: 0.0070 aux.acc_seg: 99.3621 +04/18 23:19:21 - mmengine - INFO - Iter(train) [136050/160000] lr: 1.8919e-03 eta: 3:40:48 time: 0.5526 data_time: 0.0066 memory: 7635 loss: 0.0129 decode.loss_ce: 0.0062 decode.acc_seg: 99.7650 aux.loss_ce: 0.0067 aux.acc_seg: 99.3467 +04/18 23:19:49 - mmengine - INFO - Iter(train) [136100/160000] lr: 1.8885e-03 eta: 3:40:20 time: 0.5530 data_time: 0.0072 memory: 7635 loss: 0.0137 decode.loss_ce: 0.0065 decode.acc_seg: 99.6713 aux.loss_ce: 0.0073 aux.acc_seg: 99.1525 +04/18 23:20:16 - mmengine - INFO - Iter(train) [136150/160000] lr: 1.8851e-03 eta: 3:39:53 time: 0.5518 data_time: 0.0071 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0063 decode.acc_seg: 99.7138 aux.loss_ce: 0.0072 aux.acc_seg: 99.2328 +04/18 23:20:44 - mmengine - INFO - Iter(train) [136200/160000] lr: 1.8818e-03 eta: 3:39:25 time: 0.5516 data_time: 0.0071 memory: 7635 loss: 0.0124 decode.loss_ce: 0.0058 decode.acc_seg: 99.8105 aux.loss_ce: 0.0066 aux.acc_seg: 99.4086 +04/18 23:21:12 - mmengine - INFO - Iter(train) [136250/160000] lr: 1.8784e-03 eta: 3:38:57 time: 0.5508 data_time: 0.0069 memory: 7635 loss: 0.0132 decode.loss_ce: 0.0062 decode.acc_seg: 99.7879 aux.loss_ce: 0.0070 aux.acc_seg: 99.4346 +04/18 23:21:39 - mmengine - INFO - Iter(train) [136300/160000] lr: 1.8750e-03 eta: 3:38:30 time: 0.5519 data_time: 0.0076 memory: 7635 loss: 0.0141 decode.loss_ce: 0.0068 decode.acc_seg: 99.6236 aux.loss_ce: 0.0073 aux.acc_seg: 98.8398 +04/18 23:22:07 - mmengine - INFO - Iter(train) [136350/160000] lr: 1.8716e-03 eta: 3:38:02 time: 0.5534 data_time: 0.0063 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0063 decode.acc_seg: 99.7940 aux.loss_ce: 0.0073 aux.acc_seg: 99.3380 +04/18 23:22:34 - mmengine - INFO - Iter(train) [136400/160000] lr: 1.8683e-03 eta: 3:37:34 time: 0.5536 data_time: 0.0069 memory: 7635 loss: 0.0143 decode.loss_ce: 0.0067 decode.acc_seg: 99.8033 aux.loss_ce: 0.0076 aux.acc_seg: 99.5088 +04/18 23:23:02 - mmengine - INFO - Iter(train) [136450/160000] lr: 1.8649e-03 eta: 3:37:07 time: 0.5525 data_time: 0.0070 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0062 decode.acc_seg: 99.7760 aux.loss_ce: 0.0071 aux.acc_seg: 99.1981 +04/18 23:23:30 - mmengine - INFO - Iter(train) [136500/160000] lr: 1.8615e-03 eta: 3:36:39 time: 0.5528 data_time: 0.0067 memory: 7635 loss: 0.0126 decode.loss_ce: 0.0058 decode.acc_seg: 99.7166 aux.loss_ce: 0.0068 aux.acc_seg: 99.2714 +04/18 23:23:57 - mmengine - INFO - Iter(train) [136550/160000] lr: 1.8582e-03 eta: 3:36:11 time: 0.5515 data_time: 0.0070 memory: 7635 loss: 0.0135 decode.loss_ce: 0.0063 decode.acc_seg: 99.7321 aux.loss_ce: 0.0072 aux.acc_seg: 99.1710 +04/18 23:24:25 - mmengine - INFO - Iter(train) [136600/160000] lr: 1.8548e-03 eta: 3:35:44 time: 0.5525 data_time: 0.0077 memory: 7635 loss: 0.0132 decode.loss_ce: 0.0063 decode.acc_seg: 99.7075 aux.loss_ce: 0.0070 aux.acc_seg: 99.1806 +04/18 23:24:53 - mmengine - INFO - Iter(train) [136650/160000] lr: 1.8514e-03 eta: 3:35:16 time: 0.5527 data_time: 0.0067 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0062 decode.acc_seg: 99.7082 aux.loss_ce: 0.0068 aux.acc_seg: 99.4295 +04/18 23:25:20 - mmengine - INFO - Iter(train) [136700/160000] lr: 1.8480e-03 eta: 3:34:48 time: 0.5517 data_time: 0.0067 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0070 decode.acc_seg: 99.6579 aux.loss_ce: 0.0070 aux.acc_seg: 99.0955 +04/18 23:25:48 - mmengine - INFO - Iter(train) [136750/160000] lr: 1.8447e-03 eta: 3:34:21 time: 0.5505 data_time: 0.0071 memory: 7635 loss: 0.0144 decode.loss_ce: 0.0067 decode.acc_seg: 99.7284 aux.loss_ce: 0.0077 aux.acc_seg: 99.1408 +04/18 23:26:16 - mmengine - INFO - Iter(train) [136800/160000] lr: 1.8413e-03 eta: 3:33:53 time: 0.5600 data_time: 0.0066 memory: 7635 loss: 0.0125 decode.loss_ce: 0.0060 decode.acc_seg: 99.7236 aux.loss_ce: 0.0065 aux.acc_seg: 99.3533 +04/18 23:26:43 - mmengine - INFO - Iter(train) [136850/160000] lr: 1.8379e-03 eta: 3:33:25 time: 0.5526 data_time: 0.0071 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0068 decode.acc_seg: 99.7611 aux.loss_ce: 0.0072 aux.acc_seg: 99.3172 +04/18 23:27:11 - mmengine - INFO - Iter(train) [136900/160000] lr: 1.8345e-03 eta: 3:32:58 time: 0.5513 data_time: 0.0068 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0067 decode.acc_seg: 99.6988 aux.loss_ce: 0.0072 aux.acc_seg: 99.2083 +04/18 23:27:38 - mmengine - INFO - Iter(train) [136950/160000] lr: 1.8311e-03 eta: 3:32:30 time: 0.5529 data_time: 0.0068 memory: 7635 loss: 0.0140 decode.loss_ce: 0.0065 decode.acc_seg: 99.7701 aux.loss_ce: 0.0075 aux.acc_seg: 99.2968 +04/18 23:28:06 - mmengine - INFO - Exp name: upernet_r101_4xb4-160k_cag-512x512_20240418_022212 +04/18 23:28:06 - mmengine - INFO - Iter(train) [137000/160000] lr: 1.8278e-03 eta: 3:32:02 time: 0.5530 data_time: 0.0071 memory: 7635 loss: 0.0150 decode.loss_ce: 0.0069 decode.acc_seg: 99.7358 aux.loss_ce: 0.0080 aux.acc_seg: 99.1114 +04/18 23:28:34 - mmengine - INFO - Iter(train) [137050/160000] lr: 1.8244e-03 eta: 3:31:35 time: 0.5517 data_time: 0.0063 memory: 7635 loss: 0.0130 decode.loss_ce: 0.0062 decode.acc_seg: 99.7225 aux.loss_ce: 0.0069 aux.acc_seg: 99.3691 +04/18 23:29:01 - mmengine - INFO - Iter(train) [137100/160000] lr: 1.8210e-03 eta: 3:31:07 time: 0.5517 data_time: 0.0069 memory: 7635 loss: 0.0136 decode.loss_ce: 0.0064 decode.acc_seg: 99.7283 aux.loss_ce: 0.0072 aux.acc_seg: 99.1338 +04/18 23:29:29 - mmengine - INFO - Iter(train) [137150/160000] lr: 1.8176e-03 eta: 3:30:39 time: 0.5522 data_time: 0.0068 memory: 7635 loss: 0.0138 decode.loss_ce: 0.0065 decode.acc_seg: 99.6776 aux.loss_ce: 0.0072 aux.acc_seg: 99.2564 +04/18 23:29:56 - mmengine - INFO - Iter(train) [137200/160000] lr: 1.8142e-03 eta: 3:30:12 time: 0.5502 data_time: 0.0063 memory: 7635 loss: 0.0139 decode.loss_ce: 0.0065 decode.acc_seg: 99.7935 aux.loss_ce: 0.0074 aux.acc_seg: 99.3133 +04/18 23:30:24 - mmengine - INFO - Iter(train) [137250/160000] lr: 1.8109e-03 eta: 3:29:44 time: 0.5524 data_time: 0.0064 memory: 7635 loss: 0.0134 decode.loss_ce: 0.0063 decode.acc_seg: 99.7973 aux.loss_ce: 0.0071 aux.acc_seg: 99.4480 +04/18 23:30:52 - mmengine - INFO - Iter(train) [137300/160000] lr: 1.8075e-03 eta: 3:29:16 time: 0.5521 data_time: 0.0068 memory: 7635 loss: 0.0142 decode.loss_ce: 0.0069 decode.acc_seg: 99.7424 aux.loss_ce: 0.0073 aux.acc_seg: 99.1555 +04/18 23:31:19 - mmengine - INFO - Iter(train) [137350/160000] lr: 1.8041e-03 eta: 3:28:49 time: 0.5525 data_time: 0.0065 memory: 7635 loss: 0.0131 decode.loss_ce: 0.0059 decode.acc_seg: 99.7989 aux.loss_ce: 0.0072 aux.acc_seg: 99.2894 +WARNING:torch.distributed.elastic.agent.server.api:Received 2 death signal, shutting down workers +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 594961 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 594962 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 594963 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 594967 closing signal SIGINT +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + model = self.train_loop.run() # type: ignore + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + self.run_iter(data_batch) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + outputs = self.runner.model.train_step( + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 121, in train_step + losses = self._run_forward(data, mode='loss') + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 161, in _run_forward + results = self(**data, mode=mode) + File "/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl + return forward_call(*input, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/nn/parallel/distributed.py", line 963, in forward + output = self.module(*inputs[0], **kwargs[0]) + File "/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl + return forward_call(*input, **kwargs) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/base.py", line 94, in forward + return self.loss(inputs, data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/encoder_decoder.py", line 178, in loss + loss_decode = self._decode_head_forward_train(x, data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/encoder_decoder.py", line 139, in _decode_head_forward_train + loss_decode = self.decode_head.loss(inputs, data_samples, + File "/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py", line 262, in loss + losses = self.loss_by_feat(seg_logits, batch_data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py", line 336, in loss_by_feat + loss['acc_seg'] = accuracy( + File "/workspaces/mmsegmentation-1/mmseg/models/losses/accuracy.py", line 49, in accuracy + correct = correct[:, target != ignore_index] +KeyboardInterrupt +Traceback (most recent call last): +Traceback (most recent call last): +Traceback (most recent call last): + File "tools/train.py", line 104, in + File "tools/train.py", line 104, in + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + main() + main() + File "tools/train.py", line 100, in main + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + model = self.train_loop.run() # type: ignore + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + model = self.train_loop.run() # type: ignore + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + self.run_iter(data_batch)self.run_iter(data_batch) + + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + outputs = self.runner.model.train_step(outputs = self.runner.model.train_step( + + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 121, in train_step + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 121, in train_step + losses = self._run_forward(data, mode='loss')losses = self._run_forward(data, mode='loss') + + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 161, in _run_forward + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 161, in _run_forward + results = self(**data, mode=mode)results = self(**data, mode=mode) + + File "/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl + File "/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl + model = self.train_loop.run() # type: ignore + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + return forward_call(*input, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/nn/parallel/distributed.py", line 963, in forward + return forward_call(*input, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/nn/parallel/distributed.py", line 963, in forward + self.run_iter(data_batch) + output = self.module(*inputs[0], **kwargs[0]) + File "/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + output = self.module(*inputs[0], **kwargs[0]) + File "/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl + outputs = self.runner.model.train_step( return forward_call(*input, **kwargs) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/base.py", line 94, in forward + return forward_call(*input, **kwargs) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/base.py", line 94, in forward + return self.loss(inputs, data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/encoder_decoder.py", line 178, in loss + return self.loss(inputs, data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/encoder_decoder.py", line 178, in loss + +loss_decode = self._decode_head_forward_train(x, data_samples) File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 121, in train_step + + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/encoder_decoder.py", line 139, in _decode_head_forward_train + loss_decode = self._decode_head_forward_train(x, data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/encoder_decoder.py", line 139, in _decode_head_forward_train + loss_decode = self.decode_head.loss(inputs, data_samples, + File "/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py", line 262, in loss +losses = self._run_forward(data, mode='loss')loss_decode = self.decode_head.loss(inputs, data_samples, + + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 161, in _run_forward + File "/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py", line 262, in loss + losses = self.loss_by_feat(seg_logits, batch_data_samples)results = self(**data, mode=mode) + + File "/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py", line 336, in loss_by_feat + File "/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl +losses = self.loss_by_feat(seg_logits, batch_data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py", line 336, in loss_by_feat + loss['acc_seg'] = accuracy( + File "/workspaces/mmsegmentation-1/mmseg/models/losses/accuracy.py", line 49, in accuracy +loss['acc_seg'] = accuracy( + File "/workspaces/mmsegmentation-1/mmseg/models/losses/accuracy.py", line 49, in accuracy + correct = correct[:, target != ignore_index] + KeyboardInterruptcorrect = correct[:, target != ignore_index] + +KeyboardInterrupt + return forward_call(*input, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/nn/parallel/distributed.py", line 963, in forward + output = self.module(*inputs[0], **kwargs[0]) + File "/opt/conda/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl + return forward_call(*input, **kwargs) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/base.py", line 94, in forward + return self.loss(inputs, data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/encoder_decoder.py", line 178, in loss + loss_decode = self._decode_head_forward_train(x, data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/segmentors/encoder_decoder.py", line 139, in _decode_head_forward_train + loss_decode = self.decode_head.loss(inputs, data_samples, + File "/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py", line 262, in loss + losses = self.loss_by_feat(seg_logits, batch_data_samples) + File "/workspaces/mmsegmentation-1/mmseg/models/decode_heads/decode_head.py", line 336, in loss_by_feat + loss['acc_seg'] = accuracy( + File "/workspaces/mmsegmentation-1/mmseg/models/losses/accuracy.py", line 49, in accuracy + correct = correct[:, target != ignore_index] +KeyboardInterrupt +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 236, in launch_agent + result = agent.run() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/metrics/api.py", line 125, in wrapper + result = f(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 709, in run + result = self._invoke_run(role) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 850, in _invoke_run + time.sleep(monitor_interval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 594917 got signal: 2 +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/19 00:20:59 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 1523973508 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 1523973508 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/19 00:21:00 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +fp16 = dict(loss_scale='dynamic') +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = None +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=768, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + num_convs=1, + type='FCNHead'), + backbone=dict( + act_cfg=dict(type='GELU'), + attn_drop_rate=0.0, + drop_path_rate=0.1, + embed_dims=768, + img_size=( + 512, + 512, + ), + in_channels=3, + init_values=1.0, + mlp_ratio=4, + norm_cfg=dict(eps=1e-06, type='LN'), + norm_eval=False, + num_heads=12, + num_layers=12, + out_indices=[ + 3, + 5, + 7, + 11, + ], + patch_size=16, + type='MAE'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=768, + dropout_ratio=0.1, + in_channels=[ + 768, + 768, + 768, + 768, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + neck=dict( + embed_dim=768, rescales=[ + 4, + 2, + 1, + 0.5, + ], type='Feature2Pyramid'), + pretrained= + '/workspaces/mmsegmentation-1/work_dirs/R101_4000/iter_50000.pth', + test_cfg=dict(crop_size=( + 512, + 512, + ), mode='slide', stride=( + 341, + 341, + )), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + constructor='LayerDecayOptimizerConstructor', + optimizer=dict( + betas=( + 0.9, + 0.999, + ), lr=0.0001, type='AdamW', weight_decay=0.05), + paramwise_cfg=dict(layer_decay_rate=0.65, num_layers=12), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, by_epoch=False, end=1500, start_factor=1e-06, + type='LinearLR'), + dict( + begin=1500, + by_epoch=False, + end=160000, + eta_min=0.0, + power=1.0, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=2, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = '/workspaces/mmsegmentation-1/work_dirs/MAE4000_240419' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/19 00:21:02 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.cls_token as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1set param backbone.patch_embed.projection.weight as id 0 + +set param backbone.layers.0.gamma_2 as id 1set param backbone.patch_embed.projection.bias as id 0 + +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1set param backbone.layers.0.gamma_1 as id 1 + +set param backbone.layers.0.attn.relative_position_bias_table as id 1set param backbone.layers.0.gamma_2 as id 1 + +set param backbone.layers.0.attn.qkv.weight as id 1set param backbone.layers.0.ln1.weight as id 1 + +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1set param backbone.layers.0.attn.proj.weight as id 1 + +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1set param backbone.layers.0.ln2.weight as id 1 + +set param backbone.layers.0.attn.qkv.bias as id 1set param backbone.layers.0.ln2.bias as id 1 + +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1set param backbone.layers.0.ln2.bias as id 1 + +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2set param backbone.layers.1.gamma_1 as id 2 + +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2set param backbone.layers.2.gamma_1 as id 3 + +set param backbone.layers.1.ln2.bias as id 2set param backbone.layers.2.gamma_2 as id 3 + +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.1.ffn.layers.1.weight as id 2set param backbone.layers.2.attn.qkv.weight as id 3 + +set param backbone.layers.1.ffn.layers.1.bias as id 2set param backbone.layers.2.attn.qkv.bias as id 3 + +set param backbone.layers.2.attn.proj.weight as id 3set param backbone.layers.2.gamma_1 as id 3 + +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln2.weight as id 3set param backbone.layers.2.ln1.weight as id 3 + +set param backbone.layers.2.ln2.bias as id 3set param backbone.layers.2.ln1.bias as id 3 + +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3set param backbone.layers.2.ffn.layers.0.0.bias as id 3 + +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3set param backbone.layers.2.attn.proj.weight as id 3 + +set param backbone.layers.2.ffn.layers.1.bias as id 3set param backbone.layers.2.attn.proj.bias as id 3 + +set param backbone.layers.2.ln2.weight as id 3set param backbone.layers.3.gamma_1 as id 4 + +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.2.ffn.layers.1.weight as id 3set param backbone.layers.3.attn.qkv.weight as id 4 + +set param backbone.layers.2.ffn.layers.1.bias as id 3set param backbone.layers.3.attn.qkv.bias as id 4 + +set param backbone.layers.3.attn.proj.weight as id 4set param backbone.layers.3.gamma_1 as id 4 + +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln2.weight as id 4set param backbone.layers.3.ln1.weight as id 4 + +set param backbone.layers.3.ln2.bias as id 4set param backbone.layers.3.ln1.bias as id 4 + +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4set param backbone.layers.3.attn.qkv.weight as id 4 + +set param backbone.layers.3.attn.qkv.bias as id 4set param backbone.layers.3.ffn.layers.1.weight as id 4 + +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.3.ln2.weight as id 4set param backbone.layers.4.gamma_2 as id 5 + +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.attn.proj.weight as id 5set param backbone.layers.4.gamma_2 as id 5 + +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln1.weight as id 5set param backbone.layers.4.ln2.weight as id 5 + +set param backbone.layers.4.ln2.bias as id 5set param backbone.layers.4.ln1.bias as id 5 + +set param backbone.layers.4.attn.relative_position_bias_table as id 5set param backbone.layers.4.ffn.layers.0.0.weight as id 5 + +set param backbone.layers.4.attn.qkv.weight as id 5set param backbone.layers.4.ffn.layers.0.0.bias as id 5 + +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5set param backbone.layers.4.attn.proj.weight as id 5 + +set param backbone.layers.4.attn.proj.bias as id 5set param backbone.layers.5.gamma_1 as id 6 + +set param backbone.layers.5.gamma_2 as id 6set param backbone.layers.4.ln2.weight as id 5 + +set param backbone.layers.5.ln1.weight as id 6set param backbone.layers.4.ln2.bias as id 5 + +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6set param backbone.layers.4.ffn.layers.0.0.weight as id 5 + +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6set param backbone.layers.4.ffn.layers.1.weight as id 5 + +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6set param backbone.layers.5.gamma_2 as id 6 + +set param backbone.layers.5.ffn.layers.1.weight as id 6set param backbone.layers.5.ln1.weight as id 6 + +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7set param backbone.layers.5.attn.relative_position_bias_table as id 6 + +set param backbone.layers.6.gamma_2 as id 7set param backbone.layers.5.attn.qkv.weight as id 6 + +set param backbone.layers.6.ln1.weight as id 7set param backbone.layers.5.attn.qkv.bias as id 6 + +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.5.attn.proj.weight as id 6set param backbone.layers.6.attn.relative_position_bias_table as id 7 + +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.5.ln2.bias as id 6set param backbone.layers.6.attn.proj.bias as id 7 + +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7set param backbone.layers.6.attn.qkv.weight as id 7 + +set param backbone.layers.6.attn.qkv.bias as id 7set param backbone.layers.6.ffn.layers.1.weight as id 7 + +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7set param backbone.layers.7.gamma_1 as id 8 + +set param backbone.layers.7.gamma_2 as id 8set param backbone.layers.6.ln2.weight as id 7 + +set param backbone.layers.7.ln1.weight as id 8set param backbone.layers.6.ln2.bias as id 7 + +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7set param backbone.layers.8.gamma_1 as id 9 + +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.7.gamma_1 as id 8set param backbone.layers.8.attn.proj.bias as id 9 + +set param backbone.layers.7.gamma_2 as id 8set param backbone.layers.8.ln2.weight as id 9 + +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8set param backbone.layers.8.ffn.layers.0.0.weight as id 9 + +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.9.gamma_1 as id 10set param backbone.layers.7.attn.qkv.bias as id 8 + +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.7.attn.proj.weight as id 8set param backbone.layers.9.ln1.weight as id 10 + +set param backbone.layers.9.ln1.bias as id 10set param backbone.layers.7.attn.proj.bias as id 8 + +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8set param backbone.layers.9.attn.proj.bias as id 10 + +set param backbone.layers.7.ffn.layers.0.0.bias as id 8set param backbone.layers.9.ln2.weight as id 10 + +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10set param backbone.layers.7.ffn.layers.1.bias as id 8 + +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 + +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +Param groups = { + "layer_0_decay": { + "param_names": [ + "backbone.cls_token", + "backbone.pos_embed", + "backbone.patch_embed.projection.weight" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.05 + }, + "layer_0_no_decay": { + "param_names": [ + "backbone.patch_embed.projection.bias" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.0 + }, + "layer_1_no_decay": { + "param_names": [ + "backbone.layers.0.gamma_1", + "backbone.layers.0.gamma_2", + "backbone.layers.0.ln1.weight", + "backbone.layers.0.ln1.bias", + "backbone.layers.0.attn.qkv.bias", + "backbone.layers.0.attn.proj.bias", + "backbone.layers.0.ln2.weight", + "backbone.layers.0.ln2.bias", + "backbone.layers.0.ffn.layers.0.0.bias", + "backbone.layers.0.ffn.layers.1.bias" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.0 + }, + "layer_1_decay": { + "param_names": [ + "backbone.layers.0.attn.relative_position_bias_table", + "backbone.layers.0.attn.qkv.weight", + "backbone.layers.0.attn.proj.weight", + "backbone.layers.0.ffn.layers.0.0.weight", + "backbone.layers.0.ffn.layers.1.weight" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.05 + }, + "layer_2_no_decay": { + "param_names": [ + "backbone.layers.1.gamma_1", + "backbone.layers.1.gamma_2", + "backbone.layers.1.ln1.weight", + "backbone.layers.1.ln1.bias", + "backbone.layers.1.attn.qkv.bias", + "backbone.layers.1.attn.proj.bias", + "backbone.layers.1.ln2.weight", + "backbone.layers.1.ln2.bias", + "backbone.layers.1.ffn.layers.0.0.bias", + "backbone.layers.1.ffn.layers.1.bias" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.0 + }, + "layer_2_decay": { + "param_names": [ + "backbone.layers.1.attn.relative_position_bias_table", + "backbone.layers.1.attn.qkv.weight", + "backbone.layers.1.attn.proj.weight", + "backbone.layers.1.ffn.layers.0.0.weight", + "backbone.layers.1.ffn.layers.1.weight" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.05 + }, + "layer_3_no_decay": { + "param_names": [ + "backbone.layers.2.gamma_1", + "backbone.layers.2.gamma_2", + "backbone.layers.2.ln1.weight", + "backbone.layers.2.ln1.bias", + "backbone.layers.2.attn.qkv.bias", + "backbone.layers.2.attn.proj.bias", + "backbone.layers.2.ln2.weight", + "backbone.layers.2.ln2.bias", + "backbone.layers.2.ffn.layers.0.0.bias", + "backbone.layers.2.ffn.layers.1.bias" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.0 + }, + "layer_3_decay": { + "param_names": [ + "backbone.layers.2.attn.relative_position_bias_table", + "backbone.layers.2.attn.qkv.weight", + "backbone.layers.2.attn.proj.weight", + "backbone.layers.2.ffn.layers.0.0.weight", + "backbone.layers.2.ffn.layers.1.weight" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.05 + }, + "layer_4_no_decay": { + "param_names": [ + "backbone.layers.3.gamma_1", + "backbone.layers.3.gamma_2", + "backbone.layers.3.ln1.weight", + "backbone.layers.3.ln1.bias", + "backbone.layers.3.attn.qkv.bias", + "backbone.layers.3.attn.proj.bias", + "backbone.layers.3.ln2.weight", + "backbone.layers.3.ln2.bias", + "backbone.layers.3.ffn.layers.0.0.bias", + "backbone.layers.3.ffn.layers.1.bias" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.0 + }, + "layer_4_decay": { + "param_names": [ + "backbone.layers.3.attn.relative_position_bias_table", + "backbone.layers.3.attn.qkv.weight", + "backbone.layers.3.attn.proj.weight", + "backbone.layers.3.ffn.layers.0.0.weight", + "backbone.layers.3.ffn.layers.1.weight" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.05 + }, + "layer_5_no_decay": { + "param_names": [ + "backbone.layers.4.gamma_1", + "backbone.layers.4.gamma_2", + "backbone.layers.4.ln1.weight", + "backbone.layers.4.ln1.bias", + "backbone.layers.4.attn.qkv.bias", + "backbone.layers.4.attn.proj.bias", + "backbone.layers.4.ln2.weight", + "backbone.layers.4.ln2.bias", + "backbone.layers.4.ffn.layers.0.0.bias", + "backbone.layers.4.ffn.layers.1.bias" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.0 + }, + "layer_5_decay": { + "param_names": [ + "backbone.layers.4.attn.relative_position_bias_table", + "backbone.layers.4.attn.qkv.weight", + "backbone.layers.4.attn.proj.weight", + "backbone.layers.4.ffn.layers.0.0.weight", + "backbone.layers.4.ffn.layers.1.weight" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.05 + }, + "layer_6_no_decay": { + "param_names": [ + "backbone.layers.5.gamma_1", + "backbone.layers.5.gamma_2", + "backbone.layers.5.ln1.weight", + "backbone.layers.5.ln1.bias", + "backbone.layers.5.attn.qkv.bias", + "backbone.layers.5.attn.proj.bias", + "backbone.layers.5.ln2.weight", + "backbone.layers.5.ln2.bias", + "backbone.layers.5.ffn.layers.0.0.bias", + "backbone.layers.5.ffn.layers.1.bias" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.0 + }, + "layer_6_decay": { + "param_names": [ + "backbone.layers.5.attn.relative_position_bias_table", + "backbone.layers.5.attn.qkv.weight", + "backbone.layers.5.attn.proj.weight", + "backbone.layers.5.ffn.layers.0.0.weight", + "backbone.layers.5.ffn.layers.1.weight" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.05 + }, + "layer_7_no_decay": { + "param_names": [ + "backbone.layers.6.gamma_1", + "backbone.layers.6.gamma_2", + "backbone.layers.6.ln1.weight", + "backbone.layers.6.ln1.bias", + "backbone.layers.6.attn.qkv.bias", + "backbone.layers.6.attn.proj.bias", + "backbone.layers.6.ln2.weight", + "backbone.layers.6.ln2.bias", + "backbone.layers.6.ffn.layers.0.0.bias", + "backbone.layers.6.ffn.layers.1.bias" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.0 + }, + "layer_7_decay": { + "param_names": [ + "backbone.layers.6.attn.relative_position_bias_table", + "backbone.layers.6.attn.qkv.weight", + "backbone.layers.6.attn.proj.weight", + "backbone.layers.6.ffn.layers.0.0.weight", + "backbone.layers.6.ffn.layers.1.weight" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.05 + }, + "layer_8_no_decay": { + "param_names": [ + "backbone.layers.7.gamma_1", + "backbone.layers.7.gamma_2", + "backbone.layers.7.ln1.weight", + "backbone.layers.7.ln1.bias", + "backbone.layers.7.attn.qkv.bias", + "backbone.layers.7.attn.proj.bias", + "backbone.layers.7.ln2.weight", + "backbone.layers.7.ln2.bias", + "backbone.layers.7.ffn.layers.0.0.bias", + "backbone.layers.7.ffn.layers.1.bias" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.0 + }, + "layer_8_decay": { + "param_names": [ + "backbone.layers.7.attn.relative_position_bias_table", + "backbone.layers.7.attn.qkv.weight", + "backbone.layers.7.attn.proj.weight", + "backbone.layers.7.ffn.layers.0.0.weight", + "backbone.layers.7.ffn.layers.1.weight" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.05 + }, + "layer_9_no_decay": { + "param_names": [ + "backbone.layers.8.gamma_1", + "backbone.layers.8.gamma_2", + "backbone.layers.8.ln1.weight", + "backbone.layers.8.ln1.bias", + "backbone.layers.8.attn.qkv.bias", + "backbone.layers.8.attn.proj.bias", + "backbone.layers.8.ln2.weight", + "backbone.layers.8.ln2.bias", + "backbone.layers.8.ffn.layers.0.0.bias", + "backbone.layers.8.ffn.layers.1.bias" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.0 + }, + "layer_9_decay": { + "param_names": [ + "backbone.layers.8.attn.relative_position_bias_table", + "backbone.layers.8.attn.qkv.weight", + "backbone.layers.8.attn.proj.weight", + "backbone.layers.8.ffn.layers.0.0.weight", + "backbone.layers.8.ffn.layers.1.weight" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.05 + }, + "layer_10_no_decay": { + "param_names": [ + "backbone.layers.9.gamma_1", + "backbone.layers.9.gamma_2", + "backbone.layers.9.ln1.weight", + "backbone.layers.9.ln1.bias", + "backbone.layers.9.attn.qkv.bias", + "backbone.layers.9.attn.proj.bias", + "backbone.layers.9.ln2.weight", + "backbone.layers.9.ln2.bias", + "backbone.layers.9.ffn.layers.0.0.bias", + "backbone.layers.9.ffn.layers.1.bias" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.0 + }, + "layer_10_decay": { + "param_names": [ + "backbone.layers.9.attn.relative_position_bias_table", + "backbone.layers.9.attn.qkv.weight", + "backbone.layers.9.attn.proj.weight", + "backbone.layers.9.ffn.layers.0.0.weight", + "backbone.layers.9.ffn.layers.1.weight" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.05 + }, + "layer_11_no_decay": { + "param_names": [ + "backbone.layers.10.gamma_1", + "backbone.layers.10.gamma_2", + "backbone.layers.10.ln1.weight", + "backbone.layers.10.ln1.bias", + "backbone.layers.10.attn.qkv.bias", + "backbone.layers.10.attn.proj.bias", + "backbone.layers.10.ln2.weight", + "backbone.layers.10.ln2.bias", + "backbone.layers.10.ffn.layers.0.0.bias", + "backbone.layers.10.ffn.layers.1.bias" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.0 + }, + "layer_11_decay": { + "param_names": [ + "backbone.layers.10.attn.relative_position_bias_table", + "backbone.layers.10.attn.qkv.weight", + "backbone.layers.10.attn.proj.weight", + "backbone.layers.10.ffn.layers.0.0.weight", + "backbone.layers.10.ffn.layers.1.weight" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.05 + }, + "layer_12_no_decay": { + "param_names": [ + "backbone.layers.11.gamma_1", + "backbone.layers.11.gamma_2", + "backbone.layers.11.ln1.weight", + "backbone.layers.11.ln1.bias", + "backbone.layers.11.attn.qkv.bias", + "backbone.layers.11.attn.proj.bias", + "backbone.layers.11.ln2.weight", + "backbone.layers.11.ln2.bias", + "backbone.layers.11.ffn.layers.0.0.bias", + "backbone.layers.11.ffn.layers.1.bias" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.0 + }, + "layer_12_decay": { + "param_names": [ + "backbone.layers.11.attn.relative_position_bias_table", + "backbone.layers.11.attn.qkv.weight", + "backbone.layers.11.attn.proj.weight", + "backbone.layers.11.ffn.layers.0.0.weight", + "backbone.layers.11.ffn.layers.1.weight" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.05 + }, + "layer_13_decay": { + "param_names": [ + "neck.upsample_4x.0.weight", + "neck.upsample_4x.3.weight", + "neck.upsample_2x.0.weight", + "decode_head.conv_seg.weight", + "decode_head.psp_modules.0.1.conv.weight", + "decode_head.psp_modules.1.1.conv.weight", + "decode_head.psp_modules.2.1.conv.weight", + "decode_head.psp_modules.3.1.conv.weight", + "decode_head.bottleneck.conv.weight", + "decode_head.lateral_convs.0.conv.weight", + "decode_head.lateral_convs.1.conv.weight", + "decode_head.lateral_convs.2.conv.weight", + "decode_head.fpn_convs.0.conv.weight", + "decode_head.fpn_convs.1.conv.weight", + "decode_head.fpn_convs.2.conv.weight", + "decode_head.fpn_bottleneck.conv.weight", + "auxiliary_head.conv_seg.weight", + "auxiliary_head.convs.0.conv.weight" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.05 + }, + "layer_13_no_decay": { + "param_names": [ + "neck.upsample_4x.0.bias", + "neck.upsample_4x.1.weight", + "neck.upsample_4x.1.bias", + "neck.upsample_4x.3.bias", + "neck.upsample_2x.0.bias", + "decode_head.conv_seg.bias", + "decode_head.psp_modules.0.1.bn.weight", + "decode_head.psp_modules.0.1.bn.bias", + "decode_head.psp_modules.1.1.bn.weight", + "decode_head.psp_modules.1.1.bn.bias", + "decode_head.psp_modules.2.1.bn.weight", + "decode_head.psp_modules.2.1.bn.bias", + "decode_head.psp_modules.3.1.bn.weight", + "decode_head.psp_modules.3.1.bn.bias", + "decode_head.bottleneck.bn.weight", + "decode_head.bottleneck.bn.bias", + "decode_head.lateral_convs.0.bn.weight", + "decode_head.lateral_convs.0.bn.bias", + "decode_head.lateral_convs.1.bn.weight", + "decode_head.lateral_convs.1.bn.bias", + "decode_head.lateral_convs.2.bn.weight", + "decode_head.lateral_convs.2.bn.bias", + "decode_head.fpn_convs.0.bn.weight", + "decode_head.fpn_convs.0.bn.bias", + "decode_head.fpn_convs.1.bn.weight", + "decode_head.fpn_convs.1.bn.bias", + "decode_head.fpn_convs.2.bn.weight", + "decode_head.fpn_convs.2.bn.bias", + "decode_head.fpn_bottleneck.bn.weight", + "decode_head.fpn_bottleneck.bn.bias", + "auxiliary_head.conv_seg.bias", + "auxiliary_head.convs.0.bn.weight", + "auxiliary_head.convs.0.bn.bias" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.0 + } +} +04/19 00:21:04 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/work_dirs/R101_4000/iter_50000.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/work_dirs/R101_4000/iter_50000.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/work_dirs/R101_4000/iter_50000.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/work_dirs/R101_4000/iter_50000.pth +04/19 00:21:05 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +04/19 00:21:05 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +04/19 00:21:05 - mmengine - INFO - Checkpoints will be saved to /workspaces/mmsegmentation-1/work_dirs/MAE4000_240419. +04/19 00:22:01 - mmengine - INFO - Iter(train) [ 50/160000] base_lr: 3.2689e-06 lr: 1.2086e-08 eta: 2 days, 2:05:43 time: 1.0235 data_time: 0.0041 memory: 8935 loss: 7.1308 decode.loss_ce: 5.1131 decode.acc_seg: 5.7117 aux.loss_ce: 2.0177 aux.acc_seg: 0.0000 +WARNING:torch.distributed.elastic.agent.server.api:Received 2 death signal, shutting down workers +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 38950 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 38951 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 38952 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 38953 closing signal SIGINT +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + model = self.train_loop.run() # type: ignore + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + self.run_iter(data_batch) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + outputs = self.runner.model.train_step( + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + optim_wrapper.update_params(parsed_loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 196, in update_params + self.backward(loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 220, in backward + loss.backward(**kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/_tensor.py", line 363, in backward + torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py", line 173, in backward + Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass +KeyboardInterrupt +Traceback (most recent call last): + File "tools/train.py", line 104, in +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + model = self.train_loop.run() # type: ignore + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + self.run_iter(data_batch) +model = self.train_loop.run() # type: ignore File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + outputs = self.runner.model.train_step( + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + optim_wrapper.update_params(parsed_loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 196, in update_params + self.run_iter(data_batch)self.backward(loss) + + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 220, in backward + loss.backward(**kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/_tensor.py", line 363, in backward + torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py", line 173, in backward +outputs = self.runner.model.train_step( + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass +KeyboardInterrupt + optim_wrapper.update_params(parsed_loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 196, in update_params + model = self.train_loop.run() # type: ignore + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + self.backward(loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 220, in backward + self.run_iter(data_batch) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + loss.backward(**kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/_tensor.py", line 363, in backward + outputs = self.runner.model.train_step( + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + optim_wrapper.update_params(parsed_loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 196, in update_params +torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py", line 173, in backward + self.backward(loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 220, in backward + Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass +KeyboardInterrupt + loss.backward(**kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/_tensor.py", line 363, in backward + torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py", line 173, in backward + Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass +KeyboardInterrupt +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 38950 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 38951 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 38952 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 38953 closing signal SIGTERM +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 709, in run + result = self._invoke_run(role) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 850, in _invoke_run + time.sleep(monitor_interval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 38920 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 716, in run + self._shutdown(e.sigval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 38920 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 236, in launch_agent + result = agent.run() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/metrics/api.py", line 125, in wrapper + result = f(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 721, in run + self._shutdown() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 38920 got signal: 2 +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/19 00:22:46 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 703928584 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 703928584 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/19 00:22:46 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +fp16 = dict(loss_scale='dynamic') +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = None +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=768, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + num_convs=1, + type='FCNHead'), + backbone=dict( + act_cfg=dict(type='GELU'), + attn_drop_rate=0.0, + drop_path_rate=0.1, + embed_dims=768, + img_size=( + 512, + 512, + ), + in_channels=3, + init_values=1.0, + mlp_ratio=4, + norm_cfg=dict(eps=1e-06, type='LN'), + norm_eval=False, + num_heads=12, + num_layers=12, + out_indices=[ + 3, + 5, + 7, + 11, + ], + patch_size=16, + type='MAE'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=768, + dropout_ratio=0.1, + in_channels=[ + 768, + 768, + 768, + 768, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + neck=dict( + embed_dim=768, rescales=[ + 4, + 2, + 1, + 0.5, + ], type='Feature2Pyramid'), + pretrained='/workspaces/mmsegmentation-1/work_dirs/MAE4000/iter_60000.pth', + test_cfg=dict(crop_size=( + 512, + 512, + ), mode='slide', stride=( + 341, + 341, + )), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + constructor='LayerDecayOptimizerConstructor', + optimizer=dict( + betas=( + 0.9, + 0.999, + ), lr=0.0001, type='AdamW', weight_decay=0.05), + paramwise_cfg=dict(layer_decay_rate=0.65, num_layers=12), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, by_epoch=False, end=1500, start_factor=1e-06, + type='LinearLR'), + dict( + begin=1500, + by_epoch=False, + end=160000, + eta_min=0.0, + power=1.0, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=2, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = '/workspaces/mmsegmentation-1/work_dirs/MAE4000_240419' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/19 00:22:49 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} + +set param backbone.layers.0.ln2.bias as id 1 +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2set param backbone.cls_token as id 0 + +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.pos_embed as id 0 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.0.gamma_1 as id 1set param backbone.layers.1.ffn.layers.1.bias as id 2 + +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.0.ln1.weight as id 1set param backbone.layers.2.ln1.weight as id 3 + +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3set param backbone.layers.0.attn.relative_position_bias_table as id 1 + +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3set param backbone.layers.0.attn.qkv.weight as id 1 + +set param backbone.layers.2.ln2.weight as id 3set param backbone.layers.0.attn.qkv.bias as id 1 + +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1set param backbone.layers.3.gamma_1 as id 4 + +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.4.gamma_1 as id 5set param backbone.layers.1.ln1.bias as id 2 + +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.1.attn.relative_position_bias_table as id 2set param backbone.layers.4.ln1.weight as id 5 + +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5set param backbone.layers.1.attn.proj.weight as id 2 + +set param backbone.layers.1.attn.proj.bias as id 2set param backbone.layers.4.attn.proj.weight as id 5 + +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.1.ln2.weight as id 2set param backbone.layers.4.ln2.weight as id 5 + +set param backbone.layers.4.ln2.bias as id 5set param backbone.layers.1.ln2.bias as id 2 + +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5set param backbone.layers.1.ffn.layers.1.weight as id 2 + +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.2.gamma_2 as id 3set param backbone.layers.5.ln1.bias as id 6 + +set param backbone.layers.2.ln1.weight as id 3set param backbone.layers.5.attn.relative_position_bias_table as id 6 + +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6set param backbone.layers.2.attn.relative_position_bias_table as id 3 + +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6set param backbone.layers.2.attn.proj.bias as id 3 + +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.2.ln2.bias as id 3set param backbone.layers.5.ffn.layers.1.bias as id 6 + +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.3.gamma_1 as id 4set param backbone.layers.6.attn.proj.weight as id 7 + +set param backbone.layers.6.attn.proj.bias as id 7set param backbone.layers.3.gamma_2 as id 4 + +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4set param backbone.layers.6.ffn.layers.0.0.weight as id 7 + +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.4.ln1.bias as id 5set param backbone.layers.7.attn.qkv.weight as id 8 + +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5set param backbone.layers.7.ffn.layers.0.0.weight as id 8 + +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9set param backbone.layers.4.ffn.layers.0.0.weight as id 5 + +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.5.ln1.weight as id 6set param backbone.layers.8.ln2.bias as id 9 + +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.5.attn.qkv.weight as id 6set param backbone.layers.8.ffn.layers.1.weight as id 9 + +set param backbone.layers.8.ffn.layers.1.bias as id 9set param backbone.layers.5.attn.qkv.bias as id 6 + +set param backbone.layers.9.gamma_1 as id 10set param backbone.layers.5.attn.proj.weight as id 6 + +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.5.ln2.weight as id 6set param backbone.layers.9.ln1.bias as id 10 + +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6set param backbone.layers.9.attn.qkv.bias as id 10 + +set param backbone.layers.5.ffn.layers.0.0.bias as id 6set param backbone.layers.9.attn.proj.weight as id 10 + +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.5.ffn.layers.1.bias as id 6set param backbone.layers.9.ln2.bias as id 10 + +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.6.gamma_1 as id 7set param backbone.layers.9.ffn.layers.0.0.bias as id 10 + +set param backbone.layers.6.gamma_2 as id 7set param backbone.layers.9.ffn.layers.1.weight as id 10 + +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11set param backbone.layers.6.attn.relative_position_bias_table as id 7 + +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.6.attn.qkv.weight as id 7set param backbone.layers.10.ln1.bias as id 11 + +set param backbone.layers.6.attn.qkv.bias as id 7set param backbone.layers.10.attn.relative_position_bias_table as id 11 + +set param backbone.layers.10.attn.qkv.weight as id 11set param backbone.layers.6.attn.proj.weight as id 7 + +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.6.ln2.weight as id 7set param backbone.layers.10.attn.proj.bias as id 11 + +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7set param backbone.layers.11.ffn.layers.1.bias as id 12 + +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param neck.upsample_4x.0.weight as id 13set param backbone.layers.6.ffn.layers.1.weight as id 7 + +set param neck.upsample_4x.0.bias as id 13 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param backbone.layers.7.gamma_1 as id 8 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param backbone.layers.7.gamma_2 as id 8 +set param neck.upsample_2x.0.weight as id 13 +set param backbone.layers.7.ln1.weight as id 8 +set param neck.upsample_2x.0.bias as id 13 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param decode_head.conv_seg.weight as id 13 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param decode_head.conv_seg.bias as id 13 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param decode_head.psp_modules.0.1.conv.weight as id 13set param backbone.layers.7.ln2.weight as id 8 + +set param decode_head.psp_modules.0.1.bn.weight as id 13set param backbone.layers.7.ln2.bias as id 8 + +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param decode_head.psp_modules.2.1.conv.weight as id 13set param backbone.layers.7.ffn.layers.1.bias as id 8 + +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param backbone.layers.8.gamma_1 as id 9 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param backbone.layers.8.gamma_2 as id 9 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13set param backbone.layers.8.ln1.weight as id 9 + +set param backbone.layers.8.ln1.bias as id 9 +set param decode_head.bottleneck.conv.weight as id 13 +set param backbone.layers.8.attn.relative_position_bias_table as id 9set param decode_head.bottleneck.bn.weight as id 13 + +set param decode_head.bottleneck.bn.bias as id 13 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9set param decode_head.lateral_convs.1.conv.weight as id 13 + +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param backbone.layers.8.ln2.weight as id 9set param decode_head.lateral_convs.1.bn.bias as id 13 + +set param backbone.layers.8.ln2.bias as id 9 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param decode_head.fpn_convs.0.conv.weight as id 13set param backbone.layers.8.ffn.layers.0.0.bias as id 9 + +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param backbone.layers.9.gamma_1 as id 10 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13set param backbone.layers.9.gamma_2 as id 10 + +set param decode_head.fpn_bottleneck.conv.weight as id 13set param backbone.layers.9.ln1.weight as id 10 + +set param decode_head.fpn_bottleneck.bn.weight as id 13set param backbone.layers.9.ln1.bias as id 10 + +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10set param auxiliary_head.conv_seg.weight as id 13 + +set param auxiliary_head.conv_seg.bias as id 13 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param backbone.layers.9.ln2.weight as id 10 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param backbone.layers.9.ln2.bias as id 10 +set param auxiliary_head.convs.0.bn.bias as id 13 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +Param groups = { + "layer_0_decay": { + "param_names": [ + "backbone.cls_token", + "backbone.pos_embed", + "backbone.patch_embed.projection.weight" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.05 + }, + "layer_0_no_decay": { + "param_names": [ + "backbone.patch_embed.projection.bias" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.0 + }, + "layer_1_no_decay": { + "param_names": [ + "backbone.layers.0.gamma_1", + "backbone.layers.0.gamma_2", + "backbone.layers.0.ln1.weight", + "backbone.layers.0.ln1.bias", + "backbone.layers.0.attn.qkv.bias", + "backbone.layers.0.attn.proj.bias", + "backbone.layers.0.ln2.weight", + "backbone.layers.0.ln2.bias", + "backbone.layers.0.ffn.layers.0.0.bias", + "backbone.layers.0.ffn.layers.1.bias" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.0 + }, + "layer_1_decay": { + "param_names": [ + "backbone.layers.0.attn.relative_position_bias_table", + "backbone.layers.0.attn.qkv.weight", + "backbone.layers.0.attn.proj.weight", + "backbone.layers.0.ffn.layers.0.0.weight", + "backbone.layers.0.ffn.layers.1.weight" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.05 + }, + "layer_2_no_decay": { + "param_names": [ + "backbone.layers.1.gamma_1", + "backbone.layers.1.gamma_2", + "backbone.layers.1.ln1.weight", + "backbone.layers.1.ln1.bias", + "backbone.layers.1.attn.qkv.bias", + "backbone.layers.1.attn.proj.bias", + "backbone.layers.1.ln2.weight", + "backbone.layers.1.ln2.bias", + "backbone.layers.1.ffn.layers.0.0.bias", + "backbone.layers.1.ffn.layers.1.bias" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.0 + }, + "layer_2_decay": { + "param_names": [ + "backbone.layers.1.attn.relative_position_bias_table", + "backbone.layers.1.attn.qkv.weight", + "backbone.layers.1.attn.proj.weight", + "backbone.layers.1.ffn.layers.0.0.weight", + "backbone.layers.1.ffn.layers.1.weight" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.05 + }, + "layer_3_no_decay": { + "param_names": [ + "backbone.layers.2.gamma_1", + "backbone.layers.2.gamma_2", + "backbone.layers.2.ln1.weight", + "backbone.layers.2.ln1.bias", + "backbone.layers.2.attn.qkv.bias", + "backbone.layers.2.attn.proj.bias", + "backbone.layers.2.ln2.weight", + "backbone.layers.2.ln2.bias", + "backbone.layers.2.ffn.layers.0.0.bias", + "backbone.layers.2.ffn.layers.1.bias" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.0 + }, + "layer_3_decay": { + "param_names": [ + "backbone.layers.2.attn.relative_position_bias_table", + "backbone.layers.2.attn.qkv.weight", + "backbone.layers.2.attn.proj.weight", + "backbone.layers.2.ffn.layers.0.0.weight", + "backbone.layers.2.ffn.layers.1.weight" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.05 + }, + "layer_4_no_decay": { + "param_names": [ + "backbone.layers.3.gamma_1", + "backbone.layers.3.gamma_2", + "backbone.layers.3.ln1.weight", + "backbone.layers.3.ln1.bias", + "backbone.layers.3.attn.qkv.bias", + "backbone.layers.3.attn.proj.bias", + "backbone.layers.3.ln2.weight", + "backbone.layers.3.ln2.bias", + "backbone.layers.3.ffn.layers.0.0.bias", + "backbone.layers.3.ffn.layers.1.bias" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.0 + }, + "layer_4_decay": { + "param_names": [ + "backbone.layers.3.attn.relative_position_bias_table", + "backbone.layers.3.attn.qkv.weight", + "backbone.layers.3.attn.proj.weight", + "backbone.layers.3.ffn.layers.0.0.weight", + "backbone.layers.3.ffn.layers.1.weight" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.05 + }, + "layer_5_no_decay": { + "param_names": [ + "backbone.layers.4.gamma_1", + "backbone.layers.4.gamma_2", + "backbone.layers.4.ln1.weight", + "backbone.layers.4.ln1.bias", + "backbone.layers.4.attn.qkv.bias", + "backbone.layers.4.attn.proj.bias", + "backbone.layers.4.ln2.weight", + "backbone.layers.4.ln2.bias", + "backbone.layers.4.ffn.layers.0.0.bias", + "backbone.layers.4.ffn.layers.1.bias" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.0 + }, + "layer_5_decay": { + "param_names": [ + "backbone.layers.4.attn.relative_position_bias_table", + "backbone.layers.4.attn.qkv.weight", + "backbone.layers.4.attn.proj.weight", + "backbone.layers.4.ffn.layers.0.0.weight", + "backbone.layers.4.ffn.layers.1.weight" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.05 + }, + "layer_6_no_decay": { + "param_names": [ + "backbone.layers.5.gamma_1", + "backbone.layers.5.gamma_2", + "backbone.layers.5.ln1.weight", + "backbone.layers.5.ln1.bias", + "backbone.layers.5.attn.qkv.bias", + "backbone.layers.5.attn.proj.bias", + "backbone.layers.5.ln2.weight", + "backbone.layers.5.ln2.bias", + "backbone.layers.5.ffn.layers.0.0.bias", + "backbone.layers.5.ffn.layers.1.bias" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.0 + }, + "layer_6_decay": { + "param_names": [ + "backbone.layers.5.attn.relative_position_bias_table", + "backbone.layers.5.attn.qkv.weight", + "backbone.layers.5.attn.proj.weight", + "backbone.layers.5.ffn.layers.0.0.weight", + "backbone.layers.5.ffn.layers.1.weight" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.05 + }, + "layer_7_no_decay": { + "param_names": [ + "backbone.layers.6.gamma_1", + "backbone.layers.6.gamma_2", + "backbone.layers.6.ln1.weight", + "backbone.layers.6.ln1.bias", + "backbone.layers.6.attn.qkv.bias", + "backbone.layers.6.attn.proj.bias", + "backbone.layers.6.ln2.weight", + "backbone.layers.6.ln2.bias", + "backbone.layers.6.ffn.layers.0.0.bias", + "backbone.layers.6.ffn.layers.1.bias" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.0 + }, + "layer_7_decay": { + "param_names": [ + "backbone.layers.6.attn.relative_position_bias_table", + "backbone.layers.6.attn.qkv.weight", + "backbone.layers.6.attn.proj.weight", + "backbone.layers.6.ffn.layers.0.0.weight", + "backbone.layers.6.ffn.layers.1.weight" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.05 + }, + "layer_8_no_decay": { + "param_names": [ + "backbone.layers.7.gamma_1", + "backbone.layers.7.gamma_2", + "backbone.layers.7.ln1.weight", + "backbone.layers.7.ln1.bias", + "backbone.layers.7.attn.qkv.bias", + "backbone.layers.7.attn.proj.bias", + "backbone.layers.7.ln2.weight", + "backbone.layers.7.ln2.bias", + "backbone.layers.7.ffn.layers.0.0.bias", + "backbone.layers.7.ffn.layers.1.bias" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.0 + }, + "layer_8_decay": { + "param_names": [ + "backbone.layers.7.attn.relative_position_bias_table", + "backbone.layers.7.attn.qkv.weight", + "backbone.layers.7.attn.proj.weight", + "backbone.layers.7.ffn.layers.0.0.weight", + "backbone.layers.7.ffn.layers.1.weight" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.05 + }, + "layer_9_no_decay": { + "param_names": [ + "backbone.layers.8.gamma_1", + "backbone.layers.8.gamma_2", + "backbone.layers.8.ln1.weight", + "backbone.layers.8.ln1.bias", + "backbone.layers.8.attn.qkv.bias", + "backbone.layers.8.attn.proj.bias", + "backbone.layers.8.ln2.weight", + "backbone.layers.8.ln2.bias", + "backbone.layers.8.ffn.layers.0.0.bias", + "backbone.layers.8.ffn.layers.1.bias" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.0 + }, + "layer_9_decay": { + "param_names": [ + "backbone.layers.8.attn.relative_position_bias_table", + "backbone.layers.8.attn.qkv.weight", + "backbone.layers.8.attn.proj.weight", + "backbone.layers.8.ffn.layers.0.0.weight", + "backbone.layers.8.ffn.layers.1.weight" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.05 + }, + "layer_10_no_decay": { + "param_names": [ + "backbone.layers.9.gamma_1", + "backbone.layers.9.gamma_2", + "backbone.layers.9.ln1.weight", + "backbone.layers.9.ln1.bias", + "backbone.layers.9.attn.qkv.bias", + "backbone.layers.9.attn.proj.bias", + "backbone.layers.9.ln2.weight", + "backbone.layers.9.ln2.bias", + "backbone.layers.9.ffn.layers.0.0.bias", + "backbone.layers.9.ffn.layers.1.bias" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.0 + }, + "layer_10_decay": { + "param_names": [ + "backbone.layers.9.attn.relative_position_bias_table", + "backbone.layers.9.attn.qkv.weight", + "backbone.layers.9.attn.proj.weight", + "backbone.layers.9.ffn.layers.0.0.weight", + "backbone.layers.9.ffn.layers.1.weight" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.05 + }, + "layer_11_no_decay": { + "param_names": [ + "backbone.layers.10.gamma_1", + "backbone.layers.10.gamma_2", + "backbone.layers.10.ln1.weight", + "backbone.layers.10.ln1.bias", + "backbone.layers.10.attn.qkv.bias", + "backbone.layers.10.attn.proj.bias", + "backbone.layers.10.ln2.weight", + "backbone.layers.10.ln2.bias", + "backbone.layers.10.ffn.layers.0.0.bias", + "backbone.layers.10.ffn.layers.1.bias" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.0 + }, + "layer_11_decay": { + "param_names": [ + "backbone.layers.10.attn.relative_position_bias_table", + "backbone.layers.10.attn.qkv.weight", + "backbone.layers.10.attn.proj.weight", + "backbone.layers.10.ffn.layers.0.0.weight", + "backbone.layers.10.ffn.layers.1.weight" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.05 + }, + "layer_12_no_decay": { + "param_names": [ + "backbone.layers.11.gamma_1", + "backbone.layers.11.gamma_2", + "backbone.layers.11.ln1.weight", + "backbone.layers.11.ln1.bias", + "backbone.layers.11.attn.qkv.bias", + "backbone.layers.11.attn.proj.bias", + "backbone.layers.11.ln2.weight", + "backbone.layers.11.ln2.bias", + "backbone.layers.11.ffn.layers.0.0.bias", + "backbone.layers.11.ffn.layers.1.bias" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.0 + }, + "layer_12_decay": { + "param_names": [ + "backbone.layers.11.attn.relative_position_bias_table", + "backbone.layers.11.attn.qkv.weight", + "backbone.layers.11.attn.proj.weight", + "backbone.layers.11.ffn.layers.0.0.weight", + "backbone.layers.11.ffn.layers.1.weight" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.05 + }, + "layer_13_decay": { + "param_names": [ + "neck.upsample_4x.0.weight", + "neck.upsample_4x.3.weight", + "neck.upsample_2x.0.weight", + "decode_head.conv_seg.weight", + "decode_head.psp_modules.0.1.conv.weight", + "decode_head.psp_modules.1.1.conv.weight", + "decode_head.psp_modules.2.1.conv.weight", + "decode_head.psp_modules.3.1.conv.weight", + "decode_head.bottleneck.conv.weight", + "decode_head.lateral_convs.0.conv.weight", + "decode_head.lateral_convs.1.conv.weight", + "decode_head.lateral_convs.2.conv.weight", + "decode_head.fpn_convs.0.conv.weight", + "decode_head.fpn_convs.1.conv.weight", + "decode_head.fpn_convs.2.conv.weight", + "decode_head.fpn_bottleneck.conv.weight", + "auxiliary_head.conv_seg.weight", + "auxiliary_head.convs.0.conv.weight" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.05 + }, + "layer_13_no_decay": { + "param_names": [ + "neck.upsample_4x.0.bias", + "neck.upsample_4x.1.weight", + "neck.upsample_4x.1.bias", + "neck.upsample_4x.3.bias", + "neck.upsample_2x.0.bias", + "decode_head.conv_seg.bias", + "decode_head.psp_modules.0.1.bn.weight", + "decode_head.psp_modules.0.1.bn.bias", + "decode_head.psp_modules.1.1.bn.weight", + "decode_head.psp_modules.1.1.bn.bias", + "decode_head.psp_modules.2.1.bn.weight", + "decode_head.psp_modules.2.1.bn.bias", + "decode_head.psp_modules.3.1.bn.weight", + "decode_head.psp_modules.3.1.bn.bias", + "decode_head.bottleneck.bn.weight", + "decode_head.bottleneck.bn.bias", + "decode_head.lateral_convs.0.bn.weight", + "decode_head.lateral_convs.0.bn.bias", + "decode_head.lateral_convs.1.bn.weight", + "decode_head.lateral_convs.1.bn.bias", + "decode_head.lateral_convs.2.bn.weight", + "decode_head.lateral_convs.2.bn.bias", + "decode_head.fpn_convs.0.bn.weight", + "decode_head.fpn_convs.0.bn.bias", + "decode_head.fpn_convs.1.bn.weight", + "decode_head.fpn_convs.1.bn.bias", + "decode_head.fpn_convs.2.bn.weight", + "decode_head.fpn_convs.2.bn.bias", + "decode_head.fpn_bottleneck.bn.weight", + "decode_head.fpn_bottleneck.bn.bias", + "auxiliary_head.conv_seg.bias", + "auxiliary_head.convs.0.bn.weight", + "auxiliary_head.convs.0.bn.bias" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.0 + } +} +04/19 00:22:50 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/work_dirs/MAE4000/iter_60000.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/work_dirs/MAE4000/iter_60000.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/work_dirs/MAE4000/iter_60000.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/work_dirs/MAE4000/iter_60000.pth +Traceback (most recent call last): +Traceback (most recent call last): +Traceback (most recent call last): + File "tools/train.py", line 104, in +Traceback (most recent call last): + File "tools/train.py", line 104, in + File "tools/train.py", line 104, in + File "tools/train.py", line 104, in + main() +main()main()main() File "tools/train.py", line 100, in main + + + + File "tools/train.py", line 100, in main + File "tools/train.py", line 100, in main + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1748, in train + runner.train() runner.train() +runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1748, in train + + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1748, in train + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1748, in train + self._init_model_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 923, in _init_model_weights + self._init_model_weights() +self._init_model_weights()self._init_model_weights() File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 923, in _init_model_weights + + + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 923, in _init_model_weights + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 923, in _init_model_weights + model.init_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/base_module.py", line 136, in init_weights + model.init_weights()model.init_weights() + + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/base_module.py", line 136, in init_weights + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/base_module.py", line 136, in init_weights +model.init_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/base_module.py", line 136, in init_weights + m.init_weights()m.init_weights()m.init_weights()m.init_weights() + + + + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/mae.py", line 185, in init_weights + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/mae.py", line 185, in init_weights + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/mae.py", line 185, in init_weights + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/mae.py", line 185, in init_weights + state_dict = self.resize_rel_pos_embed(checkpoint)state_dict = self.resize_rel_pos_embed(checkpoint)state_dict = self.resize_rel_pos_embed(checkpoint)state_dict = self.resize_rel_pos_embed(checkpoint) + + + + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py", line 463, in resize_rel_pos_embed + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py", line 463, in resize_rel_pos_embed + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py", line 463, in resize_rel_pos_embed + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py", line 463, in resize_rel_pos_embed + dst_num_pos, _ = self.state_dict()[key].size() dst_num_pos, _ = self.state_dict()[key].size()dst_num_pos, _ = self.state_dict()[key].size() +dst_num_pos, _ = self.state_dict()[key].size() + + +KeyErrorKeyErrorKeyError: KeyError: : 'backbone.layers.0.attn.relative_position_bias_table': 'backbone.layers.0.attn.relative_position_bias_table''backbone.layers.0.attn.relative_position_bias_table' +'backbone.layers.0.attn.relative_position_bias_table' + + +ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 1) local_rank: 0 (pid: 40810) of binary: /opt/conda/bin/python +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 245, in launch_agent + raise ChildFailedError( +torch.distributed.elastic.multiprocessing.errors.ChildFailedError: +============================================================ +tools/train.py FAILED +------------------------------------------------------------ +Failures: +[1]: + time : 2024-04-19_00:22:57 + host : 06540a9609a4 + rank : 1 (local_rank: 1) + exitcode : 1 (pid: 40811) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[2]: + time : 2024-04-19_00:22:57 + host : 06540a9609a4 + rank : 2 (local_rank: 2) + exitcode : 1 (pid: 40812) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[3]: + time : 2024-04-19_00:22:57 + host : 06540a9609a4 + rank : 3 (local_rank: 3) + exitcode : 1 (pid: 40813) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +------------------------------------------------------------ +Root Cause (first observed failure): +[0]: + time : 2024-04-19_00:22:57 + host : 06540a9609a4 + rank : 0 (local_rank: 0) + exitcode : 1 (pid: 40810) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +============================================================ +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/19 00:33:14 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 1894895559 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 1894895559 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/19 00:33:14 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +fp16 = dict(loss_scale='dynamic') +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = '' +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=768, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + num_convs=1, + type='FCNHead'), + backbone=dict( + act_cfg=dict(type='GELU'), + attn_drop_rate=0.0, + drop_path_rate=0.1, + embed_dims=768, + img_size=( + 512, + 512, + ), + in_channels=3, + init_values=1.0, + mlp_ratio=4, + norm_cfg=dict(eps=1e-06, type='LN'), + norm_eval=False, + num_heads=12, + num_layers=12, + out_indices=[ + 3, + 5, + 7, + 11, + ], + patch_size=16, + type='MAE'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=768, + dropout_ratio=0.1, + in_channels=[ + 768, + 768, + 768, + 768, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + neck=dict( + embed_dim=768, rescales=[ + 4, + 2, + 1, + 0.5, + ], type='Feature2Pyramid'), + pretrained='/workspaces/mmsegmentation-1/work_dirs/MAE4000/iter_60000.pth', + test_cfg=dict(crop_size=( + 512, + 512, + ), mode='slide', stride=( + 341, + 341, + )), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + constructor='LayerDecayOptimizerConstructor', + optimizer=dict( + betas=( + 0.9, + 0.999, + ), lr=0.0001, type='AdamW', weight_decay=0.05), + paramwise_cfg=dict(layer_decay_rate=0.65, num_layers=12), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, by_epoch=False, end=1500, start_factor=1e-06, + type='LinearLR'), + dict( + begin=1500, + by_epoch=False, + end=160000, + eta_min=0.0, + power=1.0, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=2, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = '/workspaces/mmsegmentation-1/work_dirs/MAE4000_240418' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/19 00:33:17 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +set param backbone.layers.0.ln2.weight as id 1 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.1.attn.qkv.bias as id 2set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 + +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +set param backbone.layers.11.gamma_2 as id 12 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.11.ffn.layers.1.bias as id 12set param backbone.layers.6.gamma_2 as id 7 + +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param neck.upsample_4x.0.weight as id 13set param backbone.layers.6.attn.proj.weight as id 7 + +set param backbone.layers.6.attn.proj.bias as id 7 +set param neck.upsample_4x.0.bias as id 13 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param decode_head.conv_seg.bias as id 13 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param backbone.layers.7.gamma_1 as id 8 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param backbone.layers.7.gamma_2 as id 8 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8set param decode_head.psp_modules.1.1.conv.weight as id 13 + +set param decode_head.psp_modules.1.1.bn.weight as id 13set param backbone.layers.7.attn.qkv.weight as id 8 + +set param decode_head.psp_modules.1.1.bn.bias as id 13set param backbone.layers.7.attn.qkv.bias as id 8 + +set param backbone.layers.7.attn.proj.weight as id 8 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param backbone.layers.7.attn.proj.bias as id 8 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param backbone.layers.7.ln2.weight as id 8set param decode_head.psp_modules.2.1.bn.bias as id 13 + +set param backbone.layers.7.ln2.bias as id 8 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param decode_head.psp_modules.3.1.bn.weight as id 13set param backbone.layers.7.ffn.layers.0.0.bias as id 8 + +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param decode_head.bottleneck.conv.weight as id 13set param backbone.layers.7.ffn.layers.1.bias as id 8 + +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param decode_head.lateral_convs.0.conv.weight as id 13set param backbone.layers.8.ln1.weight as id 9 + +set param decode_head.lateral_convs.0.bn.weight as id 13set param backbone.layers.8.ln1.bias as id 9 + +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9set param decode_head.lateral_convs.1.conv.weight as id 13 + +set param backbone.layers.8.attn.qkv.bias as id 9 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param backbone.layers.8.attn.proj.weight as id 9set param decode_head.lateral_convs.1.bn.bias as id 13 + +set param backbone.layers.8.attn.proj.bias as id 9 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param backbone.layers.8.ln2.weight as id 9 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param backbone.layers.8.ln2.bias as id 9 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13set param backbone.layers.8.ffn.layers.0.0.weight as id 9 + +set param decode_head.fpn_convs.0.bn.weight as id 13set param backbone.layers.8.ffn.layers.0.0.bias as id 9 + +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param decode_head.fpn_convs.1.conv.weight as id 13set param backbone.layers.8.ffn.layers.1.bias as id 9 + +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13set param backbone.layers.9.gamma_1 as id 10 + +set param backbone.layers.9.gamma_2 as id 10 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param backbone.layers.9.ln1.weight as id 10 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param backbone.layers.9.ln1.bias as id 10 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param auxiliary_head.conv_seg.weight as id 13set param backbone.layers.9.ln2.bias as id 10 + +set param auxiliary_head.conv_seg.bias as id 13 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param auxiliary_head.convs.0.conv.weight as id 13set param backbone.layers.9.ffn.layers.0.0.bias as id 10 + +set param auxiliary_head.convs.0.bn.weight as id 13set param backbone.layers.9.ffn.layers.1.weight as id 10 + +set param auxiliary_head.convs.0.bn.bias as id 13set param backbone.layers.9.ffn.layers.1.bias as id 10 + +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +Param groups = { + "layer_0_decay": { + "param_names": [ + "backbone.cls_token", + "backbone.pos_embed", + "backbone.patch_embed.projection.weight" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.05 + }, + "layer_0_no_decay": { + "param_names": [ + "backbone.patch_embed.projection.bias" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.0 + }, + "layer_1_no_decay": { + "param_names": [ + "backbone.layers.0.gamma_1", + "backbone.layers.0.gamma_2", + "backbone.layers.0.ln1.weight", + "backbone.layers.0.ln1.bias", + "backbone.layers.0.attn.qkv.bias", + "backbone.layers.0.attn.proj.bias", + "backbone.layers.0.ln2.weight", + "backbone.layers.0.ln2.bias", + "backbone.layers.0.ffn.layers.0.0.bias", + "backbone.layers.0.ffn.layers.1.bias" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.0 + }, + "layer_1_decay": { + "param_names": [ + "backbone.layers.0.attn.relative_position_bias_table", + "backbone.layers.0.attn.qkv.weight", + "backbone.layers.0.attn.proj.weight", + "backbone.layers.0.ffn.layers.0.0.weight", + "backbone.layers.0.ffn.layers.1.weight" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.05 + }, + "layer_2_no_decay": { + "param_names": [ + "backbone.layers.1.gamma_1", + "backbone.layers.1.gamma_2", + "backbone.layers.1.ln1.weight", + "backbone.layers.1.ln1.bias", + "backbone.layers.1.attn.qkv.bias", + "backbone.layers.1.attn.proj.bias", + "backbone.layers.1.ln2.weight", + "backbone.layers.1.ln2.bias", + "backbone.layers.1.ffn.layers.0.0.bias", + "backbone.layers.1.ffn.layers.1.bias" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.0 + }, + "layer_2_decay": { + "param_names": [ + "backbone.layers.1.attn.relative_position_bias_table", + "backbone.layers.1.attn.qkv.weight", + "backbone.layers.1.attn.proj.weight", + "backbone.layers.1.ffn.layers.0.0.weight", + "backbone.layers.1.ffn.layers.1.weight" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.05 + }, + "layer_3_no_decay": { + "param_names": [ + "backbone.layers.2.gamma_1", + "backbone.layers.2.gamma_2", + "backbone.layers.2.ln1.weight", + "backbone.layers.2.ln1.bias", + "backbone.layers.2.attn.qkv.bias", + "backbone.layers.2.attn.proj.bias", + "backbone.layers.2.ln2.weight", + "backbone.layers.2.ln2.bias", + "backbone.layers.2.ffn.layers.0.0.bias", + "backbone.layers.2.ffn.layers.1.bias" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.0 + }, + "layer_3_decay": { + "param_names": [ + "backbone.layers.2.attn.relative_position_bias_table", + "backbone.layers.2.attn.qkv.weight", + "backbone.layers.2.attn.proj.weight", + "backbone.layers.2.ffn.layers.0.0.weight", + "backbone.layers.2.ffn.layers.1.weight" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.05 + }, + "layer_4_no_decay": { + "param_names": [ + "backbone.layers.3.gamma_1", + "backbone.layers.3.gamma_2", + "backbone.layers.3.ln1.weight", + "backbone.layers.3.ln1.bias", + "backbone.layers.3.attn.qkv.bias", + "backbone.layers.3.attn.proj.bias", + "backbone.layers.3.ln2.weight", + "backbone.layers.3.ln2.bias", + "backbone.layers.3.ffn.layers.0.0.bias", + "backbone.layers.3.ffn.layers.1.bias" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.0 + }, + "layer_4_decay": { + "param_names": [ + "backbone.layers.3.attn.relative_position_bias_table", + "backbone.layers.3.attn.qkv.weight", + "backbone.layers.3.attn.proj.weight", + "backbone.layers.3.ffn.layers.0.0.weight", + "backbone.layers.3.ffn.layers.1.weight" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.05 + }, + "layer_5_no_decay": { + "param_names": [ + "backbone.layers.4.gamma_1", + "backbone.layers.4.gamma_2", + "backbone.layers.4.ln1.weight", + "backbone.layers.4.ln1.bias", + "backbone.layers.4.attn.qkv.bias", + "backbone.layers.4.attn.proj.bias", + "backbone.layers.4.ln2.weight", + "backbone.layers.4.ln2.bias", + "backbone.layers.4.ffn.layers.0.0.bias", + "backbone.layers.4.ffn.layers.1.bias" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.0 + }, + "layer_5_decay": { + "param_names": [ + "backbone.layers.4.attn.relative_position_bias_table", + "backbone.layers.4.attn.qkv.weight", + "backbone.layers.4.attn.proj.weight", + "backbone.layers.4.ffn.layers.0.0.weight", + "backbone.layers.4.ffn.layers.1.weight" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.05 + }, + "layer_6_no_decay": { + "param_names": [ + "backbone.layers.5.gamma_1", + "backbone.layers.5.gamma_2", + "backbone.layers.5.ln1.weight", + "backbone.layers.5.ln1.bias", + "backbone.layers.5.attn.qkv.bias", + "backbone.layers.5.attn.proj.bias", + "backbone.layers.5.ln2.weight", + "backbone.layers.5.ln2.bias", + "backbone.layers.5.ffn.layers.0.0.bias", + "backbone.layers.5.ffn.layers.1.bias" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.0 + }, + "layer_6_decay": { + "param_names": [ + "backbone.layers.5.attn.relative_position_bias_table", + "backbone.layers.5.attn.qkv.weight", + "backbone.layers.5.attn.proj.weight", + "backbone.layers.5.ffn.layers.0.0.weight", + "backbone.layers.5.ffn.layers.1.weight" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.05 + }, + "layer_7_no_decay": { + "param_names": [ + "backbone.layers.6.gamma_1", + "backbone.layers.6.gamma_2", + "backbone.layers.6.ln1.weight", + "backbone.layers.6.ln1.bias", + "backbone.layers.6.attn.qkv.bias", + "backbone.layers.6.attn.proj.bias", + "backbone.layers.6.ln2.weight", + "backbone.layers.6.ln2.bias", + "backbone.layers.6.ffn.layers.0.0.bias", + "backbone.layers.6.ffn.layers.1.bias" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.0 + }, + "layer_7_decay": { + "param_names": [ + "backbone.layers.6.attn.relative_position_bias_table", + "backbone.layers.6.attn.qkv.weight", + "backbone.layers.6.attn.proj.weight", + "backbone.layers.6.ffn.layers.0.0.weight", + "backbone.layers.6.ffn.layers.1.weight" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.05 + }, + "layer_8_no_decay": { + "param_names": [ + "backbone.layers.7.gamma_1", + "backbone.layers.7.gamma_2", + "backbone.layers.7.ln1.weight", + "backbone.layers.7.ln1.bias", + "backbone.layers.7.attn.qkv.bias", + "backbone.layers.7.attn.proj.bias", + "backbone.layers.7.ln2.weight", + "backbone.layers.7.ln2.bias", + "backbone.layers.7.ffn.layers.0.0.bias", + "backbone.layers.7.ffn.layers.1.bias" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.0 + }, + "layer_8_decay": { + "param_names": [ + "backbone.layers.7.attn.relative_position_bias_table", + "backbone.layers.7.attn.qkv.weight", + "backbone.layers.7.attn.proj.weight", + "backbone.layers.7.ffn.layers.0.0.weight", + "backbone.layers.7.ffn.layers.1.weight" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.05 + }, + "layer_9_no_decay": { + "param_names": [ + "backbone.layers.8.gamma_1", + "backbone.layers.8.gamma_2", + "backbone.layers.8.ln1.weight", + "backbone.layers.8.ln1.bias", + "backbone.layers.8.attn.qkv.bias", + "backbone.layers.8.attn.proj.bias", + "backbone.layers.8.ln2.weight", + "backbone.layers.8.ln2.bias", + "backbone.layers.8.ffn.layers.0.0.bias", + "backbone.layers.8.ffn.layers.1.bias" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.0 + }, + "layer_9_decay": { + "param_names": [ + "backbone.layers.8.attn.relative_position_bias_table", + "backbone.layers.8.attn.qkv.weight", + "backbone.layers.8.attn.proj.weight", + "backbone.layers.8.ffn.layers.0.0.weight", + "backbone.layers.8.ffn.layers.1.weight" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.05 + }, + "layer_10_no_decay": { + "param_names": [ + "backbone.layers.9.gamma_1", + "backbone.layers.9.gamma_2", + "backbone.layers.9.ln1.weight", + "backbone.layers.9.ln1.bias", + "backbone.layers.9.attn.qkv.bias", + "backbone.layers.9.attn.proj.bias", + "backbone.layers.9.ln2.weight", + "backbone.layers.9.ln2.bias", + "backbone.layers.9.ffn.layers.0.0.bias", + "backbone.layers.9.ffn.layers.1.bias" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.0 + }, + "layer_10_decay": { + "param_names": [ + "backbone.layers.9.attn.relative_position_bias_table", + "backbone.layers.9.attn.qkv.weight", + "backbone.layers.9.attn.proj.weight", + "backbone.layers.9.ffn.layers.0.0.weight", + "backbone.layers.9.ffn.layers.1.weight" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.05 + }, + "layer_11_no_decay": { + "param_names": [ + "backbone.layers.10.gamma_1", + "backbone.layers.10.gamma_2", + "backbone.layers.10.ln1.weight", + "backbone.layers.10.ln1.bias", + "backbone.layers.10.attn.qkv.bias", + "backbone.layers.10.attn.proj.bias", + "backbone.layers.10.ln2.weight", + "backbone.layers.10.ln2.bias", + "backbone.layers.10.ffn.layers.0.0.bias", + "backbone.layers.10.ffn.layers.1.bias" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.0 + }, + "layer_11_decay": { + "param_names": [ + "backbone.layers.10.attn.relative_position_bias_table", + "backbone.layers.10.attn.qkv.weight", + "backbone.layers.10.attn.proj.weight", + "backbone.layers.10.ffn.layers.0.0.weight", + "backbone.layers.10.ffn.layers.1.weight" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.05 + }, + "layer_12_no_decay": { + "param_names": [ + "backbone.layers.11.gamma_1", + "backbone.layers.11.gamma_2", + "backbone.layers.11.ln1.weight", + "backbone.layers.11.ln1.bias", + "backbone.layers.11.attn.qkv.bias", + "backbone.layers.11.attn.proj.bias", + "backbone.layers.11.ln2.weight", + "backbone.layers.11.ln2.bias", + "backbone.layers.11.ffn.layers.0.0.bias", + "backbone.layers.11.ffn.layers.1.bias" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.0 + }, + "layer_12_decay": { + "param_names": [ + "backbone.layers.11.attn.relative_position_bias_table", + "backbone.layers.11.attn.qkv.weight", + "backbone.layers.11.attn.proj.weight", + "backbone.layers.11.ffn.layers.0.0.weight", + "backbone.layers.11.ffn.layers.1.weight" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.05 + }, + "layer_13_decay": { + "param_names": [ + "neck.upsample_4x.0.weight", + "neck.upsample_4x.3.weight", + "neck.upsample_2x.0.weight", + "decode_head.conv_seg.weight", + "decode_head.psp_modules.0.1.conv.weight", + "decode_head.psp_modules.1.1.conv.weight", + "decode_head.psp_modules.2.1.conv.weight", + "decode_head.psp_modules.3.1.conv.weight", + "decode_head.bottleneck.conv.weight", + "decode_head.lateral_convs.0.conv.weight", + "decode_head.lateral_convs.1.conv.weight", + "decode_head.lateral_convs.2.conv.weight", + "decode_head.fpn_convs.0.conv.weight", + "decode_head.fpn_convs.1.conv.weight", + "decode_head.fpn_convs.2.conv.weight", + "decode_head.fpn_bottleneck.conv.weight", + "auxiliary_head.conv_seg.weight", + "auxiliary_head.convs.0.conv.weight" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.05 + }, + "layer_13_no_decay": { + "param_names": [ + "neck.upsample_4x.0.bias", + "neck.upsample_4x.1.weight", + "neck.upsample_4x.1.bias", + "neck.upsample_4x.3.bias", + "neck.upsample_2x.0.bias", + "decode_head.conv_seg.bias", + "decode_head.psp_modules.0.1.bn.weight", + "decode_head.psp_modules.0.1.bn.bias", + "decode_head.psp_modules.1.1.bn.weight", + "decode_head.psp_modules.1.1.bn.bias", + "decode_head.psp_modules.2.1.bn.weight", + "decode_head.psp_modules.2.1.bn.bias", + "decode_head.psp_modules.3.1.bn.weight", + "decode_head.psp_modules.3.1.bn.bias", + "decode_head.bottleneck.bn.weight", + "decode_head.bottleneck.bn.bias", + "decode_head.lateral_convs.0.bn.weight", + "decode_head.lateral_convs.0.bn.bias", + "decode_head.lateral_convs.1.bn.weight", + "decode_head.lateral_convs.1.bn.bias", + "decode_head.lateral_convs.2.bn.weight", + "decode_head.lateral_convs.2.bn.bias", + "decode_head.fpn_convs.0.bn.weight", + "decode_head.fpn_convs.0.bn.bias", + "decode_head.fpn_convs.1.bn.weight", + "decode_head.fpn_convs.1.bn.bias", + "decode_head.fpn_convs.2.bn.weight", + "decode_head.fpn_convs.2.bn.bias", + "decode_head.fpn_bottleneck.bn.weight", + "decode_head.fpn_bottleneck.bn.bias", + "auxiliary_head.conv_seg.bias", + "auxiliary_head.convs.0.bn.weight", + "auxiliary_head.convs.0.bn.bias" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.0 + } +} +04/19 00:33:18 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/work_dirs/MAE4000/iter_60000.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/work_dirs/MAE4000/iter_60000.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/work_dirs/MAE4000/iter_60000.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/work_dirs/MAE4000/iter_60000.pth +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1748, in train + self._init_model_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 923, in _init_model_weights + model.init_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/base_module.py", line 136, in init_weights + m.init_weights() + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/mae.py", line 185, in init_weights + state_dict = self.resize_rel_pos_embed(checkpoint) + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py", line 463, in resize_rel_pos_embed + dst_num_pos, _ = self.state_dict()[key].size() +KeyError: 'backbone.layers.0.attn.relative_position_bias_table' +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1748, in train + self._init_model_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 923, in _init_model_weights + model.init_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/base_module.py", line 136, in init_weights + m.init_weights() + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/mae.py", line 185, in init_weights + state_dict = self.resize_rel_pos_embed(checkpoint) + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py", line 463, in resize_rel_pos_embed + dst_num_pos, _ = self.state_dict()[key].size() +KeyError: 'backbone.layers.0.attn.relative_position_bias_table' +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1748, in train + self._init_model_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 923, in _init_model_weights + model.init_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/base_module.py", line 136, in init_weights + m.init_weights() + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/mae.py", line 185, in init_weights + state_dict = self.resize_rel_pos_embed(checkpoint) + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py", line 463, in resize_rel_pos_embed + dst_num_pos, _ = self.state_dict()[key].size() +KeyError: 'backbone.layers.0.attn.relative_position_bias_table' +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1748, in train + self._init_model_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 923, in _init_model_weights + model.init_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/base_module.py", line 136, in init_weights + m.init_weights() + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/mae.py", line 185, in init_weights + state_dict = self.resize_rel_pos_embed(checkpoint) + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py", line 463, in resize_rel_pos_embed + dst_num_pos, _ = self.state_dict()[key].size() +KeyError: 'backbone.layers.0.attn.relative_position_bias_table' +ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 1) local_rank: 0 (pid: 49439) of binary: /opt/conda/bin/python +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 245, in launch_agent + raise ChildFailedError( +torch.distributed.elastic.multiprocessing.errors.ChildFailedError: +============================================================ +tools/train.py FAILED +------------------------------------------------------------ +Failures: +[1]: + time : 2024-04-19_00:33:20 + host : 06540a9609a4 + rank : 1 (local_rank: 1) + exitcode : 1 (pid: 49440) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[2]: + time : 2024-04-19_00:33:20 + host : 06540a9609a4 + rank : 2 (local_rank: 2) + exitcode : 1 (pid: 49441) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[3]: + time : 2024-04-19_00:33:20 + host : 06540a9609a4 + rank : 3 (local_rank: 3) + exitcode : 1 (pid: 49445) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +------------------------------------------------------------ +Root Cause (first observed failure): +[0]: + time : 2024-04-19_00:33:20 + host : 06540a9609a4 + rank : 0 (local_rank: 0) + exitcode : 1 (pid: 49439) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +============================================================ +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/19 00:36:34 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 135406166 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 135406166 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/19 00:36:34 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +fp16 = dict(loss_scale='dynamic') +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = '' +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=768, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + num_convs=1, + type='FCNHead'), + backbone=dict( + act_cfg=dict(type='GELU'), + attn_drop_rate=0.0, + drop_path_rate=0.1, + embed_dims=768, + img_size=( + 512, + 512, + ), + in_channels=3, + init_values=1.0, + mlp_ratio=4, + norm_cfg=dict(eps=1e-06, type='LN'), + norm_eval=False, + num_heads=12, + num_layers=12, + out_indices=[ + 3, + 5, + 7, + 11, + ], + patch_size=16, + type='MAE'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=768, + dropout_ratio=0.1, + in_channels=[ + 768, + 768, + 768, + 768, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + neck=dict( + embed_dim=768, rescales=[ + 4, + 2, + 1, + 0.5, + ], type='Feature2Pyramid'), + pretrained='/workspaces/mmsegmentation-1/converted_model.pth', + test_cfg=dict(crop_size=( + 512, + 512, + ), mode='slide', stride=( + 341, + 341, + )), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + constructor='LayerDecayOptimizerConstructor', + optimizer=dict( + betas=( + 0.9, + 0.999, + ), lr=0.0001, type='AdamW', weight_decay=0.05), + paramwise_cfg=dict(layer_decay_rate=0.65, num_layers=12), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, by_epoch=False, end=1500, start_factor=1e-06, + type='LinearLR'), + dict( + begin=1500, + by_epoch=False, + end=160000, + eta_min=0.0, + power=1.0, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=2, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = '/workspaces/mmsegmentation-1/work_dirs/MAE4000_240418' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/19 00:36:37 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +Param groups = { + "layer_0_decay": { + "param_names": [ + "backbone.cls_token", + "backbone.pos_embed", + "backbone.patch_embed.projection.weight" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.05 + }, + "layer_0_no_decay": { + "param_names": [ + "backbone.patch_embed.projection.bias" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.0 + }, + "layer_1_no_decay": { + "param_names": [ + "backbone.layers.0.gamma_1", + "backbone.layers.0.gamma_2", + "backbone.layers.0.ln1.weight", + "backbone.layers.0.ln1.bias", + "backbone.layers.0.attn.qkv.bias", + "backbone.layers.0.attn.proj.bias", + "backbone.layers.0.ln2.weight", + "backbone.layers.0.ln2.bias", + "backbone.layers.0.ffn.layers.0.0.bias", + "backbone.layers.0.ffn.layers.1.bias" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.0 + }, + "layer_1_decay": { + "param_names": [ + "backbone.layers.0.attn.relative_position_bias_table", + "backbone.layers.0.attn.qkv.weight", + "backbone.layers.0.attn.proj.weight", + "backbone.layers.0.ffn.layers.0.0.weight", + "backbone.layers.0.ffn.layers.1.weight" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.05 + }, + "layer_2_no_decay": { + "param_names": [ + "backbone.layers.1.gamma_1", + "backbone.layers.1.gamma_2", + "backbone.layers.1.ln1.weight", + "backbone.layers.1.ln1.bias", + "backbone.layers.1.attn.qkv.bias", + "backbone.layers.1.attn.proj.bias", + "backbone.layers.1.ln2.weight", + "backbone.layers.1.ln2.bias", + "backbone.layers.1.ffn.layers.0.0.bias", + "backbone.layers.1.ffn.layers.1.bias" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.0 + }, + "layer_2_decay": { + "param_names": [ + "backbone.layers.1.attn.relative_position_bias_table", + "backbone.layers.1.attn.qkv.weight", + "backbone.layers.1.attn.proj.weight", + "backbone.layers.1.ffn.layers.0.0.weight", + "backbone.layers.1.ffn.layers.1.weight" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.05 + }, + "layer_3_no_decay": { + "param_names": [ + "backbone.layers.2.gamma_1", + "backbone.layers.2.gamma_2", + "backbone.layers.2.ln1.weight", + "backbone.layers.2.ln1.bias", + "backbone.layers.2.attn.qkv.bias", + "backbone.layers.2.attn.proj.bias", + "backbone.layers.2.ln2.weight", + "backbone.layers.2.ln2.bias", + "backbone.layers.2.ffn.layers.0.0.bias", + "backbone.layers.2.ffn.layers.1.bias" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.0 + }, + "layer_3_decay": { + "param_names": [ + "backbone.layers.2.attn.relative_position_bias_table", + "backbone.layers.2.attn.qkv.weight", + "backbone.layers.2.attn.proj.weight", + "backbone.layers.2.ffn.layers.0.0.weight", + "backbone.layers.2.ffn.layers.1.weight" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.05 + }, + "layer_4_no_decay": { + "param_names": [ + "backbone.layers.3.gamma_1", + "backbone.layers.3.gamma_2", + "backbone.layers.3.ln1.weight", + "backbone.layers.3.ln1.bias", + "backbone.layers.3.attn.qkv.bias", + "backbone.layers.3.attn.proj.bias", + "backbone.layers.3.ln2.weight", + "backbone.layers.3.ln2.bias", + "backbone.layers.3.ffn.layers.0.0.bias", + "backbone.layers.3.ffn.layers.1.bias" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.0 + }, + "layer_4_decay": { + "param_names": [ + "backbone.layers.3.attn.relative_position_bias_table", + "backbone.layers.3.attn.qkv.weight", + "backbone.layers.3.attn.proj.weight", + "backbone.layers.3.ffn.layers.0.0.weight", + "backbone.layers.3.ffn.layers.1.weight" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.05 + }, + "layer_5_no_decay": { + "param_names": [ + "backbone.layers.4.gamma_1", + "backbone.layers.4.gamma_2", + "backbone.layers.4.ln1.weight", + "backbone.layers.4.ln1.bias", + "backbone.layers.4.attn.qkv.bias", + "backbone.layers.4.attn.proj.bias", + "backbone.layers.4.ln2.weight", + "backbone.layers.4.ln2.bias", + "backbone.layers.4.ffn.layers.0.0.bias", + "backbone.layers.4.ffn.layers.1.bias" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.0 + }, + "layer_5_decay": { + "param_names": [ + "backbone.layers.4.attn.relative_position_bias_table", + "backbone.layers.4.attn.qkv.weight", + "backbone.layers.4.attn.proj.weight", + "backbone.layers.4.ffn.layers.0.0.weight", + "backbone.layers.4.ffn.layers.1.weight" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.05 + }, + "layer_6_no_decay": { + "param_names": [ + "backbone.layers.5.gamma_1", + "backbone.layers.5.gamma_2", + "backbone.layers.5.ln1.weight", + "backbone.layers.5.ln1.bias", + "backbone.layers.5.attn.qkv.bias", + "backbone.layers.5.attn.proj.bias", + "backbone.layers.5.ln2.weight", + "backbone.layers.5.ln2.bias", + "backbone.layers.5.ffn.layers.0.0.bias", + "backbone.layers.5.ffn.layers.1.bias" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.0 + }, + "layer_6_decay": { + "param_names": [ + "backbone.layers.5.attn.relative_position_bias_table", + "backbone.layers.5.attn.qkv.weight", + "backbone.layers.5.attn.proj.weight", + "backbone.layers.5.ffn.layers.0.0.weight", + "backbone.layers.5.ffn.layers.1.weight" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.05 + }, + "layer_7_no_decay": { + "param_names": [ + "backbone.layers.6.gamma_1", + "backbone.layers.6.gamma_2", + "backbone.layers.6.ln1.weight", + "backbone.layers.6.ln1.bias", + "backbone.layers.6.attn.qkv.bias", + "backbone.layers.6.attn.proj.bias", + "backbone.layers.6.ln2.weight", + "backbone.layers.6.ln2.bias", + "backbone.layers.6.ffn.layers.0.0.bias", + "backbone.layers.6.ffn.layers.1.bias" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.0 + }, + "layer_7_decay": { + "param_names": [ + "backbone.layers.6.attn.relative_position_bias_table", + "backbone.layers.6.attn.qkv.weight", + "backbone.layers.6.attn.proj.weight", + "backbone.layers.6.ffn.layers.0.0.weight", + "backbone.layers.6.ffn.layers.1.weight" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.05 + }, + "layer_8_no_decay": { + "param_names": [ + "backbone.layers.7.gamma_1", + "backbone.layers.7.gamma_2", + "backbone.layers.7.ln1.weight", + "backbone.layers.7.ln1.bias", + "backbone.layers.7.attn.qkv.bias", + "backbone.layers.7.attn.proj.bias", + "backbone.layers.7.ln2.weight", + "backbone.layers.7.ln2.bias", + "backbone.layers.7.ffn.layers.0.0.bias", + "backbone.layers.7.ffn.layers.1.bias" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.0 + }, + "layer_8_decay": { + "param_names": [ + "backbone.layers.7.attn.relative_position_bias_table", + "backbone.layers.7.attn.qkv.weight", + "backbone.layers.7.attn.proj.weight", + "backbone.layers.7.ffn.layers.0.0.weight", + "backbone.layers.7.ffn.layers.1.weight" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.05 + }, + "layer_9_no_decay": { + "param_names": [ + "backbone.layers.8.gamma_1", + "backbone.layers.8.gamma_2", + "backbone.layers.8.ln1.weight", + "backbone.layers.8.ln1.bias", + "backbone.layers.8.attn.qkv.bias", + "backbone.layers.8.attn.proj.bias", + "backbone.layers.8.ln2.weight", + "backbone.layers.8.ln2.bias", + "backbone.layers.8.ffn.layers.0.0.bias", + "backbone.layers.8.ffn.layers.1.bias" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.0 + }, + "layer_9_decay": { + "param_names": [ + "backbone.layers.8.attn.relative_position_bias_table", + "backbone.layers.8.attn.qkv.weight", + "backbone.layers.8.attn.proj.weight", + "backbone.layers.8.ffn.layers.0.0.weight", + "backbone.layers.8.ffn.layers.1.weight" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.05 + }, + "layer_10_no_decay": { + "param_names": [ + "backbone.layers.9.gamma_1", + "backbone.layers.9.gamma_2", + "backbone.layers.9.ln1.weight", + "backbone.layers.9.ln1.bias", + "backbone.layers.9.attn.qkv.bias", + "backbone.layers.9.attn.proj.bias", + "backbone.layers.9.ln2.weight", + "backbone.layers.9.ln2.bias", + "backbone.layers.9.ffn.layers.0.0.bias", + "backbone.layers.9.ffn.layers.1.bias" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.0 + }, + "layer_10_decay": { + "param_names": [ + "backbone.layers.9.attn.relative_position_bias_table", + "backbone.layers.9.attn.qkv.weight", + "backbone.layers.9.attn.proj.weight", + "backbone.layers.9.ffn.layers.0.0.weight", + "backbone.layers.9.ffn.layers.1.weight" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.05 + }, + "layer_11_no_decay": { + "param_names": [ + "backbone.layers.10.gamma_1", + "backbone.layers.10.gamma_2", + "backbone.layers.10.ln1.weight", + "backbone.layers.10.ln1.bias", + "backbone.layers.10.attn.qkv.bias", + "backbone.layers.10.attn.proj.bias", + "backbone.layers.10.ln2.weight", + "backbone.layers.10.ln2.bias", + "backbone.layers.10.ffn.layers.0.0.bias", + "backbone.layers.10.ffn.layers.1.bias" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.0 + }, + "layer_11_decay": { + "param_names": [ + "backbone.layers.10.attn.relative_position_bias_table", + "backbone.layers.10.attn.qkv.weight", + "backbone.layers.10.attn.proj.weight", + "backbone.layers.10.ffn.layers.0.0.weight", + "backbone.layers.10.ffn.layers.1.weight" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.05 + }, + "layer_12_no_decay": { + "param_names": [ + "backbone.layers.11.gamma_1", + "backbone.layers.11.gamma_2", + "backbone.layers.11.ln1.weight", + "backbone.layers.11.ln1.bias", + "backbone.layers.11.attn.qkv.bias", + "backbone.layers.11.attn.proj.bias", + "backbone.layers.11.ln2.weight", + "backbone.layers.11.ln2.bias", + "backbone.layers.11.ffn.layers.0.0.bias", + "backbone.layers.11.ffn.layers.1.bias" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.0 + }, + "layer_12_decay": { + "param_names": [ + "backbone.layers.11.attn.relative_position_bias_table", + "backbone.layers.11.attn.qkv.weight", + "backbone.layers.11.attn.proj.weight", + "backbone.layers.11.ffn.layers.0.0.weight", + "backbone.layers.11.ffn.layers.1.weight" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.05 + }, + "layer_13_decay": { + "param_names": [ + "neck.upsample_4x.0.weight", + "neck.upsample_4x.3.weight", + "neck.upsample_2x.0.weight", + "decode_head.conv_seg.weight", + "decode_head.psp_modules.0.1.conv.weight", + "decode_head.psp_modules.1.1.conv.weight", + "decode_head.psp_modules.2.1.conv.weight", + "decode_head.psp_modules.3.1.conv.weight", + "decode_head.bottleneck.conv.weight", + "decode_head.lateral_convs.0.conv.weight", + "decode_head.lateral_convs.1.conv.weight", + "decode_head.lateral_convs.2.conv.weight", + "decode_head.fpn_convs.0.conv.weight", + "decode_head.fpn_convs.1.conv.weight", + "decode_head.fpn_convs.2.conv.weight", + "decode_head.fpn_bottleneck.conv.weight", + "auxiliary_head.conv_seg.weight", + "auxiliary_head.convs.0.conv.weight" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.05 + }, + "layer_13_no_decay": { + "param_names": [ + "neck.upsample_4x.0.bias", + "neck.upsample_4x.1.weight", + "neck.upsample_4x.1.bias", + "neck.upsample_4x.3.bias", + "neck.upsample_2x.0.bias", + "decode_head.conv_seg.bias", + "decode_head.psp_modules.0.1.bn.weight", + "decode_head.psp_modules.0.1.bn.bias", + "decode_head.psp_modules.1.1.bn.weight", + "decode_head.psp_modules.1.1.bn.bias", + "decode_head.psp_modules.2.1.bn.weight", + "decode_head.psp_modules.2.1.bn.bias", + "decode_head.psp_modules.3.1.bn.weight", + "decode_head.psp_modules.3.1.bn.bias", + "decode_head.bottleneck.bn.weight", + "decode_head.bottleneck.bn.bias", + "decode_head.lateral_convs.0.bn.weight", + "decode_head.lateral_convs.0.bn.bias", + "decode_head.lateral_convs.1.bn.weight", + "decode_head.lateral_convs.1.bn.bias", + "decode_head.lateral_convs.2.bn.weight", + "decode_head.lateral_convs.2.bn.bias", + "decode_head.fpn_convs.0.bn.weight", + "decode_head.fpn_convs.0.bn.bias", + "decode_head.fpn_convs.1.bn.weight", + "decode_head.fpn_convs.1.bn.bias", + "decode_head.fpn_convs.2.bn.weight", + "decode_head.fpn_convs.2.bn.bias", + "decode_head.fpn_bottleneck.bn.weight", + "decode_head.fpn_bottleneck.bn.bias", + "auxiliary_head.conv_seg.bias", + "auxiliary_head.convs.0.bn.weight", + "auxiliary_head.convs.0.bn.bias" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.0 + } +} +04/19 00:36:38 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1765, in train + self.load_or_resume() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1699, in load_or_resume + self.load_checkpoint(self._load_from) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 2127, in load_checkpoint + checkpoint = _load_checkpoint(filename, map_location=map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint +Loads checkpoint by local backend from path: +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1765, in train +Loads checkpoint by local backend from path: +Traceback (most recent call last): + File "tools/train.py", line 104, in + self.load_or_resume()main() + + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1699, in load_or_resume + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1765, in train + self.load_checkpoint(self._load_from) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 2127, in load_checkpoint + self.load_or_resume() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1699, in load_or_resume + checkpoint = _load_checkpoint(filename, map_location=map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + self.load_checkpoint(self._load_from) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 2127, in load_checkpoint + checkpoint = _load_checkpoint(filename, map_location=map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint +Loads checkpoint by local backend from path: +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1765, in train + self.load_or_resume() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1699, in load_or_resume + self.load_checkpoint(self._load_from) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 2127, in load_checkpoint + checkpoint = _load_checkpoint(filename, map_location=map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) return CheckpointLoader.load_checkpoint(filename, map_location, logger) +return CheckpointLoader.load_checkpoint(filename, map_location, logger) + + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError : return checkpoint_loader(filename, map_location) can not be found.return checkpoint_loader(filename, map_location)return checkpoint_loader(filename, map_location) + + + + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.')raise FileNotFoundError(f'{filename} can not be found.')raise FileNotFoundError(f'{filename} can not be found.') + + +FileNotFoundErrorFileNotFoundError: FileNotFoundError: can not be found.: can not be found. + can not be found. + +ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 1) local_rank: 0 (pid: 52295) of binary: /opt/conda/bin/python +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 245, in launch_agent + raise ChildFailedError( +torch.distributed.elastic.multiprocessing.errors.ChildFailedError: +============================================================ +tools/train.py FAILED +------------------------------------------------------------ +Failures: +[1]: + time : 2024-04-19_00:36:40 + host : 06540a9609a4 + rank : 1 (local_rank: 1) + exitcode : 1 (pid: 52296) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[2]: + time : 2024-04-19_00:36:40 + host : 06540a9609a4 + rank : 2 (local_rank: 2) + exitcode : 1 (pid: 52297) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[3]: + time : 2024-04-19_00:36:40 + host : 06540a9609a4 + rank : 3 (local_rank: 3) + exitcode : 1 (pid: 52298) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +------------------------------------------------------------ +Root Cause (first observed failure): +[0]: + time : 2024-04-19_00:36:40 + host : 06540a9609a4 + rank : 0 (local_rank: 0) + exitcode : 1 (pid: 52295) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +============================================================ +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +WARNING:torch.distributed.elastic.agent.server.api:Received 2 death signal, shutting down workers +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 55354 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 55355 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 55356 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 55357 closing signal SIGINT +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 93, in main +Traceback (most recent call last): + File "tools/train.py", line 104, in + runner = Runner.from_cfg(cfg) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 462, in from_cfg + runner = cls( + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 403, in __init__ + main()self._log_env(env_cfg) + + File "tools/train.py", line 93, in main + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 2368, in _log_env + runner = Runner.from_cfg(cfg) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 462, in from_cfg + runner = cls( + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 403, in __init__ + self._log_env(env_cfg) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 2368, in _log_env + env = collect_env() + File "/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/collect_env.py", line 125, in collect_env + env = collect_env() + File "/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/collect_env.py", line 125, in collect_env +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 93, in main + runner = Runner.from_cfg(cfg) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 462, in from_cfg + runner = cls( + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 403, in __init__ + self._log_env(env_cfg) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 2368, in _log_env + env = collect_env() + File "/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/collect_env.py", line 125, in collect_env + cc_info = subprocess.check_output(f'{cc} --version', shell=True)cc_info = subprocess.check_output(f'{cc} --version', shell=True) + + File "/opt/conda/lib/python3.8/subprocess.py", line 415, in check_output + File "/opt/conda/lib/python3.8/subprocess.py", line 415, in check_output +cc_info = subprocess.check_output(f'{cc} --version', shell=True) + File "/opt/conda/lib/python3.8/subprocess.py", line 415, in check_output + return run(*popenargs, stdout=PIPE, timeout=timeout, check=True, + File "/opt/conda/lib/python3.8/subprocess.py", line 493, in run + return run(*popenargs, stdout=PIPE, timeout=timeout, check=True,return run(*popenargs, stdout=PIPE, timeout=timeout, check=True, + + File "/opt/conda/lib/python3.8/subprocess.py", line 493, in run + File "/opt/conda/lib/python3.8/subprocess.py", line 493, in run + with Popen(*popenargs, **kwargs) as process: + File "/opt/conda/lib/python3.8/subprocess.py", line 858, in __init__ + with Popen(*popenargs, **kwargs) as process: + File "/opt/conda/lib/python3.8/subprocess.py", line 858, in __init__ + with Popen(*popenargs, **kwargs) as process: + File "/opt/conda/lib/python3.8/subprocess.py", line 858, in __init__ + self._execute_child(args, executable, preexec_fn, close_fds, + File "/opt/conda/lib/python3.8/subprocess.py", line 1639, in _execute_child + self._execute_child(args, executable, preexec_fn, close_fds, + File "/opt/conda/lib/python3.8/subprocess.py", line 1639, in _execute_child + self._execute_child(args, executable, preexec_fn, close_fds, + File "/opt/conda/lib/python3.8/subprocess.py", line 1639, in _execute_child + self.pid = _posixsubprocess.fork_exec(self.pid = _posixsubprocess.fork_exec( + +self.pid = _posixsubprocess.fork_exec( +KeyboardInterruptKeyboardInterrupt + +KeyboardInterrupt +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 93, in main + runner = Runner.from_cfg(cfg) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 462, in from_cfg + runner = cls( + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 403, in __init__ + self._log_env(env_cfg) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 2368, in _log_env + env = collect_env() + File "/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/collect_env.py", line 125, in collect_env + cc_info = subprocess.check_output(f'{cc} --version', shell=True) + File "/opt/conda/lib/python3.8/subprocess.py", line 415, in check_output + return run(*popenargs, stdout=PIPE, timeout=timeout, check=True, + File "/opt/conda/lib/python3.8/subprocess.py", line 493, in run + with Popen(*popenargs, **kwargs) as process: + File "/opt/conda/lib/python3.8/subprocess.py", line 858, in __init__ + self._execute_child(args, executable, preexec_fn, close_fds, + File "/opt/conda/lib/python3.8/subprocess.py", line 1639, in _execute_child + self.pid = _posixsubprocess.fork_exec( +KeyboardInterrupt +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 55354 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 55355 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 55356 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 55357 closing signal SIGTERM +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 709, in run + result = self._invoke_run(role) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 850, in _invoke_run + time.sleep(monitor_interval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 55324 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 716, in run + self._shutdown(e.sigval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 55324 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 236, in launch_agent + result = agent.run() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/metrics/api.py", line 125, in wrapper + result = f(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 721, in run + self._shutdown() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 55324 got signal: 2 +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/19 00:38:45 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 1757742453 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 1757742453 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/19 00:38:45 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +fp16 = dict(loss_scale='dynamic') +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = '' +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=768, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + num_convs=1, + type='FCNHead'), + backbone=dict( + act_cfg=dict(type='GELU'), + attn_drop_rate=0.0, + drop_path_rate=0.1, + embed_dims=768, + img_size=( + 512, + 512, + ), + in_channels=3, + init_values=1.0, + mlp_ratio=4, + norm_cfg=dict(eps=1e-06, type='LN'), + norm_eval=False, + num_heads=12, + num_layers=12, + out_indices=[ + 3, + 5, + 7, + 11, + ], + patch_size=16, + type='MAE'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=768, + dropout_ratio=0.1, + in_channels=[ + 768, + 768, + 768, + 768, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + neck=dict( + embed_dim=768, rescales=[ + 4, + 2, + 1, + 0.5, + ], type='Feature2Pyramid'), + pretrained='/workspaces/mmsegmentation-1/converted_model.pth', + test_cfg=dict(crop_size=( + 512, + 512, + ), mode='slide', stride=( + 341, + 341, + )), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + constructor='LayerDecayOptimizerConstructor', + optimizer=dict( + betas=( + 0.9, + 0.999, + ), lr=0.0001, type='AdamW', weight_decay=0.05), + paramwise_cfg=dict(layer_decay_rate=0.65, num_layers=12), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, by_epoch=False, end=1500, start_factor=1e-06, + type='LinearLR'), + dict( + begin=1500, + by_epoch=False, + end=160000, + eta_min=0.0, + power=1.0, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=2, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = '/workspaces/mmsegmentation-1/work_dirs/MAE4000' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/19 00:38:48 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14set param backbone.patch_embed.projection.weight as id 0 + +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.cls_token as id 0 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.pos_embed as id 0 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.layers.0.attn.proj.weight as id 1set param backbone.patch_embed.projection.bias as id 0 + +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.1.gamma_2 as id 2set param backbone.layers.0.attn.proj.bias as id 1 + +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.1.attn.proj.weight as id 2set param backbone.layers.0.ffn.layers.0.0.bias as id 1 + +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2set param backbone.layers.0.ffn.layers.1.weight as id 1 + +set param backbone.layers.1.ln2.bias as id 2set param backbone.layers.0.ffn.layers.1.bias as id 1 + +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2set param backbone.layers.1.gamma_2 as id 2 + +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.2.ln1.weight as id 3set param backbone.layers.1.attn.qkv.bias as id 2 + +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2set param backbone.layers.2.attn.proj.bias as id 3 + +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.1.ffn.layers.1.weight as id 2set param backbone.layers.2.ln2.bias as id 3 + +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.gamma_1 as id 3set param backbone.layers.2.ffn.layers.0.0.bias as id 3 + +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ln1.weight as id 3set param backbone.layers.2.ffn.layers.1.bias as id 3 + +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3set param backbone.layers.3.gamma_1 as id 4 + +set param backbone.layers.2.attn.qkv.weight as id 3set param backbone.layers.3.gamma_2 as id 4 + +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4set param backbone.layers.2.attn.proj.weight as id 3 + +set param backbone.layers.2.attn.proj.bias as id 3set param backbone.layers.3.attn.relative_position_bias_table as id 4 + +set param backbone.layers.2.ln2.weight as id 3set param backbone.layers.3.attn.qkv.weight as id 4 + +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4set param backbone.layers.2.ffn.layers.0.0.weight as id 3 + +set param backbone.layers.3.attn.proj.bias as id 4set param backbone.layers.2.ffn.layers.0.0.bias as id 3 + +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4set param backbone.layers.3.ln1.weight as id 4 + +set param backbone.layers.3.ffn.layers.1.bias as id 4set param backbone.layers.3.ln1.bias as id 4 + +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.4.attn.qkv.weight as id 5set param backbone.layers.3.ln2.bias as id 4 + +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4set param backbone.layers.4.ln2.weight as id 5 + +set param backbone.layers.3.ffn.layers.0.0.bias as id 4set param backbone.layers.4.ln2.bias as id 5 + +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.gamma_1 as id 5set param backbone.layers.4.ffn.layers.1.weight as id 5 + +set param backbone.layers.4.gamma_2 as id 5set param backbone.layers.4.ffn.layers.1.bias as id 5 + +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5set param backbone.layers.5.gamma_1 as id 6 + +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.5.ln1.weight as id 6set param backbone.layers.4.attn.qkv.bias as id 5 + +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5set param backbone.layers.5.attn.proj.bias as id 6 + +set param backbone.layers.4.ffn.layers.0.0.bias as id 5set param backbone.layers.5.ln2.weight as id 6 + +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6set param backbone.layers.5.gamma_2 as id 6 + +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7set param backbone.layers.5.ffn.layers.0.0.weight as id 6 + +set param backbone.layers.5.ffn.layers.0.0.bias as id 6set param backbone.layers.6.ln2.weight as id 7 + +set param backbone.layers.6.ln2.bias as id 7set param backbone.layers.5.ffn.layers.1.weight as id 6 + +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7set param backbone.layers.6.ln2.weight as id 7 + +set param backbone.layers.6.ln2.bias as id 7set param backbone.layers.6.ffn.layers.0.0.bias as id 7 + +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7set param backbone.layers.7.ln2.bias as id 8 + +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.8.gamma_2 as id 9set param backbone.layers.7.ln1.bias as id 8 + +set param backbone.layers.8.ln1.weight as id 9set param backbone.layers.7.attn.relative_position_bias_table as id 8 + +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.8.attn.relative_position_bias_table as id 9set param backbone.layers.7.attn.qkv.bias as id 8 + +set param backbone.layers.8.attn.qkv.weight as id 9set param backbone.layers.7.attn.proj.weight as id 8 + +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8set param backbone.layers.8.ln2.bias as id 9 + +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9set param backbone.layers.9.gamma_1 as id 10 + +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.8.attn.relative_position_bias_table as id 9set param backbone.layers.9.ln1.weight as id 10 + +set param backbone.layers.9.ln1.bias as id 10set param backbone.layers.8.attn.qkv.weight as id 9 + +set param backbone.layers.8.attn.qkv.bias as id 9set param backbone.layers.9.attn.relative_position_bias_table as id 10 + +set param backbone.layers.8.attn.proj.weight as id 9set param backbone.layers.9.attn.qkv.weight as id 10 + +set param backbone.layers.8.attn.proj.bias as id 9set param backbone.layers.9.attn.qkv.bias as id 10 + +set param backbone.layers.8.ln2.weight as id 9set param backbone.layers.9.attn.proj.weight as id 10 + +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10set param backbone.layers.8.ffn.layers.0.0.weight as id 9 + +set param backbone.layers.9.ln2.bias as id 10set param backbone.layers.8.ffn.layers.0.0.bias as id 9 + +set param backbone.layers.8.ffn.layers.1.weight as id 9set param backbone.layers.9.ffn.layers.0.0.weight as id 10 + +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10set param backbone.layers.9.gamma_1 as id 10 + +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.9.attn.qkv.weight as id 10set param backbone.layers.10.ln1.bias as id 11 + +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11set param backbone.layers.9.ffn.layers.0.0.weight as id 10 + +set param backbone.layers.10.ln2.bias as id 11set param backbone.layers.9.ffn.layers.0.0.bias as id 10 + +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11set param backbone.layers.10.gamma_1 as id 11 + +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12set param backbone.layers.10.attn.qkv.weight as id 11 + +set param backbone.layers.10.attn.qkv.bias as id 11set param backbone.layers.11.attn.relative_position_bias_table as id 12 + +set param backbone.layers.10.attn.proj.weight as id 11set param backbone.layers.11.attn.qkv.weight as id 12 + +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.10.ln2.weight as id 11set param backbone.layers.11.attn.proj.weight as id 12 + +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12set param backbone.layers.10.ffn.layers.1.weight as id 11 + +set param backbone.layers.11.ffn.layers.0.0.bias as id 12set param backbone.layers.10.ffn.layers.1.bias as id 11 + +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param neck.upsample_4x.0.bias as id 13 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param neck.upsample_4x.1.weight as id 13set param backbone.layers.11.attn.proj.weight as id 12 + +set param neck.upsample_4x.1.bias as id 13set param backbone.layers.11.attn.proj.bias as id 12 + +set param neck.upsample_4x.3.weight as id 13 +set param backbone.layers.11.ln2.weight as id 12 +set param neck.upsample_4x.3.bias as id 13 +set param backbone.layers.11.ln2.bias as id 12 +set param neck.upsample_2x.0.weight as id 13 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12set param neck.upsample_2x.0.bias as id 13 + +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13set param neck.upsample_4x.1.bias as id 13 + +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13set param neck.upsample_2x.0.weight as id 13 + +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13set param decode_head.psp_modules.3.1.conv.weight as id 13 + +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13set param decode_head.psp_modules.0.1.conv.weight as id 13 + +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13set param decode_head.lateral_convs.1.conv.weight as id 13 + +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13set param decode_head.bottleneck.conv.weight as id 13 + +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13set param decode_head.fpn_convs.1.conv.weight as id 13 + +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13set param decode_head.fpn_convs.1.bn.bias as id 13 + +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13set param decode_head.lateral_convs.1.conv.weight as id 13 + +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13set param decode_head.lateral_convs.2.conv.weight as id 13 + +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13set param decode_head.fpn_convs.0.conv.weight as id 13 + +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13set param auxiliary_head.convs.0.conv.weight as id 13 + +set param auxiliary_head.convs.0.bn.weight as id 13set param decode_head.fpn_convs.1.conv.weight as id 13 + +set param auxiliary_head.convs.0.bn.bias as id 13set param decode_head.fpn_convs.1.bn.weight as id 13 + +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +Param groups = { + "layer_0_decay": { + "param_names": [ + "backbone.cls_token", + "backbone.pos_embed", + "backbone.patch_embed.projection.weight" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.05 + }, + "layer_0_no_decay": { + "param_names": [ + "backbone.patch_embed.projection.bias" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.0 + }, + "layer_1_no_decay": { + "param_names": [ + "backbone.layers.0.gamma_1", + "backbone.layers.0.gamma_2", + "backbone.layers.0.ln1.weight", + "backbone.layers.0.ln1.bias", + "backbone.layers.0.attn.qkv.bias", + "backbone.layers.0.attn.proj.bias", + "backbone.layers.0.ln2.weight", + "backbone.layers.0.ln2.bias", + "backbone.layers.0.ffn.layers.0.0.bias", + "backbone.layers.0.ffn.layers.1.bias" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.0 + }, + "layer_1_decay": { + "param_names": [ + "backbone.layers.0.attn.relative_position_bias_table", + "backbone.layers.0.attn.qkv.weight", + "backbone.layers.0.attn.proj.weight", + "backbone.layers.0.ffn.layers.0.0.weight", + "backbone.layers.0.ffn.layers.1.weight" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.05 + }, + "layer_2_no_decay": { + "param_names": [ + "backbone.layers.1.gamma_1", + "backbone.layers.1.gamma_2", + "backbone.layers.1.ln1.weight", + "backbone.layers.1.ln1.bias", + "backbone.layers.1.attn.qkv.bias", + "backbone.layers.1.attn.proj.bias", + "backbone.layers.1.ln2.weight", + "backbone.layers.1.ln2.bias", + "backbone.layers.1.ffn.layers.0.0.bias", + "backbone.layers.1.ffn.layers.1.bias" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.0 + }, + "layer_2_decay": { + "param_names": [ + "backbone.layers.1.attn.relative_position_bias_table", + "backbone.layers.1.attn.qkv.weight", + "backbone.layers.1.attn.proj.weight", + "backbone.layers.1.ffn.layers.0.0.weight", + "backbone.layers.1.ffn.layers.1.weight" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.05 + }, + "layer_3_no_decay": { + "param_names": [ + "backbone.layers.2.gamma_1", + "backbone.layers.2.gamma_2", + "backbone.layers.2.ln1.weight", + "backbone.layers.2.ln1.bias", + "backbone.layers.2.attn.qkv.bias", + "backbone.layers.2.attn.proj.bias", + "backbone.layers.2.ln2.weight", + "backbone.layers.2.ln2.bias", + "backbone.layers.2.ffn.layers.0.0.bias", + "backbone.layers.2.ffn.layers.1.bias" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.0 + }, + "layer_3_decay": { + "param_names": [ + "backbone.layers.2.attn.relative_position_bias_table", + "backbone.layers.2.attn.qkv.weight", + "backbone.layers.2.attn.proj.weight", + "backbone.layers.2.ffn.layers.0.0.weight", + "backbone.layers.2.ffn.layers.1.weight" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.05 + }, + "layer_4_no_decay": { + "param_names": [ + "backbone.layers.3.gamma_1", + "backbone.layers.3.gamma_2", + "backbone.layers.3.ln1.weight", + "backbone.layers.3.ln1.bias", + "backbone.layers.3.attn.qkv.bias", + "backbone.layers.3.attn.proj.bias", + "backbone.layers.3.ln2.weight", + "backbone.layers.3.ln2.bias", + "backbone.layers.3.ffn.layers.0.0.bias", + "backbone.layers.3.ffn.layers.1.bias" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.0 + }, + "layer_4_decay": { + "param_names": [ + "backbone.layers.3.attn.relative_position_bias_table", + "backbone.layers.3.attn.qkv.weight", + "backbone.layers.3.attn.proj.weight", + "backbone.layers.3.ffn.layers.0.0.weight", + "backbone.layers.3.ffn.layers.1.weight" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.05 + }, + "layer_5_no_decay": { + "param_names": [ + "backbone.layers.4.gamma_1", + "backbone.layers.4.gamma_2", + "backbone.layers.4.ln1.weight", + "backbone.layers.4.ln1.bias", + "backbone.layers.4.attn.qkv.bias", + "backbone.layers.4.attn.proj.bias", + "backbone.layers.4.ln2.weight", + "backbone.layers.4.ln2.bias", + "backbone.layers.4.ffn.layers.0.0.bias", + "backbone.layers.4.ffn.layers.1.bias" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.0 + }, + "layer_5_decay": { + "param_names": [ + "backbone.layers.4.attn.relative_position_bias_table", + "backbone.layers.4.attn.qkv.weight", + "backbone.layers.4.attn.proj.weight", + "backbone.layers.4.ffn.layers.0.0.weight", + "backbone.layers.4.ffn.layers.1.weight" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.05 + }, + "layer_6_no_decay": { + "param_names": [ + "backbone.layers.5.gamma_1", + "backbone.layers.5.gamma_2", + "backbone.layers.5.ln1.weight", + "backbone.layers.5.ln1.bias", + "backbone.layers.5.attn.qkv.bias", + "backbone.layers.5.attn.proj.bias", + "backbone.layers.5.ln2.weight", + "backbone.layers.5.ln2.bias", + "backbone.layers.5.ffn.layers.0.0.bias", + "backbone.layers.5.ffn.layers.1.bias" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.0 + }, + "layer_6_decay": { + "param_names": [ + "backbone.layers.5.attn.relative_position_bias_table", + "backbone.layers.5.attn.qkv.weight", + "backbone.layers.5.attn.proj.weight", + "backbone.layers.5.ffn.layers.0.0.weight", + "backbone.layers.5.ffn.layers.1.weight" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.05 + }, + "layer_7_no_decay": { + "param_names": [ + "backbone.layers.6.gamma_1", + "backbone.layers.6.gamma_2", + "backbone.layers.6.ln1.weight", + "backbone.layers.6.ln1.bias", + "backbone.layers.6.attn.qkv.bias", + "backbone.layers.6.attn.proj.bias", + "backbone.layers.6.ln2.weight", + "backbone.layers.6.ln2.bias", + "backbone.layers.6.ffn.layers.0.0.bias", + "backbone.layers.6.ffn.layers.1.bias" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.0 + }, + "layer_7_decay": { + "param_names": [ + "backbone.layers.6.attn.relative_position_bias_table", + "backbone.layers.6.attn.qkv.weight", + "backbone.layers.6.attn.proj.weight", + "backbone.layers.6.ffn.layers.0.0.weight", + "backbone.layers.6.ffn.layers.1.weight" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.05 + }, + "layer_8_no_decay": { + "param_names": [ + "backbone.layers.7.gamma_1", + "backbone.layers.7.gamma_2", + "backbone.layers.7.ln1.weight", + "backbone.layers.7.ln1.bias", + "backbone.layers.7.attn.qkv.bias", + "backbone.layers.7.attn.proj.bias", + "backbone.layers.7.ln2.weight", + "backbone.layers.7.ln2.bias", + "backbone.layers.7.ffn.layers.0.0.bias", + "backbone.layers.7.ffn.layers.1.bias" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.0 + }, + "layer_8_decay": { + "param_names": [ + "backbone.layers.7.attn.relative_position_bias_table", + "backbone.layers.7.attn.qkv.weight", + "backbone.layers.7.attn.proj.weight", + "backbone.layers.7.ffn.layers.0.0.weight", + "backbone.layers.7.ffn.layers.1.weight" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.05 + }, + "layer_9_no_decay": { + "param_names": [ + "backbone.layers.8.gamma_1", + "backbone.layers.8.gamma_2", + "backbone.layers.8.ln1.weight", + "backbone.layers.8.ln1.bias", + "backbone.layers.8.attn.qkv.bias", + "backbone.layers.8.attn.proj.bias", + "backbone.layers.8.ln2.weight", + "backbone.layers.8.ln2.bias", + "backbone.layers.8.ffn.layers.0.0.bias", + "backbone.layers.8.ffn.layers.1.bias" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.0 + }, + "layer_9_decay": { + "param_names": [ + "backbone.layers.8.attn.relative_position_bias_table", + "backbone.layers.8.attn.qkv.weight", + "backbone.layers.8.attn.proj.weight", + "backbone.layers.8.ffn.layers.0.0.weight", + "backbone.layers.8.ffn.layers.1.weight" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.05 + }, + "layer_10_no_decay": { + "param_names": [ + "backbone.layers.9.gamma_1", + "backbone.layers.9.gamma_2", + "backbone.layers.9.ln1.weight", + "backbone.layers.9.ln1.bias", + "backbone.layers.9.attn.qkv.bias", + "backbone.layers.9.attn.proj.bias", + "backbone.layers.9.ln2.weight", + "backbone.layers.9.ln2.bias", + "backbone.layers.9.ffn.layers.0.0.bias", + "backbone.layers.9.ffn.layers.1.bias" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.0 + }, + "layer_10_decay": { + "param_names": [ + "backbone.layers.9.attn.relative_position_bias_table", + "backbone.layers.9.attn.qkv.weight", + "backbone.layers.9.attn.proj.weight", + "backbone.layers.9.ffn.layers.0.0.weight", + "backbone.layers.9.ffn.layers.1.weight" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.05 + }, + "layer_11_no_decay": { + "param_names": [ + "backbone.layers.10.gamma_1", + "backbone.layers.10.gamma_2", + "backbone.layers.10.ln1.weight", + "backbone.layers.10.ln1.bias", + "backbone.layers.10.attn.qkv.bias", + "backbone.layers.10.attn.proj.bias", + "backbone.layers.10.ln2.weight", + "backbone.layers.10.ln2.bias", + "backbone.layers.10.ffn.layers.0.0.bias", + "backbone.layers.10.ffn.layers.1.bias" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.0 + }, + "layer_11_decay": { + "param_names": [ + "backbone.layers.10.attn.relative_position_bias_table", + "backbone.layers.10.attn.qkv.weight", + "backbone.layers.10.attn.proj.weight", + "backbone.layers.10.ffn.layers.0.0.weight", + "backbone.layers.10.ffn.layers.1.weight" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.05 + }, + "layer_12_no_decay": { + "param_names": [ + "backbone.layers.11.gamma_1", + "backbone.layers.11.gamma_2", + "backbone.layers.11.ln1.weight", + "backbone.layers.11.ln1.bias", + "backbone.layers.11.attn.qkv.bias", + "backbone.layers.11.attn.proj.bias", + "backbone.layers.11.ln2.weight", + "backbone.layers.11.ln2.bias", + "backbone.layers.11.ffn.layers.0.0.bias", + "backbone.layers.11.ffn.layers.1.bias" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.0 + }, + "layer_12_decay": { + "param_names": [ + "backbone.layers.11.attn.relative_position_bias_table", + "backbone.layers.11.attn.qkv.weight", + "backbone.layers.11.attn.proj.weight", + "backbone.layers.11.ffn.layers.0.0.weight", + "backbone.layers.11.ffn.layers.1.weight" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.05 + }, + "layer_13_decay": { + "param_names": [ + "neck.upsample_4x.0.weight", + "neck.upsample_4x.3.weight", + "neck.upsample_2x.0.weight", + "decode_head.conv_seg.weight", + "decode_head.psp_modules.0.1.conv.weight", + "decode_head.psp_modules.1.1.conv.weight", + "decode_head.psp_modules.2.1.conv.weight", + "decode_head.psp_modules.3.1.conv.weight", + "decode_head.bottleneck.conv.weight", + "decode_head.lateral_convs.0.conv.weight", + "decode_head.lateral_convs.1.conv.weight", + "decode_head.lateral_convs.2.conv.weight", + "decode_head.fpn_convs.0.conv.weight", + "decode_head.fpn_convs.1.conv.weight", + "decode_head.fpn_convs.2.conv.weight", + "decode_head.fpn_bottleneck.conv.weight", + "auxiliary_head.conv_seg.weight", + "auxiliary_head.convs.0.conv.weight" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.05 + }, + "layer_13_no_decay": { + "param_names": [ + "neck.upsample_4x.0.bias", + "neck.upsample_4x.1.weight", + "neck.upsample_4x.1.bias", + "neck.upsample_4x.3.bias", + "neck.upsample_2x.0.bias", + "decode_head.conv_seg.bias", + "decode_head.psp_modules.0.1.bn.weight", + "decode_head.psp_modules.0.1.bn.bias", + "decode_head.psp_modules.1.1.bn.weight", + "decode_head.psp_modules.1.1.bn.bias", + "decode_head.psp_modules.2.1.bn.weight", + "decode_head.psp_modules.2.1.bn.bias", + "decode_head.psp_modules.3.1.bn.weight", + "decode_head.psp_modules.3.1.bn.bias", + "decode_head.bottleneck.bn.weight", + "decode_head.bottleneck.bn.bias", + "decode_head.lateral_convs.0.bn.weight", + "decode_head.lateral_convs.0.bn.bias", + "decode_head.lateral_convs.1.bn.weight", + "decode_head.lateral_convs.1.bn.bias", + "decode_head.lateral_convs.2.bn.weight", + "decode_head.lateral_convs.2.bn.bias", + "decode_head.fpn_convs.0.bn.weight", + "decode_head.fpn_convs.0.bn.bias", + "decode_head.fpn_convs.1.bn.weight", + "decode_head.fpn_convs.1.bn.bias", + "decode_head.fpn_convs.2.bn.weight", + "decode_head.fpn_convs.2.bn.bias", + "decode_head.fpn_bottleneck.bn.weight", + "decode_head.fpn_bottleneck.bn.bias", + "auxiliary_head.conv_seg.bias", + "auxiliary_head.convs.0.bn.weight", + "auxiliary_head.convs.0.bn.bias" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.0 + } +} +04/19 00:38:49 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1765, in train + self.load_or_resume() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1699, in load_or_resume + self.load_checkpoint(self._load_from) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 2127, in load_checkpoint + checkpoint = _load_checkpoint(filename, map_location=map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError: can not be found. +Loads checkpoint by local backend from path: +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1765, in train + self.load_or_resume() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1699, in load_or_resume + self.load_checkpoint(self._load_from) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 2127, in load_checkpoint + checkpoint = _load_checkpoint(filename, map_location=map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError: can not be found. +Loads checkpoint by local backend from path: +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1765, in train + self.load_or_resume() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1699, in load_or_resume + self.load_checkpoint(self._load_from) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 2127, in load_checkpoint + checkpoint = _load_checkpoint(filename, map_location=map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError: can not be found. +Loads checkpoint by local backend from path: +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1765, in train + self.load_or_resume() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1699, in load_or_resume + self.load_checkpoint(self._load_from) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 2127, in load_checkpoint + checkpoint = _load_checkpoint(filename, map_location=map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError: can not be found. +ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 1) local_rank: 0 (pid: 55633) of binary: /opt/conda/bin/python +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 245, in launch_agent + raise ChildFailedError( +torch.distributed.elastic.multiprocessing.errors.ChildFailedError: +============================================================ +tools/train.py FAILED +------------------------------------------------------------ +Failures: +[1]: + time : 2024-04-19_00:38:51 + host : 06540a9609a4 + rank : 1 (local_rank: 1) + exitcode : 1 (pid: 55634) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[2]: + time : 2024-04-19_00:38:51 + host : 06540a9609a4 + rank : 2 (local_rank: 2) + exitcode : 1 (pid: 55635) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[3]: + time : 2024-04-19_00:38:51 + host : 06540a9609a4 + rank : 3 (local_rank: 3) + exitcode : 1 (pid: 55636) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +------------------------------------------------------------ +Root Cause (first observed failure): +[0]: + time : 2024-04-19_00:38:51 + host : 06540a9609a4 + rank : 0 (local_rank: 0) + exitcode : 1 (pid: 55633) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +============================================================ +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/19 00:41:23 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 856117106 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 856117106 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/19 00:41:23 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +fp16 = dict(loss_scale='dynamic') +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = '' +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=768, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + num_convs=1, + type='FCNHead'), + backbone=dict( + act_cfg=dict(type='GELU'), + attn_drop_rate=0.0, + drop_path_rate=0.1, + embed_dims=768, + img_size=( + 512, + 512, + ), + in_channels=3, + init_values=1.0, + mlp_ratio=4, + norm_cfg=dict(eps=1e-06, type='LN'), + norm_eval=False, + num_heads=12, + num_layers=12, + out_indices=[ + 3, + 5, + 7, + 11, + ], + patch_size=16, + type='MAE'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=768, + dropout_ratio=0.1, + in_channels=[ + 768, + 768, + 768, + 768, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + neck=dict( + embed_dim=768, rescales=[ + 4, + 2, + 1, + 0.5, + ], type='Feature2Pyramid'), + pretrained='/workspaces/mmsegmentation-1/converted_model.pth', + test_cfg=dict(crop_size=( + 512, + 512, + ), mode='slide', stride=( + 341, + 341, + )), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + constructor='LayerDecayOptimizerConstructor', + optimizer=dict( + betas=( + 0.9, + 0.999, + ), lr=0.0001, type='AdamW', weight_decay=0.05), + paramwise_cfg=dict(layer_decay_rate=0.65, num_layers=12), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, by_epoch=False, end=1500, start_factor=1e-06, + type='LinearLR'), + dict( + begin=1500, + by_epoch=False, + end=160000, + eta_min=0.0, + power=1.0, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=2, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = '/workspaces/mmsegmentation-1/work_dirs/MAE4000' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/19 00:41:26 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65}set param backbone.cls_token as id 0 + +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.cls_token as id 0 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.pos_embed as id 0 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1set param backbone.patch_embed.projection.bias as id 0 + +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1set param backbone.layers.0.ffn.layers.1.bias as id 1 + +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.0.attn.proj.weight as id 1set param backbone.layers.1.gamma_2 as id 2 + +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1set param backbone.layers.1.attn.proj.weight as id 2 + +set param backbone.layers.0.ffn.layers.0.0.bias as id 1set param backbone.layers.1.attn.proj.bias as id 2 + +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2set param backbone.layers.1.gamma_1 as id 2 + +set param backbone.layers.1.gamma_2 as id 2set param backbone.layers.1.ffn.layers.1.weight as id 2 + +set param backbone.layers.1.ln1.weight as id 2set param backbone.layers.1.ffn.layers.1.bias as id 2 + +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2set param backbone.layers.2.gamma_1 as id 3 + +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.2.ln1.weight as id 3set param backbone.layers.1.attn.qkv.bias as id 2 + +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.1.attn.proj.weight as id 2set param backbone.layers.2.attn.relative_position_bias_table as id 3 + +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3set param backbone.layers.1.ln2.weight as id 2 + +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.2.ln2.weight as id 3set param backbone.layers.1.ffn.layers.0.0.bias as id 2 + +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +Param groups = { + "layer_0_decay": { + "param_names": [ + "backbone.cls_token", + "backbone.pos_embed", + "backbone.patch_embed.projection.weight" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.05 + }, + "layer_0_no_decay": { + "param_names": [ + "backbone.patch_embed.projection.bias" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.0 + }, + "layer_1_no_decay": { + "param_names": [ + "backbone.layers.0.gamma_1", + "backbone.layers.0.gamma_2", + "backbone.layers.0.ln1.weight", + "backbone.layers.0.ln1.bias", + "backbone.layers.0.attn.qkv.bias", + "backbone.layers.0.attn.proj.bias", + "backbone.layers.0.ln2.weight", + "backbone.layers.0.ln2.bias", + "backbone.layers.0.ffn.layers.0.0.bias", + "backbone.layers.0.ffn.layers.1.bias" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.0 + }, + "layer_1_decay": { + "param_names": [ + "backbone.layers.0.attn.relative_position_bias_table", + "backbone.layers.0.attn.qkv.weight", + "backbone.layers.0.attn.proj.weight", + "backbone.layers.0.ffn.layers.0.0.weight", + "backbone.layers.0.ffn.layers.1.weight" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.05 + }, + "layer_2_no_decay": { + "param_names": [ + "backbone.layers.1.gamma_1", + "backbone.layers.1.gamma_2", + "backbone.layers.1.ln1.weight", + "backbone.layers.1.ln1.bias", + "backbone.layers.1.attn.qkv.bias", + "backbone.layers.1.attn.proj.bias", + "backbone.layers.1.ln2.weight", + "backbone.layers.1.ln2.bias", + "backbone.layers.1.ffn.layers.0.0.bias", + "backbone.layers.1.ffn.layers.1.bias" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.0 + }, + "layer_2_decay": { + "param_names": [ + "backbone.layers.1.attn.relative_position_bias_table", + "backbone.layers.1.attn.qkv.weight", + "backbone.layers.1.attn.proj.weight", + "backbone.layers.1.ffn.layers.0.0.weight", + "backbone.layers.1.ffn.layers.1.weight" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.05 + }, + "layer_3_no_decay": { + "param_names": [ + "backbone.layers.2.gamma_1", + "backbone.layers.2.gamma_2", + "backbone.layers.2.ln1.weight", + "backbone.layers.2.ln1.bias", + "backbone.layers.2.attn.qkv.bias", + "backbone.layers.2.attn.proj.bias", + "backbone.layers.2.ln2.weight", + "backbone.layers.2.ln2.bias", + "backbone.layers.2.ffn.layers.0.0.bias", + "backbone.layers.2.ffn.layers.1.bias" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.0 + }, + "layer_3_decay": { + "param_names": [ + "backbone.layers.2.attn.relative_position_bias_table", + "backbone.layers.2.attn.qkv.weight", + "backbone.layers.2.attn.proj.weight", + "backbone.layers.2.ffn.layers.0.0.weight", + "backbone.layers.2.ffn.layers.1.weight" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.05 + }, + "layer_4_no_decay": { + "param_names": [ + "backbone.layers.3.gamma_1", + "backbone.layers.3.gamma_2", + "backbone.layers.3.ln1.weight", + "backbone.layers.3.ln1.bias", + "backbone.layers.3.attn.qkv.bias", + "backbone.layers.3.attn.proj.bias", + "backbone.layers.3.ln2.weight", + "backbone.layers.3.ln2.bias", + "backbone.layers.3.ffn.layers.0.0.bias", + "backbone.layers.3.ffn.layers.1.bias" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.0 + }, + "layer_4_decay": { + "param_names": [ + "backbone.layers.3.attn.relative_position_bias_table", + "backbone.layers.3.attn.qkv.weight", + "backbone.layers.3.attn.proj.weight", + "backbone.layers.3.ffn.layers.0.0.weight", + "backbone.layers.3.ffn.layers.1.weight" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.05 + }, + "layer_5_no_decay": { + "param_names": [ + "backbone.layers.4.gamma_1", + "backbone.layers.4.gamma_2", + "backbone.layers.4.ln1.weight", + "backbone.layers.4.ln1.bias", + "backbone.layers.4.attn.qkv.bias", + "backbone.layers.4.attn.proj.bias", + "backbone.layers.4.ln2.weight", + "backbone.layers.4.ln2.bias", + "backbone.layers.4.ffn.layers.0.0.bias", + "backbone.layers.4.ffn.layers.1.bias" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.0 + }, + "layer_5_decay": { + "param_names": [ + "backbone.layers.4.attn.relative_position_bias_table", + "backbone.layers.4.attn.qkv.weight", + "backbone.layers.4.attn.proj.weight", + "backbone.layers.4.ffn.layers.0.0.weight", + "backbone.layers.4.ffn.layers.1.weight" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.05 + }, + "layer_6_no_decay": { + "param_names": [ + "backbone.layers.5.gamma_1", + "backbone.layers.5.gamma_2", + "backbone.layers.5.ln1.weight", + "backbone.layers.5.ln1.bias", + "backbone.layers.5.attn.qkv.bias", + "backbone.layers.5.attn.proj.bias", + "backbone.layers.5.ln2.weight", + "backbone.layers.5.ln2.bias", + "backbone.layers.5.ffn.layers.0.0.bias", + "backbone.layers.5.ffn.layers.1.bias" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.0 + }, + "layer_6_decay": { + "param_names": [ + "backbone.layers.5.attn.relative_position_bias_table", + "backbone.layers.5.attn.qkv.weight", + "backbone.layers.5.attn.proj.weight", + "backbone.layers.5.ffn.layers.0.0.weight", + "backbone.layers.5.ffn.layers.1.weight" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.05 + }, + "layer_7_no_decay": { + "param_names": [ + "backbone.layers.6.gamma_1", + "backbone.layers.6.gamma_2", + "backbone.layers.6.ln1.weight", + "backbone.layers.6.ln1.bias", + "backbone.layers.6.attn.qkv.bias", + "backbone.layers.6.attn.proj.bias", + "backbone.layers.6.ln2.weight", + "backbone.layers.6.ln2.bias", + "backbone.layers.6.ffn.layers.0.0.bias", + "backbone.layers.6.ffn.layers.1.bias" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.0 + }, + "layer_7_decay": { + "param_names": [ + "backbone.layers.6.attn.relative_position_bias_table", + "backbone.layers.6.attn.qkv.weight", + "backbone.layers.6.attn.proj.weight", + "backbone.layers.6.ffn.layers.0.0.weight", + "backbone.layers.6.ffn.layers.1.weight" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.05 + }, + "layer_8_no_decay": { + "param_names": [ + "backbone.layers.7.gamma_1", + "backbone.layers.7.gamma_2", + "backbone.layers.7.ln1.weight", + "backbone.layers.7.ln1.bias", + "backbone.layers.7.attn.qkv.bias", + "backbone.layers.7.attn.proj.bias", + "backbone.layers.7.ln2.weight", + "backbone.layers.7.ln2.bias", + "backbone.layers.7.ffn.layers.0.0.bias", + "backbone.layers.7.ffn.layers.1.bias" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.0 + }, + "layer_8_decay": { + "param_names": [ + "backbone.layers.7.attn.relative_position_bias_table", + "backbone.layers.7.attn.qkv.weight", + "backbone.layers.7.attn.proj.weight", + "backbone.layers.7.ffn.layers.0.0.weight", + "backbone.layers.7.ffn.layers.1.weight" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.05 + }, + "layer_9_no_decay": { + "param_names": [ + "backbone.layers.8.gamma_1", + "backbone.layers.8.gamma_2", + "backbone.layers.8.ln1.weight", + "backbone.layers.8.ln1.bias", + "backbone.layers.8.attn.qkv.bias", + "backbone.layers.8.attn.proj.bias", + "backbone.layers.8.ln2.weight", + "backbone.layers.8.ln2.bias", + "backbone.layers.8.ffn.layers.0.0.bias", + "backbone.layers.8.ffn.layers.1.bias" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.0 + }, + "layer_9_decay": { + "param_names": [ + "backbone.layers.8.attn.relative_position_bias_table", + "backbone.layers.8.attn.qkv.weight", + "backbone.layers.8.attn.proj.weight", + "backbone.layers.8.ffn.layers.0.0.weight", + "backbone.layers.8.ffn.layers.1.weight" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.05 + }, + "layer_10_no_decay": { + "param_names": [ + "backbone.layers.9.gamma_1", + "backbone.layers.9.gamma_2", + "backbone.layers.9.ln1.weight", + "backbone.layers.9.ln1.bias", + "backbone.layers.9.attn.qkv.bias", + "backbone.layers.9.attn.proj.bias", + "backbone.layers.9.ln2.weight", + "backbone.layers.9.ln2.bias", + "backbone.layers.9.ffn.layers.0.0.bias", + "backbone.layers.9.ffn.layers.1.bias" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.0 + }, + "layer_10_decay": { + "param_names": [ + "backbone.layers.9.attn.relative_position_bias_table", + "backbone.layers.9.attn.qkv.weight", + "backbone.layers.9.attn.proj.weight", + "backbone.layers.9.ffn.layers.0.0.weight", + "backbone.layers.9.ffn.layers.1.weight" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.05 + }, + "layer_11_no_decay": { + "param_names": [ + "backbone.layers.10.gamma_1", + "backbone.layers.10.gamma_2", + "backbone.layers.10.ln1.weight", + "backbone.layers.10.ln1.bias", + "backbone.layers.10.attn.qkv.bias", + "backbone.layers.10.attn.proj.bias", + "backbone.layers.10.ln2.weight", + "backbone.layers.10.ln2.bias", + "backbone.layers.10.ffn.layers.0.0.bias", + "backbone.layers.10.ffn.layers.1.bias" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.0 + }, + "layer_11_decay": { + "param_names": [ + "backbone.layers.10.attn.relative_position_bias_table", + "backbone.layers.10.attn.qkv.weight", + "backbone.layers.10.attn.proj.weight", + "backbone.layers.10.ffn.layers.0.0.weight", + "backbone.layers.10.ffn.layers.1.weight" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.05 + }, + "layer_12_no_decay": { + "param_names": [ + "backbone.layers.11.gamma_1", + "backbone.layers.11.gamma_2", + "backbone.layers.11.ln1.weight", + "backbone.layers.11.ln1.bias", + "backbone.layers.11.attn.qkv.bias", + "backbone.layers.11.attn.proj.bias", + "backbone.layers.11.ln2.weight", + "backbone.layers.11.ln2.bias", + "backbone.layers.11.ffn.layers.0.0.bias", + "backbone.layers.11.ffn.layers.1.bias" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.0 + }, + "layer_12_decay": { + "param_names": [ + "backbone.layers.11.attn.relative_position_bias_table", + "backbone.layers.11.attn.qkv.weight", + "backbone.layers.11.attn.proj.weight", + "backbone.layers.11.ffn.layers.0.0.weight", + "backbone.layers.11.ffn.layers.1.weight" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.05 + }, + "layer_13_decay": { + "param_names": [ + "neck.upsample_4x.0.weight", + "neck.upsample_4x.3.weight", + "neck.upsample_2x.0.weight", + "decode_head.conv_seg.weight", + "decode_head.psp_modules.0.1.conv.weight", + "decode_head.psp_modules.1.1.conv.weight", + "decode_head.psp_modules.2.1.conv.weight", + "decode_head.psp_modules.3.1.conv.weight", + "decode_head.bottleneck.conv.weight", + "decode_head.lateral_convs.0.conv.weight", + "decode_head.lateral_convs.1.conv.weight", + "decode_head.lateral_convs.2.conv.weight", + "decode_head.fpn_convs.0.conv.weight", + "decode_head.fpn_convs.1.conv.weight", + "decode_head.fpn_convs.2.conv.weight", + "decode_head.fpn_bottleneck.conv.weight", + "auxiliary_head.conv_seg.weight", + "auxiliary_head.convs.0.conv.weight" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.05 + }, + "layer_13_no_decay": { + "param_names": [ + "neck.upsample_4x.0.bias", + "neck.upsample_4x.1.weight", + "neck.upsample_4x.1.bias", + "neck.upsample_4x.3.bias", + "neck.upsample_2x.0.bias", + "decode_head.conv_seg.bias", + "decode_head.psp_modules.0.1.bn.weight", + "decode_head.psp_modules.0.1.bn.bias", + "decode_head.psp_modules.1.1.bn.weight", + "decode_head.psp_modules.1.1.bn.bias", + "decode_head.psp_modules.2.1.bn.weight", + "decode_head.psp_modules.2.1.bn.bias", + "decode_head.psp_modules.3.1.bn.weight", + "decode_head.psp_modules.3.1.bn.bias", + "decode_head.bottleneck.bn.weight", + "decode_head.bottleneck.bn.bias", + "decode_head.lateral_convs.0.bn.weight", + "decode_head.lateral_convs.0.bn.bias", + "decode_head.lateral_convs.1.bn.weight", + "decode_head.lateral_convs.1.bn.bias", + "decode_head.lateral_convs.2.bn.weight", + "decode_head.lateral_convs.2.bn.bias", + "decode_head.fpn_convs.0.bn.weight", + "decode_head.fpn_convs.0.bn.bias", + "decode_head.fpn_convs.1.bn.weight", + "decode_head.fpn_convs.1.bn.bias", + "decode_head.fpn_convs.2.bn.weight", + "decode_head.fpn_convs.2.bn.bias", + "decode_head.fpn_bottleneck.bn.weight", + "decode_head.fpn_bottleneck.bn.bias", + "auxiliary_head.conv_seg.bias", + "auxiliary_head.convs.0.bn.weight", + "auxiliary_head.convs.0.bn.bias" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.0 + } +} +04/19 00:41:27 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1765, in train + self.load_or_resume() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1699, in load_or_resume + self.load_checkpoint(self._load_from) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 2127, in load_checkpoint + checkpoint = _load_checkpoint(filename, map_location=map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError: can not be found. +Loads checkpoint by local backend from path: +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1765, in train + self.load_or_resume() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1699, in load_or_resume + self.load_checkpoint(self._load_from) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 2127, in load_checkpoint + checkpoint = _load_checkpoint(filename, map_location=map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError: can not be found. +Loads checkpoint by local backend from path: +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1765, in train + self.load_or_resume() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1699, in load_or_resume + self.load_checkpoint(self._load_from) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 2127, in load_checkpoint + checkpoint = _load_checkpoint(filename, map_location=map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError: can not be found. +Loads checkpoint by local backend from path: +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1765, in train + self.load_or_resume() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1699, in load_or_resume + self.load_checkpoint(self._load_from) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 2127, in load_checkpoint + checkpoint = _load_checkpoint(filename, map_location=map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError: can not be found. +ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 1) local_rank: 0 (pid: 58086) of binary: /opt/conda/bin/python +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 245, in launch_agent + raise ChildFailedError( +torch.distributed.elastic.multiprocessing.errors.ChildFailedError: +============================================================ +tools/train.py FAILED +------------------------------------------------------------ +Failures: +[1]: + time : 2024-04-19_00:41:29 + host : 06540a9609a4 + rank : 1 (local_rank: 1) + exitcode : 1 (pid: 58087) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[2]: + time : 2024-04-19_00:41:29 + host : 06540a9609a4 + rank : 2 (local_rank: 2) + exitcode : 1 (pid: 58088) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[3]: + time : 2024-04-19_00:41:29 + host : 06540a9609a4 + rank : 3 (local_rank: 3) + exitcode : 1 (pid: 58089) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +------------------------------------------------------------ +Root Cause (first observed failure): +[0]: + time : 2024-04-19_00:41:29 + host : 06540a9609a4 + rank : 0 (local_rank: 0) + exitcode : 1 (pid: 58086) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +============================================================ +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +usage: train.py [-h] [--work-dir WORK_DIR] [--resume] [--amp] + [--cfg-options CFG_OPTIONS [CFG_OPTIONS ...]] + [--launcher {none,pytorch,slurm,mpi}] + [--local_rank LOCAL_RANK] + config +train.py: error: argument --resume: ignored explicit argument 'auto' +usage: train.py [-h] [--work-dir WORK_DIR] [--resume] [--amp] + [--cfg-options CFG_OPTIONS [CFG_OPTIONS ...]] + [--launcher {none,pytorch,slurm,mpi}] + [--local_rank LOCAL_RANK] + config +train.py: error: argument --resume: ignored explicit argument 'auto' +usage: train.py [-h] [--work-dir WORK_DIR] [--resume] [--amp] + [--cfg-options CFG_OPTIONS [CFG_OPTIONS ...]] + [--launcher {none,pytorch,slurm,mpi}] + [--local_rank LOCAL_RANK] + config +train.py: error: argument --resume: ignored explicit argument 'auto' +usage: train.py [-h] [--work-dir WORK_DIR] [--resume] [--amp] + [--cfg-options CFG_OPTIONS [CFG_OPTIONS ...]] + [--launcher {none,pytorch,slurm,mpi}] + [--local_rank LOCAL_RANK] + config +train.py: error: argument --resume: ignored explicit argument 'auto' +ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 2) local_rank: 0 (pid: 59317) of binary: /opt/conda/bin/python +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 245, in launch_agent + raise ChildFailedError( +torch.distributed.elastic.multiprocessing.errors.ChildFailedError: +============================================================ +tools/train.py FAILED +------------------------------------------------------------ +Failures: +[1]: + time : 2024-04-19_00:42:18 + host : 06540a9609a4 + rank : 1 (local_rank: 1) + exitcode : 2 (pid: 59318) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[2]: + time : 2024-04-19_00:42:18 + host : 06540a9609a4 + rank : 2 (local_rank: 2) + exitcode : 2 (pid: 59319) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[3]: + time : 2024-04-19_00:42:18 + host : 06540a9609a4 + rank : 3 (local_rank: 3) + exitcode : 2 (pid: 59320) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +------------------------------------------------------------ +Root Cause (first observed failure): +[0]: + time : 2024-04-19_00:42:18 + host : 06540a9609a4 + rank : 0 (local_rank: 0) + exitcode : 2 (pid: 59317) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +============================================================ +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +usage: train.py [-h] [--work-dir WORK_DIR] [--resume] [--amp] + [--cfg-options CFG_OPTIONS [CFG_OPTIONS ...]] + [--launcher {none,pytorch,slurm,mpi}] + [--local_rank LOCAL_RANK] + config +train.py: error: unrecognized arguments: --load_from= +usage: train.py [-h] [--work-dir WORK_DIR] [--resume] [--amp] + [--cfg-options CFG_OPTIONS [CFG_OPTIONS ...]] + [--launcher {none,pytorch,slurm,mpi}] + [--local_rank LOCAL_RANK] + config +train.py: error: unrecognized arguments: --load_from= +usage: train.py [-h] [--work-dir WORK_DIR] [--resume] [--amp] + [--cfg-options CFG_OPTIONS [CFG_OPTIONS ...]] + [--launcher {none,pytorch,slurm,mpi}] + [--local_rank LOCAL_RANK] + config +train.py: error: unrecognized arguments: --load_from= +usage: train.py [-h] [--work-dir WORK_DIR] [--resume] [--amp] + [--cfg-options CFG_OPTIONS [CFG_OPTIONS ...]] + [--launcher {none,pytorch,slurm,mpi}] + [--local_rank LOCAL_RANK] + config +train.py: error: unrecognized arguments: --load_from= +ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 2) local_rank: 0 (pid: 61127) of binary: /opt/conda/bin/python +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 245, in launch_agent + raise ChildFailedError( +torch.distributed.elastic.multiprocessing.errors.ChildFailedError: +============================================================ +tools/train.py FAILED +------------------------------------------------------------ +Failures: +[1]: + time : 2024-04-19_00:44:54 + host : 06540a9609a4 + rank : 1 (local_rank: 1) + exitcode : 2 (pid: 61128) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[2]: + time : 2024-04-19_00:44:54 + host : 06540a9609a4 + rank : 2 (local_rank: 2) + exitcode : 2 (pid: 61129) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[3]: + time : 2024-04-19_00:44:54 + host : 06540a9609a4 + rank : 3 (local_rank: 3) + exitcode : 2 (pid: 61130) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +------------------------------------------------------------ +Root Cause (first observed failure): +[0]: + time : 2024-04-19_00:44:54 + host : 06540a9609a4 + rank : 0 (local_rank: 0) + exitcode : 2 (pid: 61127) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +============================================================ +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/19 00:45:35 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 172032146 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 172032146 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/19 00:45:35 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +fp16 = dict(loss_scale='dynamic') +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = '' +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=768, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + num_convs=1, + type='FCNHead'), + backbone=dict( + act_cfg=dict(type='GELU'), + attn_drop_rate=0.0, + drop_path_rate=0.1, + embed_dims=768, + img_size=( + 512, + 512, + ), + in_channels=3, + init_values=1.0, + mlp_ratio=4, + norm_cfg=dict(eps=1e-06, type='LN'), + norm_eval=False, + num_heads=12, + num_layers=12, + out_indices=[ + 3, + 5, + 7, + 11, + ], + patch_size=16, + type='MAE'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=768, + dropout_ratio=0.1, + in_channels=[ + 768, + 768, + 768, + 768, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + neck=dict( + embed_dim=768, rescales=[ + 4, + 2, + 1, + 0.5, + ], type='Feature2Pyramid'), + pretrained='/workspaces/mmsegmentation-1/converted_model.pth', + test_cfg=dict(crop_size=( + 512, + 512, + ), mode='slide', stride=( + 341, + 341, + )), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + constructor='LayerDecayOptimizerConstructor', + optimizer=dict( + betas=( + 0.9, + 0.999, + ), lr=0.0001, type='AdamW', weight_decay=0.05), + paramwise_cfg=dict(layer_decay_rate=0.65, num_layers=12), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, by_epoch=False, end=1500, start_factor=1e-06, + type='LinearLR'), + dict( + begin=1500, + by_epoch=False, + end=160000, + eta_min=0.0, + power=1.0, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=2, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = '/workspaces/mmsegmentation-1/work_dirs/MAE4000' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/19 00:45:38 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65}set param backbone.layers.2.ln1.weight as id 3 + +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 + +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4set param backbone.cls_token as id 0 + +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.pos_embed as id 0 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.cls_token as id 0set param backbone.layers.3.attn.qkv.weight as id 4 + +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.pos_embed as id 0set param backbone.patch_embed.projection.weight as id 0set param backbone.layers.3.attn.proj.weight as id 4 + + +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.patch_embed.projection.weight as id 0set param backbone.layers.3.ffn.layers.0.0.weight as id 4 + +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4set param backbone.patch_embed.projection.bias as id 0 + +set param backbone.layers.0.gamma_2 as id 1set param backbone.layers.3.ffn.layers.1.weight as id 4 + +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.gamma_1 as id 1 + +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.0.ln1.bias as id 1set param backbone.layers.4.ln1.weight as id 5set param backbone.layers.0.gamma_2 as id 1 + + +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5set param backbone.layers.0.attn.relative_position_bias_table as id 1 + +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1set param backbone.layers.4.attn.proj.weight as id 5 + +set param backbone.layers.4.attn.proj.bias as id 5set param backbone.layers.0.attn.qkv.bias as id 1set param backbone.layers.0.attn.relative_position_bias_table as id 1 + + +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1set param backbone.layers.4.ffn.layers.0.0.weight as id 5 + +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1set param backbone.layers.4.ffn.layers.0.0.bias as id 5 + +set param backbone.layers.0.ln2.weight as id 1set param backbone.layers.4.ffn.layers.1.weight as id 5 + +set param backbone.layers.4.ffn.layers.1.bias as id 5set param backbone.layers.0.attn.proj.weight as id 1set param backbone.layers.0.ln2.bias as id 1 + + +set param backbone.layers.0.attn.proj.bias as id 1set param backbone.layers.5.gamma_1 as id 6 + +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.0.ln2.weight as id 1set param backbone.layers.5.ln1.weight as id 6 + +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.5.ln1.bias as id 6set param backbone.layers.0.ln2.bias as id 1 + +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.0.ffn.layers.1.bias as id 1 + +set param backbone.layers.0.ffn.layers.0.0.bias as id 1set param backbone.layers.5.ln2.weight as id 6 + +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.0.ffn.layers.1.weight as id 1 + +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.1.gamma_2 as id 2set param backbone.layers.0.ffn.layers.1.bias as id 1set param backbone.layers.5.ffn.layers.1.weight as id 6 + + +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.1.ln1.bias as id 2set param backbone.layers.6.gamma_2 as id 7 + +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.6.ln1.weight as id 7set param backbone.layers.1.attn.relative_position_bias_table as id 2 + +set param backbone.layers.1.gamma_2 as id 2set param backbone.layers.6.ln1.bias as id 7 + +set param backbone.layers.1.attn.qkv.weight as id 2set param backbone.layers.6.attn.relative_position_bias_table as id 7 + +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.6.attn.qkv.weight as id 7set param backbone.layers.1.attn.qkv.bias as id 2 + +set param backbone.layers.1.ln1.bias as id 2set param backbone.layers.6.attn.qkv.bias as id 7 + +set param backbone.layers.1.attn.proj.weight as id 2set param backbone.layers.6.attn.proj.weight as id 7set param backbone.layers.1.attn.relative_position_bias_table as id 2 + + +set param backbone.layers.6.attn.proj.bias as id 7set param backbone.layers.1.attn.proj.bias as id 2 + +set param backbone.layers.6.ln2.weight as id 7set param backbone.layers.1.attn.qkv.weight as id 2 + +set param backbone.layers.1.ln2.weight as id 2set param backbone.layers.6.ln2.bias as id 7 + +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.2.gamma_2 as id 3set param backbone.layers.6.ffn.layers.0.0.weight as id 7 + +set param backbone.layers.1.ffn.layers.1.weight as id 2set param backbone.layers.2.ln1.weight as id 3set param backbone.layers.6.ffn.layers.0.0.bias as id 7 + + +set param backbone.layers.1.ffn.layers.1.bias as id 2set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.6.ffn.layers.1.weight as id 7 + +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.7.gamma_1 as id 8set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.gamma_2 as id 3 + +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.ln1.weight as id 3set param backbone.layers.7.ln1.weight as id 8 + +set param backbone.layers.7.ln1.bias as id 8set param backbone.layers.2.ln1.bias as id 3set param backbone.layers.2.attn.proj.weight as id 3 + + +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.2.attn.proj.bias as id 3set param backbone.layers.2.attn.relative_position_bias_table as id 3 + +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.2.ln2.weight as id 3set param backbone.layers.2.attn.qkv.weight as id 3 + +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.2.ln2.bias as id 3set param backbone.layers.7.attn.proj.bias as id 8set param backbone.layers.2.attn.qkv.bias as id 3 + + +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 + +set param backbone.layers.7.ffn.layers.0.0.weight as id 8set param backbone.layers.2.ffn.layers.0.0.bias as id 3set param backbone.layers.2.attn.proj.bias as id 3 + + +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.2.ln2.weight as id 3set param backbone.layers.2.ffn.layers.1.weight as id 3 + +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.2.ln2.bias as id 3set param backbone.layers.2.ffn.layers.1.bias as id 3 + +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3set param backbone.layers.3.gamma_1 as id 4set param backbone.layers.8.ln1.weight as id 9 + + +set param backbone.layers.8.ln1.bias as id 9set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.3.gamma_2 as id 4 + +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.3.ln1.weight as id 4set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.2.ffn.layers.1.weight as id 3 + +set param backbone.layers.3.ln1.bias as id 4set param backbone.layers.8.attn.qkv.bias as id 9 + +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.8.attn.proj.weight as id 9set param backbone.layers.3.attn.relative_position_bias_table as id 4 + +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4set param backbone.layers.8.ln2.weight as id 9 + +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.8.ln2.bias as id 9set param backbone.layers.3.attn.qkv.bias as id 4 + +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.3.ln1.bias as id 4set param backbone.layers.8.ffn.layers.0.0.bias as id 9set param backbone.layers.3.attn.proj.weight as id 4 + + +set param backbone.layers.8.ffn.layers.1.weight as id 9set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.8.ffn.layers.1.bias as id 9 + +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.ln2.weight as id 4set param backbone.layers.9.gamma_1 as id 10 + +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.9.gamma_2 as id 10set param backbone.layers.3.ln2.bias as id 4 + +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.3.attn.proj.bias as id 4set param backbone.layers.9.attn.relative_position_bias_table as id 10 + +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.9.attn.qkv.weight as id 10set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 + +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.3.ffn.layers.1.weight as id 4set param backbone.layers.9.attn.proj.bias as id 10 + +set param backbone.layers.3.ffn.layers.1.bias as id 4set param backbone.layers.9.ln2.weight as id 10set param backbone.layers.3.ffn.layers.0.0.weight as id 4 + + +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10set param backbone.layers.4.gamma_1 as id 5 + +set param backbone.layers.9.ffn.layers.0.0.bias as id 10set param backbone.layers.3.ffn.layers.1.weight as id 4 + +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.gamma_1 as id 5set param backbone.layers.10.gamma_2 as id 11 + +set param backbone.layers.4.attn.relative_position_bias_table as id 5set param backbone.layers.10.ln1.weight as id 11 + +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5set param backbone.layers.10.attn.qkv.weight as id 11 + +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.4.attn.proj.weight as id 5set param backbone.layers.10.attn.proj.weight as id 11set param backbone.layers.4.attn.relative_position_bias_table as id 5 + + +set param backbone.layers.10.attn.proj.bias as id 11set param backbone.layers.4.attn.proj.bias as id 5 + +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.4.ln2.weight as id 5set param backbone.layers.4.attn.qkv.bias as id 5set param backbone.layers.10.ln2.bias as id 11 + + +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5set param backbone.layers.10.ffn.layers.0.0.weight as id 11 + +set param backbone.layers.10.ffn.layers.0.0.bias as id 11set param backbone.layers.4.attn.proj.bias as id 5 + +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.4.ln2.weight as id 5set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 + +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.4.ffn.layers.1.bias as id 5set param backbone.layers.11.ln1.weight as id 12set param backbone.layers.4.ffn.layers.0.0.weight as id 5 + + +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12set param backbone.layers.4.ffn.layers.0.0.bias as id 5set param backbone.layers.5.gamma_1 as id 6 + + +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.4.ffn.layers.1.weight as id 5set param backbone.layers.5.gamma_2 as id 6set param backbone.layers.11.attn.qkv.bias as id 12 + + +set param backbone.layers.4.ffn.layers.1.bias as id 5set param backbone.layers.11.attn.proj.weight as id 12 + +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.5.ln1.bias as id 6set param backbone.layers.11.ln2.weight as id 12set param backbone.layers.5.gamma_1 as id 6 + + +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.5.attn.relative_position_bias_table as id 6set param backbone.layers.5.gamma_2 as id 6 + +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.5.attn.qkv.weight as id 6set param backbone.layers.5.ln1.weight as id 6 + +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.5.attn.qkv.bias as id 6set param backbone.layers.11.ffn.layers.1.bias as id 12set param backbone.layers.5.ln1.bias as id 6 + + +set param backbone.layers.5.attn.proj.weight as id 6set param backbone.layers.5.attn.relative_position_bias_table as id 6 + +set param backbone.layers.5.attn.proj.bias as id 6 +set param neck.upsample_4x.0.weight as id 13set param backbone.layers.5.attn.qkv.weight as id 6 + +set param backbone.layers.5.ln2.weight as id 6set param neck.upsample_4x.0.bias as id 13 + +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param neck.upsample_4x.1.weight as id 13 +set param backbone.layers.5.attn.proj.weight as id 6set param neck.upsample_4x.1.bias as id 13 + +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6set param neck.upsample_4x.3.weight as id 13 + +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param neck.upsample_4x.3.bias as id 13 +set param backbone.layers.5.ln2.weight as id 6 +set param neck.upsample_2x.0.weight as id 13set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 + +set param neck.upsample_2x.0.bias as id 13 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param decode_head.conv_seg.weight as id 13set param backbone.layers.5.ffn.layers.0.0.bias as id 6 + +set param backbone.layers.6.gamma_2 as id 7set param decode_head.conv_seg.bias as id 13 + +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.ln1.bias as id 7 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.gamma_1 as id 7 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13set param backbone.layers.6.gamma_2 as id 7 + +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.ln1.weight as id 7set param backbone.layers.6.attn.qkv.bias as id 7set param decode_head.psp_modules.1.1.conv.weight as id 13 + + +set param backbone.layers.6.ln1.bias as id 7set param decode_head.psp_modules.1.1.bn.weight as id 13 + +set param backbone.layers.6.attn.proj.weight as id 7 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param decode_head.psp_modules.2.1.bn.bias as id 13set param backbone.layers.6.ln2.bias as id 7 + +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7set param decode_head.psp_modules.3.1.conv.weight as id 13 + +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param backbone.layers.6.ln2.weight as id 7set param decode_head.psp_modules.3.1.bn.bias as id 13 + +set param backbone.layers.6.ln2.bias as id 7 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13set param backbone.layers.6.ffn.layers.0.0.weight as id 7 + +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param backbone.layers.6.ffn.layers.1.weight as id 7set param decode_head.fpn_convs.0.bn.bias as id 13 + +set param backbone.layers.6.ffn.layers.1.bias as id 7set param decode_head.fpn_convs.1.conv.weight as id 13 + +set param backbone.layers.6.ffn.layers.0.0.weight as id 7set param decode_head.fpn_convs.1.bn.weight as id 13 + +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7set param backbone.layers.7.gamma_1 as id 8 +set param decode_head.fpn_convs.2.conv.weight as id 13 + +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param backbone.layers.7.gamma_2 as id 8set param backbone.layers.6.ffn.layers.1.weight as id 7set param decode_head.fpn_convs.2.bn.bias as id 13 + + +set param backbone.layers.6.ffn.layers.1.bias as id 7set param backbone.layers.7.ln1.weight as id 8set param decode_head.fpn_bottleneck.conv.weight as id 13 + + +set param backbone.layers.7.ln1.bias as id 8set param decode_head.fpn_bottleneck.bn.weight as id 13 + +set param decode_head.fpn_bottleneck.bn.bias as id 13set param backbone.layers.7.gamma_1 as id 8 + +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8set param auxiliary_head.conv_seg.weight as id 13 + +set param backbone.layers.7.ln1.weight as id 8 +set param auxiliary_head.conv_seg.bias as id 13set param backbone.layers.7.attn.qkv.bias as id 8 + +set param backbone.layers.7.ln1.bias as id 8 +set param auxiliary_head.convs.0.conv.weight as id 13set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 + +set param auxiliary_head.convs.0.bn.weight as id 13 +set param backbone.layers.7.attn.proj.bias as id 8set param auxiliary_head.convs.0.bn.bias as id 13 + +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.ln2.weight as id 8set param backbone.layers.7.attn.qkv.bias as id 8 + +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8set param backbone.layers.7.ln2.weight as id 8 + +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9set param backbone.layers.7.ffn.layers.1.weight as id 8 + +set param backbone.layers.7.ffn.layers.1.bias as id 8set param backbone.layers.8.ln1.weight as id 9 + +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9set param backbone.layers.8.ln1.weight as id 9 + +set param backbone.layers.8.attn.qkv.bias as id 9set param backbone.layers.8.ln1.bias as id 9 + +set param backbone.layers.8.attn.proj.weight as id 9set param backbone.layers.8.attn.relative_position_bias_table as id 9 + +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9set param backbone.layers.8.ln2.weight as id 9 + +set param backbone.layers.8.ln2.bias as id 9set param backbone.layers.8.attn.proj.weight as id 9 + +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9set param backbone.layers.8.ffn.layers.0.0.weight as id 9 + +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10set param backbone.layers.9.gamma_2 as id 10 + +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10set param backbone.layers.9.attn.proj.bias as id 10 + +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10set param backbone.layers.9.ln2.weight as id 10 + +set param backbone.layers.9.ffn.layers.0.0.bias as id 10set param backbone.layers.9.ln2.bias as id 10 + +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10set param backbone.layers.10.gamma_1 as id 11 + +set param backbone.layers.9.ffn.layers.1.bias as id 10set param backbone.layers.10.gamma_2 as id 11 + +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11set param backbone.layers.10.ffn.layers.0.0.weight as id 11 + +set param backbone.layers.10.ln2.bias as id 11set param backbone.layers.10.ffn.layers.0.0.bias as id 11 + +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11set param backbone.layers.11.gamma_1 as id 12 + +set param backbone.layers.10.ffn.layers.1.bias as id 11set param backbone.layers.11.gamma_2 as id 12 + +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12set param backbone.layers.11.ln1.bias as id 12 + +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12set param backbone.layers.11.attn.proj.weight as id 12 + +set param backbone.layers.11.ln2.bias as id 12set param backbone.layers.11.attn.proj.bias as id 12 + +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12set param backbone.layers.11.ffn.layers.0.0.weight as id 12 + +set param backbone.layers.11.ffn.layers.1.bias as id 12set param backbone.layers.11.ffn.layers.0.0.bias as id 12 + +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13set param neck.upsample_4x.1.weight as id 13 + +set param neck.upsample_4x.3.bias as id 13set param neck.upsample_4x.1.bias as id 13 + +set param neck.upsample_4x.3.weight as id 13set param neck.upsample_2x.0.weight as id 13 + +set param neck.upsample_4x.3.bias as id 13set param neck.upsample_2x.0.bias as id 13 + +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13set param decode_head.psp_modules.1.1.bn.bias as id 13 + +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13set param decode_head.psp_modules.1.1.bn.bias as id 13 + +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13set param decode_head.psp_modules.2.1.conv.weight as id 13 + +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.bottleneck.conv.weight as id 13set param decode_head.psp_modules.3.1.bn.weight as id 13 + +set param decode_head.psp_modules.3.1.bn.bias as id 13set param decode_head.bottleneck.bn.weight as id 13 + +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13set param decode_head.bottleneck.bn.bias as id 13 + +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13set param decode_head.lateral_convs.0.conv.weight as id 13 + +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13set param decode_head.lateral_convs.1.conv.weight as id 13 + +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13set param decode_head.fpn_convs.0.conv.weight as id 13 + +set param decode_head.fpn_convs.0.bn.weight as id 13set param decode_head.fpn_convs.1.conv.weight as id 13 + +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13set param decode_head.fpn_convs.1.conv.weight as id 13 + +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13set param decode_head.fpn_convs.2.conv.weight as id 13 + +set param decode_head.fpn_convs.2.bn.weight as id 13set param decode_head.fpn_bottleneck.conv.weight as id 13 + +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +Param groups = { + "layer_0_decay": { + "param_names": [ + "backbone.cls_token", + "backbone.pos_embed", + "backbone.patch_embed.projection.weight" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.05 + }, + "layer_0_no_decay": { + "param_names": [ + "backbone.patch_embed.projection.bias" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.0 + }, + "layer_1_no_decay": { + "param_names": [ + "backbone.layers.0.gamma_1", + "backbone.layers.0.gamma_2", + "backbone.layers.0.ln1.weight", + "backbone.layers.0.ln1.bias", + "backbone.layers.0.attn.qkv.bias", + "backbone.layers.0.attn.proj.bias", + "backbone.layers.0.ln2.weight", + "backbone.layers.0.ln2.bias", + "backbone.layers.0.ffn.layers.0.0.bias", + "backbone.layers.0.ffn.layers.1.bias" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.0 + }, + "layer_1_decay": { + "param_names": [ + "backbone.layers.0.attn.relative_position_bias_table", + "backbone.layers.0.attn.qkv.weight", + "backbone.layers.0.attn.proj.weight", + "backbone.layers.0.ffn.layers.0.0.weight", + "backbone.layers.0.ffn.layers.1.weight" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.05 + }, + "layer_2_no_decay": { + "param_names": [ + "backbone.layers.1.gamma_1", + "backbone.layers.1.gamma_2", + "backbone.layers.1.ln1.weight", + "backbone.layers.1.ln1.bias", + "backbone.layers.1.attn.qkv.bias", + "backbone.layers.1.attn.proj.bias", + "backbone.layers.1.ln2.weight", + "backbone.layers.1.ln2.bias", + "backbone.layers.1.ffn.layers.0.0.bias", + "backbone.layers.1.ffn.layers.1.bias" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.0 + }, + "layer_2_decay": { + "param_names": [ + "backbone.layers.1.attn.relative_position_bias_table", + "backbone.layers.1.attn.qkv.weight", + "backbone.layers.1.attn.proj.weight", + "backbone.layers.1.ffn.layers.0.0.weight", + "backbone.layers.1.ffn.layers.1.weight" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.05 + }, + "layer_3_no_decay": { + "param_names": [ + "backbone.layers.2.gamma_1", + "backbone.layers.2.gamma_2", + "backbone.layers.2.ln1.weight", + "backbone.layers.2.ln1.bias", + "backbone.layers.2.attn.qkv.bias", + "backbone.layers.2.attn.proj.bias", + "backbone.layers.2.ln2.weight", + "backbone.layers.2.ln2.bias", + "backbone.layers.2.ffn.layers.0.0.bias", + "backbone.layers.2.ffn.layers.1.bias" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.0 + }, + "layer_3_decay": { + "param_names": [ + "backbone.layers.2.attn.relative_position_bias_table", + "backbone.layers.2.attn.qkv.weight", + "backbone.layers.2.attn.proj.weight", + "backbone.layers.2.ffn.layers.0.0.weight", + "backbone.layers.2.ffn.layers.1.weight" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.05 + }, + "layer_4_no_decay": { + "param_names": [ + "backbone.layers.3.gamma_1", + "backbone.layers.3.gamma_2", + "backbone.layers.3.ln1.weight", + "backbone.layers.3.ln1.bias", + "backbone.layers.3.attn.qkv.bias", + "backbone.layers.3.attn.proj.bias", + "backbone.layers.3.ln2.weight", + "backbone.layers.3.ln2.bias", + "backbone.layers.3.ffn.layers.0.0.bias", + "backbone.layers.3.ffn.layers.1.bias" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.0 + }, + "layer_4_decay": { + "param_names": [ + "backbone.layers.3.attn.relative_position_bias_table", + "backbone.layers.3.attn.qkv.weight", + "backbone.layers.3.attn.proj.weight", + "backbone.layers.3.ffn.layers.0.0.weight", + "backbone.layers.3.ffn.layers.1.weight" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.05 + }, + "layer_5_no_decay": { + "param_names": [ + "backbone.layers.4.gamma_1", + "backbone.layers.4.gamma_2", + "backbone.layers.4.ln1.weight", + "backbone.layers.4.ln1.bias", + "backbone.layers.4.attn.qkv.bias", + "backbone.layers.4.attn.proj.bias", + "backbone.layers.4.ln2.weight", + "backbone.layers.4.ln2.bias", + "backbone.layers.4.ffn.layers.0.0.bias", + "backbone.layers.4.ffn.layers.1.bias" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.0 + }, + "layer_5_decay": { + "param_names": [ + "backbone.layers.4.attn.relative_position_bias_table", + "backbone.layers.4.attn.qkv.weight", + "backbone.layers.4.attn.proj.weight", + "backbone.layers.4.ffn.layers.0.0.weight", + "backbone.layers.4.ffn.layers.1.weight" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.05 + }, + "layer_6_no_decay": { + "param_names": [ + "backbone.layers.5.gamma_1", + "backbone.layers.5.gamma_2", + "backbone.layers.5.ln1.weight", + "backbone.layers.5.ln1.bias", + "backbone.layers.5.attn.qkv.bias", + "backbone.layers.5.attn.proj.bias", + "backbone.layers.5.ln2.weight", + "backbone.layers.5.ln2.bias", + "backbone.layers.5.ffn.layers.0.0.bias", + "backbone.layers.5.ffn.layers.1.bias" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.0 + }, + "layer_6_decay": { + "param_names": [ + "backbone.layers.5.attn.relative_position_bias_table", + "backbone.layers.5.attn.qkv.weight", + "backbone.layers.5.attn.proj.weight", + "backbone.layers.5.ffn.layers.0.0.weight", + "backbone.layers.5.ffn.layers.1.weight" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.05 + }, + "layer_7_no_decay": { + "param_names": [ + "backbone.layers.6.gamma_1", + "backbone.layers.6.gamma_2", + "backbone.layers.6.ln1.weight", + "backbone.layers.6.ln1.bias", + "backbone.layers.6.attn.qkv.bias", + "backbone.layers.6.attn.proj.bias", + "backbone.layers.6.ln2.weight", + "backbone.layers.6.ln2.bias", + "backbone.layers.6.ffn.layers.0.0.bias", + "backbone.layers.6.ffn.layers.1.bias" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.0 + }, + "layer_7_decay": { + "param_names": [ + "backbone.layers.6.attn.relative_position_bias_table", + "backbone.layers.6.attn.qkv.weight", + "backbone.layers.6.attn.proj.weight", + "backbone.layers.6.ffn.layers.0.0.weight", + "backbone.layers.6.ffn.layers.1.weight" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.05 + }, + "layer_8_no_decay": { + "param_names": [ + "backbone.layers.7.gamma_1", + "backbone.layers.7.gamma_2", + "backbone.layers.7.ln1.weight", + "backbone.layers.7.ln1.bias", + "backbone.layers.7.attn.qkv.bias", + "backbone.layers.7.attn.proj.bias", + "backbone.layers.7.ln2.weight", + "backbone.layers.7.ln2.bias", + "backbone.layers.7.ffn.layers.0.0.bias", + "backbone.layers.7.ffn.layers.1.bias" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.0 + }, + "layer_8_decay": { + "param_names": [ + "backbone.layers.7.attn.relative_position_bias_table", + "backbone.layers.7.attn.qkv.weight", + "backbone.layers.7.attn.proj.weight", + "backbone.layers.7.ffn.layers.0.0.weight", + "backbone.layers.7.ffn.layers.1.weight" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.05 + }, + "layer_9_no_decay": { + "param_names": [ + "backbone.layers.8.gamma_1", + "backbone.layers.8.gamma_2", + "backbone.layers.8.ln1.weight", + "backbone.layers.8.ln1.bias", + "backbone.layers.8.attn.qkv.bias", + "backbone.layers.8.attn.proj.bias", + "backbone.layers.8.ln2.weight", + "backbone.layers.8.ln2.bias", + "backbone.layers.8.ffn.layers.0.0.bias", + "backbone.layers.8.ffn.layers.1.bias" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.0 + }, + "layer_9_decay": { + "param_names": [ + "backbone.layers.8.attn.relative_position_bias_table", + "backbone.layers.8.attn.qkv.weight", + "backbone.layers.8.attn.proj.weight", + "backbone.layers.8.ffn.layers.0.0.weight", + "backbone.layers.8.ffn.layers.1.weight" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.05 + }, + "layer_10_no_decay": { + "param_names": [ + "backbone.layers.9.gamma_1", + "backbone.layers.9.gamma_2", + "backbone.layers.9.ln1.weight", + "backbone.layers.9.ln1.bias", + "backbone.layers.9.attn.qkv.bias", + "backbone.layers.9.attn.proj.bias", + "backbone.layers.9.ln2.weight", + "backbone.layers.9.ln2.bias", + "backbone.layers.9.ffn.layers.0.0.bias", + "backbone.layers.9.ffn.layers.1.bias" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.0 + }, + "layer_10_decay": { + "param_names": [ + "backbone.layers.9.attn.relative_position_bias_table", + "backbone.layers.9.attn.qkv.weight", + "backbone.layers.9.attn.proj.weight", + "backbone.layers.9.ffn.layers.0.0.weight", + "backbone.layers.9.ffn.layers.1.weight" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.05 + }, + "layer_11_no_decay": { + "param_names": [ + "backbone.layers.10.gamma_1", + "backbone.layers.10.gamma_2", + "backbone.layers.10.ln1.weight", + "backbone.layers.10.ln1.bias", + "backbone.layers.10.attn.qkv.bias", + "backbone.layers.10.attn.proj.bias", + "backbone.layers.10.ln2.weight", + "backbone.layers.10.ln2.bias", + "backbone.layers.10.ffn.layers.0.0.bias", + "backbone.layers.10.ffn.layers.1.bias" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.0 + }, + "layer_11_decay": { + "param_names": [ + "backbone.layers.10.attn.relative_position_bias_table", + "backbone.layers.10.attn.qkv.weight", + "backbone.layers.10.attn.proj.weight", + "backbone.layers.10.ffn.layers.0.0.weight", + "backbone.layers.10.ffn.layers.1.weight" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.05 + }, + "layer_12_no_decay": { + "param_names": [ + "backbone.layers.11.gamma_1", + "backbone.layers.11.gamma_2", + "backbone.layers.11.ln1.weight", + "backbone.layers.11.ln1.bias", + "backbone.layers.11.attn.qkv.bias", + "backbone.layers.11.attn.proj.bias", + "backbone.layers.11.ln2.weight", + "backbone.layers.11.ln2.bias", + "backbone.layers.11.ffn.layers.0.0.bias", + "backbone.layers.11.ffn.layers.1.bias" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.0 + }, + "layer_12_decay": { + "param_names": [ + "backbone.layers.11.attn.relative_position_bias_table", + "backbone.layers.11.attn.qkv.weight", + "backbone.layers.11.attn.proj.weight", + "backbone.layers.11.ffn.layers.0.0.weight", + "backbone.layers.11.ffn.layers.1.weight" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.05 + }, + "layer_13_decay": { + "param_names": [ + "neck.upsample_4x.0.weight", + "neck.upsample_4x.3.weight", + "neck.upsample_2x.0.weight", + "decode_head.conv_seg.weight", + "decode_head.psp_modules.0.1.conv.weight", + "decode_head.psp_modules.1.1.conv.weight", + "decode_head.psp_modules.2.1.conv.weight", + "decode_head.psp_modules.3.1.conv.weight", + "decode_head.bottleneck.conv.weight", + "decode_head.lateral_convs.0.conv.weight", + "decode_head.lateral_convs.1.conv.weight", + "decode_head.lateral_convs.2.conv.weight", + "decode_head.fpn_convs.0.conv.weight", + "decode_head.fpn_convs.1.conv.weight", + "decode_head.fpn_convs.2.conv.weight", + "decode_head.fpn_bottleneck.conv.weight", + "auxiliary_head.conv_seg.weight", + "auxiliary_head.convs.0.conv.weight" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.05 + }, + "layer_13_no_decay": { + "param_names": [ + "neck.upsample_4x.0.bias", + "neck.upsample_4x.1.weight", + "neck.upsample_4x.1.bias", + "neck.upsample_4x.3.bias", + "neck.upsample_2x.0.bias", + "decode_head.conv_seg.bias", + "decode_head.psp_modules.0.1.bn.weight", + "decode_head.psp_modules.0.1.bn.bias", + "decode_head.psp_modules.1.1.bn.weight", + "decode_head.psp_modules.1.1.bn.bias", + "decode_head.psp_modules.2.1.bn.weight", + "decode_head.psp_modules.2.1.bn.bias", + "decode_head.psp_modules.3.1.bn.weight", + "decode_head.psp_modules.3.1.bn.bias", + "decode_head.bottleneck.bn.weight", + "decode_head.bottleneck.bn.bias", + "decode_head.lateral_convs.0.bn.weight", + "decode_head.lateral_convs.0.bn.bias", + "decode_head.lateral_convs.1.bn.weight", + "decode_head.lateral_convs.1.bn.bias", + "decode_head.lateral_convs.2.bn.weight", + "decode_head.lateral_convs.2.bn.bias", + "decode_head.fpn_convs.0.bn.weight", + "decode_head.fpn_convs.0.bn.bias", + "decode_head.fpn_convs.1.bn.weight", + "decode_head.fpn_convs.1.bn.bias", + "decode_head.fpn_convs.2.bn.weight", + "decode_head.fpn_convs.2.bn.bias", + "decode_head.fpn_bottleneck.bn.weight", + "decode_head.fpn_bottleneck.bn.bias", + "auxiliary_head.conv_seg.bias", + "auxiliary_head.convs.0.bn.weight", + "auxiliary_head.convs.0.bn.bias" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.0 + } +} +04/19 00:45:39 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1765, in train + self.load_or_resume() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1699, in load_or_resume + self.load_checkpoint(self._load_from) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 2127, in load_checkpoint + checkpoint = _load_checkpoint(filename, map_location=map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError: can not be found. +Loads checkpoint by local backend from path: +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1765, in train + self.load_or_resume() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1699, in load_or_resume + self.load_checkpoint(self._load_from) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 2127, in load_checkpoint + checkpoint = _load_checkpoint(filename, map_location=map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError: can not be found. +Loads checkpoint by local backend from path: +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1765, in train + self.load_or_resume() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1699, in load_or_resume + self.load_checkpoint(self._load_from) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 2127, in load_checkpoint + checkpoint = _load_checkpoint(filename, map_location=map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError: can not be found. +Loads checkpoint by local backend from path: +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1765, in train + self.load_or_resume() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1699, in load_or_resume + self.load_checkpoint(self._load_from) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 2127, in load_checkpoint + checkpoint = _load_checkpoint(filename, map_location=map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError: can not be found. +ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 1) local_rank: 0 (pid: 61752) of binary: /opt/conda/bin/python +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 245, in launch_agent + raise ChildFailedError( +torch.distributed.elastic.multiprocessing.errors.ChildFailedError: +============================================================ +tools/train.py FAILED +------------------------------------------------------------ +Failures: +[1]: + time : 2024-04-19_00:45:41 + host : 06540a9609a4 + rank : 1 (local_rank: 1) + exitcode : 1 (pid: 61753) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[2]: + time : 2024-04-19_00:45:41 + host : 06540a9609a4 + rank : 2 (local_rank: 2) + exitcode : 1 (pid: 61754) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[3]: + time : 2024-04-19_00:45:41 + host : 06540a9609a4 + rank : 3 (local_rank: 3) + exitcode : 1 (pid: 61755) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +------------------------------------------------------------ +Root Cause (first observed failure): +[0]: + time : 2024-04-19_00:45:41 + host : 06540a9609a4 + rank : 0 (local_rank: 0) + exitcode : 1 (pid: 61752) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +============================================================ +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +usage: train.py [-h] [--work-dir WORK_DIR] [--resume] [--amp] + [--cfg-options CFG_OPTIONS [CFG_OPTIONS ...]] + [--launcher {none,pytorch,slurm,mpi}] + [--local_rank LOCAL_RANK] + config +usage: train.py [-h] [--work-dir WORK_DIR] [--resume] [--amp] + [--cfg-options CFG_OPTIONS [CFG_OPTIONS ...]] + [--launcher {none,pytorch,slurm,mpi}] + [--local_rank LOCAL_RANK] + config +train.py: error: unrecognized arguments: auto +train.py: error: unrecognized arguments: auto +usage: train.py [-h] [--work-dir WORK_DIR] [--resume] [--amp] + [--cfg-options CFG_OPTIONS [CFG_OPTIONS ...]] + [--launcher {none,pytorch,slurm,mpi}] + [--local_rank LOCAL_RANK] + config +train.py: error: unrecognized arguments: auto +usage: train.py [-h] [--work-dir WORK_DIR] [--resume] [--amp] + [--cfg-options CFG_OPTIONS [CFG_OPTIONS ...]] + [--launcher {none,pytorch,slurm,mpi}] + [--local_rank LOCAL_RANK] + config +train.py: error: unrecognized arguments: auto +ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 2) local_rank: 0 (pid: 62754) of binary: /opt/conda/bin/python +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 245, in launch_agent + raise ChildFailedError( +torch.distributed.elastic.multiprocessing.errors.ChildFailedError: +============================================================ +tools/train.py FAILED +------------------------------------------------------------ +Failures: +[1]: + time : 2024-04-19_00:46:07 + host : 06540a9609a4 + rank : 1 (local_rank: 1) + exitcode : 2 (pid: 62755) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[2]: + time : 2024-04-19_00:46:07 + host : 06540a9609a4 + rank : 2 (local_rank: 2) + exitcode : 2 (pid: 62756) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[3]: + time : 2024-04-19_00:46:07 + host : 06540a9609a4 + rank : 3 (local_rank: 3) + exitcode : 2 (pid: 62757) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +------------------------------------------------------------ +Root Cause (first observed failure): +[0]: + time : 2024-04-19_00:46:07 + host : 06540a9609a4 + rank : 0 (local_rank: 0) + exitcode : 2 (pid: 62754) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +============================================================ +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/19 00:49:00 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 1281961478 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 1281961478 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/19 00:49:00 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +fp16 = dict(loss_scale='dynamic') +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = None +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=768, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + num_convs=1, + type='FCNHead'), + backbone=dict( + act_cfg=dict(type='GELU'), + attn_drop_rate=0.0, + drop_path_rate=0.1, + embed_dims=768, + img_size=( + 512, + 512, + ), + in_channels=3, + init_values=1.0, + mlp_ratio=4, + norm_cfg=dict(eps=1e-06, type='LN'), + norm_eval=False, + num_heads=12, + num_layers=12, + out_indices=[ + 3, + 5, + 7, + 11, + ], + patch_size=16, + type='MAE'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=768, + dropout_ratio=0.1, + in_channels=[ + 768, + 768, + 768, + 768, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + neck=dict( + embed_dim=768, rescales=[ + 4, + 2, + 1, + 0.5, + ], type='Feature2Pyramid'), + pretrained='/workspaces/mmsegmentation-1/converted_model.pth', + test_cfg=dict(crop_size=( + 512, + 512, + ), mode='slide', stride=( + 341, + 341, + )), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + constructor='LayerDecayOptimizerConstructor', + optimizer=dict( + betas=( + 0.9, + 0.999, + ), lr=0.0001, type='AdamW', weight_decay=0.05), + paramwise_cfg=dict(layer_decay_rate=0.65, num_layers=12), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, by_epoch=False, end=1500, start_factor=1e-06, + type='LinearLR'), + dict( + begin=1500, + by_epoch=False, + end=160000, + eta_min=0.0, + power=1.0, + type='PolyLR'), +] +resume = True +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=2, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = '/workspaces/mmsegmentation-1/work_dirs/MAE4000' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/19 00:49:03 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +set param backbone.layers.0.gamma_1 as id 1 +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.cls_token as id 0 +set param backbone.layers.0.attn.qkv.weight as id 1set param backbone.pos_embed as id 0 + +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.patch_embed.projection.weight as id 0set param backbone.layers.0.attn.proj.weight as id 1 + +set param backbone.patch_embed.projection.bias as id 0set param backbone.layers.0.attn.proj.bias as id 1 + +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.gamma_1 as id 1set param backbone.layers.0.ln2.bias as id 1 + +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1set param backbone.layers.0.attn.qkv.bias as id 1 + +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1set param backbone.layers.1.gamma_1 as id 2 + +set param backbone.layers.1.gamma_2 as id 2set param backbone.layers.0.ln2.weight as id 1 + +set param backbone.layers.0.ln2.bias as id 1set param backbone.layers.1.ln1.weight as id 2 + +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2set param backbone.layers.0.ffn.layers.0.0.weight as id 1 + +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2set param backbone.layers.0.ffn.layers.1.weight as id 1 + +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2set param backbone.layers.1.attn.relative_position_bias_table as id 2 + +set param backbone.layers.1.ffn.layers.0.0.bias as id 2set param backbone.layers.1.attn.qkv.weight as id 2 + +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.attn.proj.weight as id 2set param backbone.layers.1.ffn.layers.1.bias as id 2 + +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2set param backbone.layers.2.gamma_1 as id 3 + +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3set param backbone.layers.1.ffn.layers.0.0.weight as id 2 + +set param backbone.layers.2.ln1.bias as id 3set param backbone.layers.1.ffn.layers.0.0.bias as id 2 + +set param backbone.layers.2.attn.relative_position_bias_table as id 3set param backbone.layers.1.ffn.layers.1.weight as id 2 + +set param backbone.layers.1.ffn.layers.1.bias as id 2set param backbone.layers.2.attn.qkv.weight as id 3 + +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3set param backbone.layers.2.attn.proj.weight as id 3 + +set param backbone.layers.2.ln1.weight as id 3set param backbone.layers.2.attn.proj.bias as id 3 + +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3set param backbone.layers.2.attn.relative_position_bias_table as id 3 + +set param backbone.layers.2.ln2.bias as id 3set param backbone.layers.2.attn.qkv.weight as id 3 + +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3set param backbone.layers.2.attn.proj.weight as id 3 + +set param backbone.layers.2.attn.proj.bias as id 3set param backbone.layers.2.ffn.layers.0.0.bias as id 3 + +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3set param backbone.layers.3.gamma_1 as id 4 + +set param backbone.layers.3.gamma_2 as id 4set param backbone.layers.2.ffn.layers.1.weight as id 3 + +set param backbone.layers.2.ffn.layers.1.bias as id 3set param backbone.layers.3.ln1.weight as id 4 + +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4set param backbone.layers.3.gamma_2 as id 4 + +set param backbone.layers.3.ln1.weight as id 4set param backbone.layers.3.attn.qkv.weight as id 4 + +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4set param backbone.layers.3.attn.proj.weight as id 4 + +set param backbone.layers.3.attn.qkv.bias as id 4set param backbone.layers.3.attn.proj.bias as id 4 + +set param backbone.layers.3.attn.proj.weight as id 4set param backbone.layers.3.ln2.weight as id 4 + +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4set param backbone.layers.3.ffn.layers.1.weight as id 4 + +set param backbone.layers.3.ffn.layers.1.bias as id 4set param backbone.layers.3.ffn.layers.1.bias as id 4 + +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.gamma_2 as id 5set param backbone.layers.4.ln1.weight as id 5 + +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5set param backbone.layers.4.ln1.bias as id 5 + +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.weight as id 5set param backbone.layers.4.attn.proj.bias as id 5 + +set param backbone.layers.4.attn.proj.bias as id 5set param backbone.layers.4.ln2.weight as id 5 + +set param backbone.layers.4.ln2.bias as id 5set param backbone.layers.4.ln2.weight as id 5 + +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5set param backbone.layers.4.ffn.layers.0.0.bias as id 5 + +set param backbone.layers.5.gamma_1 as id 6set param backbone.layers.4.ffn.layers.1.weight as id 5 + +set param backbone.layers.5.gamma_2 as id 6set param backbone.layers.4.ffn.layers.1.bias as id 5 + +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6set param backbone.layers.5.gamma_1 as id 6 + +set param backbone.layers.5.attn.relative_position_bias_table as id 6set param backbone.layers.5.gamma_2 as id 6 + +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6set param backbone.layers.5.attn.proj.bias as id 6 + +set param backbone.layers.5.ln2.weight as id 6set param backbone.layers.5.attn.qkv.weight as id 6 + +set param backbone.layers.5.ln2.bias as id 6set param backbone.layers.5.attn.qkv.bias as id 6 + +set param backbone.layers.5.ffn.layers.0.0.weight as id 6set param backbone.layers.5.attn.proj.weight as id 6 + +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ln2.weight as id 6set param backbone.layers.5.ffn.layers.1.bias as id 6 + +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6set param backbone.layers.6.ln1.bias as id 7 + +set param backbone.layers.6.attn.relative_position_bias_table as id 7set param backbone.layers.5.ffn.layers.1.weight as id 6 + +set param backbone.layers.6.attn.qkv.weight as id 7set param backbone.layers.5.ffn.layers.1.bias as id 6 + +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7set param backbone.layers.6.gamma_1 as id 7 + +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7set param backbone.layers.6.ln1.weight as id 7 + +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8set param backbone.layers.6.ffn.layers.0.0.weight as id 7 + +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8set param backbone.layers.6.ffn.layers.1.weight as id 7 + +set param backbone.layers.6.ffn.layers.1.bias as id 7set param backbone.layers.7.ffn.layers.0.0.weight as id 8 + +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8set param backbone.layers.7.gamma_1 as id 8 + +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9set param backbone.layers.7.attn.relative_position_bias_table as id 8 + +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.7.attn.qkv.weight as id 8set param backbone.layers.8.attn.relative_position_bias_table as id 9 + +set param backbone.layers.7.attn.qkv.bias as id 8set param backbone.layers.8.attn.qkv.weight as id 9 + +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9set param backbone.layers.7.ln2.weight as id 8 + +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8set param backbone.layers.8.ffn.layers.1.weight as id 9 + +set param backbone.layers.8.ffn.layers.1.bias as id 9set param backbone.layers.7.ffn.layers.0.0.bias as id 8 + +set param backbone.layers.7.ffn.layers.1.weight as id 8set param backbone.layers.9.gamma_1 as id 10 + +set param backbone.layers.7.ffn.layers.1.bias as id 8set param backbone.layers.9.gamma_2 as id 10 + +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10set param backbone.layers.8.gamma_1 as id 9 + +set param backbone.layers.9.attn.relative_position_bias_table as id 10set param backbone.layers.8.gamma_2 as id 9 + +set param backbone.layers.9.attn.qkv.weight as id 10set param backbone.layers.8.ln1.weight as id 9 + +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10set param backbone.layers.8.attn.relative_position_bias_table as id 9 + +set param backbone.layers.9.ln2.weight as id 10set param backbone.layers.8.attn.qkv.weight as id 9 + +set param backbone.layers.9.ln2.bias as id 10set param backbone.layers.8.attn.qkv.bias as id 9 + +set param backbone.layers.9.ffn.layers.0.0.weight as id 10set param backbone.layers.8.attn.proj.weight as id 9 + +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.8.ln2.weight as id 9set param backbone.layers.9.ffn.layers.1.bias as id 10 + +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11set param backbone.layers.8.ffn.layers.1.weight as id 9 + +set param backbone.layers.8.ffn.layers.1.bias as id 9set param backbone.layers.10.attn.qkv.weight as id 11 + +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.9.gamma_1 as id 10set param backbone.layers.10.attn.proj.weight as id 11 + +set param backbone.layers.10.attn.proj.bias as id 11set param backbone.layers.9.gamma_2 as id 10 + +set param backbone.layers.10.ln2.weight as id 11set param backbone.layers.9.ln1.weight as id 10 + +set param backbone.layers.10.ln2.bias as id 11set param backbone.layers.9.ln1.bias as id 10 + +set param backbone.layers.9.attn.relative_position_bias_table as id 10set param backbone.layers.10.ffn.layers.0.0.weight as id 11 + +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.9.attn.qkv.bias as id 10set param backbone.layers.10.ffn.layers.1.bias as id 11 + +set param backbone.layers.9.attn.proj.weight as id 10set param backbone.layers.11.gamma_1 as id 12 + +set param backbone.layers.9.attn.proj.bias as id 10set param backbone.layers.11.gamma_2 as id 12 + +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.9.ln2.bias as id 10set param backbone.layers.11.attn.relative_position_bias_table as id 12 + +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.11.ln2.weight as id 12set param backbone.layers.9.ffn.layers.1.bias as id 10 + +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.10.gamma_1 as id 11set param backbone.layers.11.ffn.layers.0.0.weight as id 12 + +set param backbone.layers.11.ffn.layers.0.0.bias as id 12set param backbone.layers.10.gamma_2 as id 11 + +set param backbone.layers.11.ffn.layers.1.weight as id 12set param backbone.layers.10.ln1.weight as id 11 + +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param neck.upsample_4x.0.weight as id 13set param backbone.layers.10.attn.qkv.weight as id 11 + +set param neck.upsample_4x.0.bias as id 13 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param neck.upsample_4x.1.weight as id 13set param backbone.layers.10.attn.proj.weight as id 11 + +set param neck.upsample_4x.1.bias as id 13 +set param backbone.layers.10.attn.proj.bias as id 11 +set param neck.upsample_4x.3.weight as id 13 +set param backbone.layers.10.ln2.weight as id 11 +set param neck.upsample_4x.3.bias as id 13 +set param backbone.layers.10.ln2.bias as id 11 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13set param backbone.layers.10.ffn.layers.0.0.weight as id 11 + +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param decode_head.conv_seg.weight as id 13 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param decode_head.conv_seg.bias as id 13 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12set param decode_head.psp_modules.0.1.conv.weight as id 13 + +set param backbone.layers.11.ln1.weight as id 12set param decode_head.psp_modules.0.1.bn.weight as id 13 + +set param backbone.layers.11.ln1.bias as id 12set param decode_head.psp_modules.0.1.bn.bias as id 13 + +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12set param decode_head.psp_modules.1.1.conv.weight as id 13 + +set param backbone.layers.11.attn.qkv.bias as id 12set param decode_head.psp_modules.1.1.bn.weight as id 13 + +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param backbone.layers.11.attn.proj.weight as id 12 +set param decode_head.psp_modules.2.1.conv.weight as id 13set param backbone.layers.11.attn.proj.bias as id 12 + +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13set param backbone.layers.11.ln2.weight as id 12 + +set param backbone.layers.11.ln2.bias as id 12 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param decode_head.bottleneck.conv.weight as id 13set param backbone.layers.11.ffn.layers.0.0.bias as id 12 + +set param decode_head.bottleneck.bn.weight as id 13 +set param backbone.layers.11.ffn.layers.1.weight as id 12set param decode_head.bottleneck.bn.bias as id 13 + +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param neck.upsample_4x.0.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param neck.upsample_4x.0.bias as id 13set param decode_head.lateral_convs.1.bn.bias as id 13 + +set param decode_head.lateral_convs.2.conv.weight as id 13set param neck.upsample_4x.1.weight as id 13 + +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param neck.upsample_4x.3.bias as id 13set param decode_head.fpn_convs.0.bn.weight as id 13 + +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13set param auxiliary_head.conv_seg.bias as id 13 + +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +Param groups = { + "layer_0_decay": { + "param_names": [ + "backbone.cls_token", + "backbone.pos_embed", + "backbone.patch_embed.projection.weight" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.05 + }, + "layer_0_no_decay": { + "param_names": [ + "backbone.patch_embed.projection.bias" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.0 + }, + "layer_1_no_decay": { + "param_names": [ + "backbone.layers.0.gamma_1", + "backbone.layers.0.gamma_2", + "backbone.layers.0.ln1.weight", + "backbone.layers.0.ln1.bias", + "backbone.layers.0.attn.qkv.bias", + "backbone.layers.0.attn.proj.bias", + "backbone.layers.0.ln2.weight", + "backbone.layers.0.ln2.bias", + "backbone.layers.0.ffn.layers.0.0.bias", + "backbone.layers.0.ffn.layers.1.bias" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.0 + }, + "layer_1_decay": { + "param_names": [ + "backbone.layers.0.attn.relative_position_bias_table", + "backbone.layers.0.attn.qkv.weight", + "backbone.layers.0.attn.proj.weight", + "backbone.layers.0.ffn.layers.0.0.weight", + "backbone.layers.0.ffn.layers.1.weight" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.05 + }, + "layer_2_no_decay": { + "param_names": [ + "backbone.layers.1.gamma_1", + "backbone.layers.1.gamma_2", + "backbone.layers.1.ln1.weight", + "backbone.layers.1.ln1.bias", + "backbone.layers.1.attn.qkv.bias", + "backbone.layers.1.attn.proj.bias", + "backbone.layers.1.ln2.weight", + "backbone.layers.1.ln2.bias", + "backbone.layers.1.ffn.layers.0.0.bias", + "backbone.layers.1.ffn.layers.1.bias" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.0 + }, + "layer_2_decay": { + "param_names": [ + "backbone.layers.1.attn.relative_position_bias_table", + "backbone.layers.1.attn.qkv.weight", + "backbone.layers.1.attn.proj.weight", + "backbone.layers.1.ffn.layers.0.0.weight", + "backbone.layers.1.ffn.layers.1.weight" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.05 + }, + "layer_3_no_decay": { + "param_names": [ + "backbone.layers.2.gamma_1", + "backbone.layers.2.gamma_2", + "backbone.layers.2.ln1.weight", + "backbone.layers.2.ln1.bias", + "backbone.layers.2.attn.qkv.bias", + "backbone.layers.2.attn.proj.bias", + "backbone.layers.2.ln2.weight", + "backbone.layers.2.ln2.bias", + "backbone.layers.2.ffn.layers.0.0.bias", + "backbone.layers.2.ffn.layers.1.bias" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.0 + }, + "layer_3_decay": { + "param_names": [ + "backbone.layers.2.attn.relative_position_bias_table", + "backbone.layers.2.attn.qkv.weight", + "backbone.layers.2.attn.proj.weight", + "backbone.layers.2.ffn.layers.0.0.weight", + "backbone.layers.2.ffn.layers.1.weight" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.05 + }, + "layer_4_no_decay": { + "param_names": [ + "backbone.layers.3.gamma_1", + "backbone.layers.3.gamma_2", + "backbone.layers.3.ln1.weight", + "backbone.layers.3.ln1.bias", + "backbone.layers.3.attn.qkv.bias", + "backbone.layers.3.attn.proj.bias", + "backbone.layers.3.ln2.weight", + "backbone.layers.3.ln2.bias", + "backbone.layers.3.ffn.layers.0.0.bias", + "backbone.layers.3.ffn.layers.1.bias" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.0 + }, + "layer_4_decay": { + "param_names": [ + "backbone.layers.3.attn.relative_position_bias_table", + "backbone.layers.3.attn.qkv.weight", + "backbone.layers.3.attn.proj.weight", + "backbone.layers.3.ffn.layers.0.0.weight", + "backbone.layers.3.ffn.layers.1.weight" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.05 + }, + "layer_5_no_decay": { + "param_names": [ + "backbone.layers.4.gamma_1", + "backbone.layers.4.gamma_2", + "backbone.layers.4.ln1.weight", + "backbone.layers.4.ln1.bias", + "backbone.layers.4.attn.qkv.bias", + "backbone.layers.4.attn.proj.bias", + "backbone.layers.4.ln2.weight", + "backbone.layers.4.ln2.bias", + "backbone.layers.4.ffn.layers.0.0.bias", + "backbone.layers.4.ffn.layers.1.bias" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.0 + }, + "layer_5_decay": { + "param_names": [ + "backbone.layers.4.attn.relative_position_bias_table", + "backbone.layers.4.attn.qkv.weight", + "backbone.layers.4.attn.proj.weight", + "backbone.layers.4.ffn.layers.0.0.weight", + "backbone.layers.4.ffn.layers.1.weight" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.05 + }, + "layer_6_no_decay": { + "param_names": [ + "backbone.layers.5.gamma_1", + "backbone.layers.5.gamma_2", + "backbone.layers.5.ln1.weight", + "backbone.layers.5.ln1.bias", + "backbone.layers.5.attn.qkv.bias", + "backbone.layers.5.attn.proj.bias", + "backbone.layers.5.ln2.weight", + "backbone.layers.5.ln2.bias", + "backbone.layers.5.ffn.layers.0.0.bias", + "backbone.layers.5.ffn.layers.1.bias" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.0 + }, + "layer_6_decay": { + "param_names": [ + "backbone.layers.5.attn.relative_position_bias_table", + "backbone.layers.5.attn.qkv.weight", + "backbone.layers.5.attn.proj.weight", + "backbone.layers.5.ffn.layers.0.0.weight", + "backbone.layers.5.ffn.layers.1.weight" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.05 + }, + "layer_7_no_decay": { + "param_names": [ + "backbone.layers.6.gamma_1", + "backbone.layers.6.gamma_2", + "backbone.layers.6.ln1.weight", + "backbone.layers.6.ln1.bias", + "backbone.layers.6.attn.qkv.bias", + "backbone.layers.6.attn.proj.bias", + "backbone.layers.6.ln2.weight", + "backbone.layers.6.ln2.bias", + "backbone.layers.6.ffn.layers.0.0.bias", + "backbone.layers.6.ffn.layers.1.bias" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.0 + }, + "layer_7_decay": { + "param_names": [ + "backbone.layers.6.attn.relative_position_bias_table", + "backbone.layers.6.attn.qkv.weight", + "backbone.layers.6.attn.proj.weight", + "backbone.layers.6.ffn.layers.0.0.weight", + "backbone.layers.6.ffn.layers.1.weight" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.05 + }, + "layer_8_no_decay": { + "param_names": [ + "backbone.layers.7.gamma_1", + "backbone.layers.7.gamma_2", + "backbone.layers.7.ln1.weight", + "backbone.layers.7.ln1.bias", + "backbone.layers.7.attn.qkv.bias", + "backbone.layers.7.attn.proj.bias", + "backbone.layers.7.ln2.weight", + "backbone.layers.7.ln2.bias", + "backbone.layers.7.ffn.layers.0.0.bias", + "backbone.layers.7.ffn.layers.1.bias" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.0 + }, + "layer_8_decay": { + "param_names": [ + "backbone.layers.7.attn.relative_position_bias_table", + "backbone.layers.7.attn.qkv.weight", + "backbone.layers.7.attn.proj.weight", + "backbone.layers.7.ffn.layers.0.0.weight", + "backbone.layers.7.ffn.layers.1.weight" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.05 + }, + "layer_9_no_decay": { + "param_names": [ + "backbone.layers.8.gamma_1", + "backbone.layers.8.gamma_2", + "backbone.layers.8.ln1.weight", + "backbone.layers.8.ln1.bias", + "backbone.layers.8.attn.qkv.bias", + "backbone.layers.8.attn.proj.bias", + "backbone.layers.8.ln2.weight", + "backbone.layers.8.ln2.bias", + "backbone.layers.8.ffn.layers.0.0.bias", + "backbone.layers.8.ffn.layers.1.bias" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.0 + }, + "layer_9_decay": { + "param_names": [ + "backbone.layers.8.attn.relative_position_bias_table", + "backbone.layers.8.attn.qkv.weight", + "backbone.layers.8.attn.proj.weight", + "backbone.layers.8.ffn.layers.0.0.weight", + "backbone.layers.8.ffn.layers.1.weight" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.05 + }, + "layer_10_no_decay": { + "param_names": [ + "backbone.layers.9.gamma_1", + "backbone.layers.9.gamma_2", + "backbone.layers.9.ln1.weight", + "backbone.layers.9.ln1.bias", + "backbone.layers.9.attn.qkv.bias", + "backbone.layers.9.attn.proj.bias", + "backbone.layers.9.ln2.weight", + "backbone.layers.9.ln2.bias", + "backbone.layers.9.ffn.layers.0.0.bias", + "backbone.layers.9.ffn.layers.1.bias" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.0 + }, + "layer_10_decay": { + "param_names": [ + "backbone.layers.9.attn.relative_position_bias_table", + "backbone.layers.9.attn.qkv.weight", + "backbone.layers.9.attn.proj.weight", + "backbone.layers.9.ffn.layers.0.0.weight", + "backbone.layers.9.ffn.layers.1.weight" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.05 + }, + "layer_11_no_decay": { + "param_names": [ + "backbone.layers.10.gamma_1", + "backbone.layers.10.gamma_2", + "backbone.layers.10.ln1.weight", + "backbone.layers.10.ln1.bias", + "backbone.layers.10.attn.qkv.bias", + "backbone.layers.10.attn.proj.bias", + "backbone.layers.10.ln2.weight", + "backbone.layers.10.ln2.bias", + "backbone.layers.10.ffn.layers.0.0.bias", + "backbone.layers.10.ffn.layers.1.bias" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.0 + }, + "layer_11_decay": { + "param_names": [ + "backbone.layers.10.attn.relative_position_bias_table", + "backbone.layers.10.attn.qkv.weight", + "backbone.layers.10.attn.proj.weight", + "backbone.layers.10.ffn.layers.0.0.weight", + "backbone.layers.10.ffn.layers.1.weight" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.05 + }, + "layer_12_no_decay": { + "param_names": [ + "backbone.layers.11.gamma_1", + "backbone.layers.11.gamma_2", + "backbone.layers.11.ln1.weight", + "backbone.layers.11.ln1.bias", + "backbone.layers.11.attn.qkv.bias", + "backbone.layers.11.attn.proj.bias", + "backbone.layers.11.ln2.weight", + "backbone.layers.11.ln2.bias", + "backbone.layers.11.ffn.layers.0.0.bias", + "backbone.layers.11.ffn.layers.1.bias" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.0 + }, + "layer_12_decay": { + "param_names": [ + "backbone.layers.11.attn.relative_position_bias_table", + "backbone.layers.11.attn.qkv.weight", + "backbone.layers.11.attn.proj.weight", + "backbone.layers.11.ffn.layers.0.0.weight", + "backbone.layers.11.ffn.layers.1.weight" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.05 + }, + "layer_13_decay": { + "param_names": [ + "neck.upsample_4x.0.weight", + "neck.upsample_4x.3.weight", + "neck.upsample_2x.0.weight", + "decode_head.conv_seg.weight", + "decode_head.psp_modules.0.1.conv.weight", + "decode_head.psp_modules.1.1.conv.weight", + "decode_head.psp_modules.2.1.conv.weight", + "decode_head.psp_modules.3.1.conv.weight", + "decode_head.bottleneck.conv.weight", + "decode_head.lateral_convs.0.conv.weight", + "decode_head.lateral_convs.1.conv.weight", + "decode_head.lateral_convs.2.conv.weight", + "decode_head.fpn_convs.0.conv.weight", + "decode_head.fpn_convs.1.conv.weight", + "decode_head.fpn_convs.2.conv.weight", + "decode_head.fpn_bottleneck.conv.weight", + "auxiliary_head.conv_seg.weight", + "auxiliary_head.convs.0.conv.weight" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.05 + }, + "layer_13_no_decay": { + "param_names": [ + "neck.upsample_4x.0.bias", + "neck.upsample_4x.1.weight", + "neck.upsample_4x.1.bias", + "neck.upsample_4x.3.bias", + "neck.upsample_2x.0.bias", + "decode_head.conv_seg.bias", + "decode_head.psp_modules.0.1.bn.weight", + "decode_head.psp_modules.0.1.bn.bias", + "decode_head.psp_modules.1.1.bn.weight", + "decode_head.psp_modules.1.1.bn.bias", + "decode_head.psp_modules.2.1.bn.weight", + "decode_head.psp_modules.2.1.bn.bias", + "decode_head.psp_modules.3.1.bn.weight", + "decode_head.psp_modules.3.1.bn.bias", + "decode_head.bottleneck.bn.weight", + "decode_head.bottleneck.bn.bias", + "decode_head.lateral_convs.0.bn.weight", + "decode_head.lateral_convs.0.bn.bias", + "decode_head.lateral_convs.1.bn.weight", + "decode_head.lateral_convs.1.bn.bias", + "decode_head.lateral_convs.2.bn.weight", + "decode_head.lateral_convs.2.bn.bias", + "decode_head.fpn_convs.0.bn.weight", + "decode_head.fpn_convs.0.bn.bias", + "decode_head.fpn_convs.1.bn.weight", + "decode_head.fpn_convs.1.bn.bias", + "decode_head.fpn_convs.2.bn.weight", + "decode_head.fpn_convs.2.bn.bias", + "decode_head.fpn_bottleneck.bn.weight", + "decode_head.fpn_bottleneck.bn.bias", + "auxiliary_head.conv_seg.bias", + "auxiliary_head.convs.0.bn.weight", + "auxiliary_head.convs.0.bn.bias" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.0 + } +} +04/19 00:49:04 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/work_dirs/MAE4000/iter_60000.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/work_dirs/MAE4000/iter_60000.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/work_dirs/MAE4000/iter_60000.pth +04/19 00:49:04 - mmengine - INFO - Auto resumed from the latest checkpoint /workspaces/mmsegmentation-1/work_dirs/MAE4000/iter_60000.pth. +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/work_dirs/MAE4000/iter_60000.pth +04/19 00:49:06 - mmengine - INFO - Load checkpoint from /workspaces/mmsegmentation-1/work_dirs/MAE4000/iter_60000.pth +04/19 00:49:06 - mmengine - INFO - resumed epoch: 0, iter: 60000 +04/19 00:49:06 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +04/19 00:49:06 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +04/19 00:49:06 - mmengine - INFO - Checkpoints will be saved to /workspaces/mmsegmentation-1/work_dirs/MAE4000. +04/19 00:49:06 - mmengine - WARNING - Advance dataloader 60000 steps to skip data that has already been trained +04/19 01:03:46 - mmengine - INFO - Iter(train) [ 60050/160000] base_lr: 6.3060e-05 lr: 2.3315e-07 eta: 20 days, 9:05:08 time: 0.9809 data_time: 0.0044 memory: 8930 loss: 0.0164 decode.loss_ce: 0.0071 decode.acc_seg: 99.5728 aux.loss_ce: 0.0092 aux.acc_seg: 98.9300 +04/19 01:04:36 - mmengine - INFO - Iter(train) [ 60100/160000] base_lr: 6.3029e-05 lr: 2.3303e-07 eta: 10 days, 18:05:59 time: 0.9869 data_time: 0.0046 memory: 8457 loss: 0.0139 decode.loss_ce: 0.0063 decode.acc_seg: 99.6572 aux.loss_ce: 0.0076 aux.acc_seg: 98.8302 +04/19 01:05:25 - mmengine - INFO - Iter(train) [ 60150/160000] base_lr: 6.2997e-05 lr: 2.3291e-07 eta: 7 days, 13:07:00 time: 0.9888 data_time: 0.0043 memory: 8457 loss: 0.0134 decode.loss_ce: 0.0062 decode.acc_seg: 99.7871 aux.loss_ce: 0.0071 aux.acc_seg: 99.3629 +04/19 01:06:15 - mmengine - INFO - Iter(train) [ 60200/160000] base_lr: 6.2966e-05 lr: 2.3280e-07 eta: 5 days, 22:38:11 time: 0.9928 data_time: 0.0046 memory: 8457 loss: 0.0138 decode.loss_ce: 0.0060 decode.acc_seg: 99.8487 aux.loss_ce: 0.0078 aux.acc_seg: 99.4499 +04/19 01:07:04 - mmengine - INFO - Iter(train) [ 60250/160000] base_lr: 6.2934e-05 lr: 2.3268e-07 eta: 4 days, 23:33:04 time: 0.9917 data_time: 0.0044 memory: 8457 loss: 0.0126 decode.loss_ce: 0.0056 decode.acc_seg: 99.8344 aux.loss_ce: 0.0069 aux.acc_seg: 99.4980 +04/19 01:07:54 - mmengine - INFO - Iter(train) [ 60300/160000] base_lr: 6.2903e-05 lr: 2.3256e-07 eta: 4 days, 8:09:31 time: 0.9937 data_time: 0.0045 memory: 8457 loss: 0.0132 decode.loss_ce: 0.0060 decode.acc_seg: 99.7702 aux.loss_ce: 0.0072 aux.acc_seg: 99.2840 +04/19 01:08:44 - mmengine - INFO - Iter(train) [ 60350/160000] base_lr: 6.2871e-05 lr: 2.3245e-07 eta: 3 days, 21:09:39 time: 0.9942 data_time: 0.0044 memory: 8457 loss: 0.0135 decode.loss_ce: 0.0062 decode.acc_seg: 99.6994 aux.loss_ce: 0.0073 aux.acc_seg: 99.1688 +04/19 01:09:33 - mmengine - INFO - Iter(train) [ 60400/160000] base_lr: 6.2840e-05 lr: 2.3233e-07 eta: 3 days, 12:54:36 time: 0.9924 data_time: 0.0044 memory: 8457 loss: 0.0137 decode.loss_ce: 0.0061 decode.acc_seg: 99.7244 aux.loss_ce: 0.0076 aux.acc_seg: 99.0751 +04/19 01:10:23 - mmengine - INFO - Iter(train) [ 60450/160000] base_lr: 6.2808e-05 lr: 2.3221e-07 eta: 3 days, 6:29:25 time: 0.9936 data_time: 0.0041 memory: 8457 loss: 0.0134 decode.loss_ce: 0.0061 decode.acc_seg: 99.6300 aux.loss_ce: 0.0072 aux.acc_seg: 98.7514 +04/19 01:11:13 - mmengine - INFO - Iter(train) [ 60500/160000] base_lr: 6.2776e-05 lr: 2.3210e-07 eta: 3 days, 1:21:13 time: 0.9934 data_time: 0.0047 memory: 8457 loss: 0.0129 decode.loss_ce: 0.0059 decode.acc_seg: 99.6748 aux.loss_ce: 0.0070 aux.acc_seg: 99.3412 +04/19 01:12:02 - mmengine - INFO - Iter(train) [ 60550/160000] base_lr: 6.2745e-05 lr: 2.3198e-07 eta: 2 days, 21:08:55 time: 0.9952 data_time: 0.0043 memory: 8457 loss: 0.0139 decode.loss_ce: 0.0063 decode.acc_seg: 99.7656 aux.loss_ce: 0.0076 aux.acc_seg: 99.2676 +04/19 01:12:52 - mmengine - INFO - Iter(train) [ 60600/160000] base_lr: 6.2713e-05 lr: 2.3186e-07 eta: 2 days, 17:38:32 time: 0.9947 data_time: 0.0046 memory: 8457 loss: 0.0157 decode.loss_ce: 0.0070 decode.acc_seg: 99.6742 aux.loss_ce: 0.0086 aux.acc_seg: 98.7873 +04/19 01:13:42 - mmengine - INFO - Iter(train) [ 60650/160000] base_lr: 6.2682e-05 lr: 2.3175e-07 eta: 2 days, 14:40:24 time: 0.9943 data_time: 0.0048 memory: 8457 loss: 0.0146 decode.loss_ce: 0.0071 decode.acc_seg: 99.7372 aux.loss_ce: 0.0075 aux.acc_seg: 99.4942 +04/19 01:14:32 - mmengine - INFO - Iter(train) [ 60700/160000] base_lr: 6.2650e-05 lr: 2.3163e-07 eta: 2 days, 12:07:39 time: 0.9953 data_time: 0.0050 memory: 8457 loss: 0.0126 decode.loss_ce: 0.0057 decode.acc_seg: 99.8459 aux.loss_ce: 0.0069 aux.acc_seg: 99.5266 +04/19 01:15:21 - mmengine - INFO - Iter(train) [ 60750/160000] base_lr: 6.2619e-05 lr: 2.3151e-07 eta: 2 days, 9:55:12 time: 0.9950 data_time: 0.0046 memory: 8457 loss: 0.0128 decode.loss_ce: 0.0062 decode.acc_seg: 99.8405 aux.loss_ce: 0.0066 aux.acc_seg: 99.4726 +04/19 01:16:11 - mmengine - INFO - Iter(train) [ 60800/160000] base_lr: 6.2587e-05 lr: 2.3140e-07 eta: 2 days, 7:59:14 time: 0.9963 data_time: 0.0047 memory: 8457 loss: 0.0143 decode.loss_ce: 0.0066 decode.acc_seg: 99.6958 aux.loss_ce: 0.0077 aux.acc_seg: 98.9441 +04/19 01:17:01 - mmengine - INFO - Iter(train) [ 60850/160000] base_lr: 6.2556e-05 lr: 2.3128e-07 eta: 2 days, 6:16:51 time: 0.9958 data_time: 0.0045 memory: 8457 loss: 0.0141 decode.loss_ce: 0.0063 decode.acc_seg: 99.7522 aux.loss_ce: 0.0077 aux.acc_seg: 98.9187 +04/19 01:17:51 - mmengine - INFO - Iter(train) [ 60900/160000] base_lr: 6.2524e-05 lr: 2.3116e-07 eta: 2 days, 4:45:41 time: 0.9943 data_time: 0.0047 memory: 8457 loss: 0.0123 decode.loss_ce: 0.0056 decode.acc_seg: 99.8035 aux.loss_ce: 0.0067 aux.acc_seg: 99.2311 +04/19 01:18:40 - mmengine - INFO - Iter(train) [ 60950/160000] base_lr: 6.2493e-05 lr: 2.3105e-07 eta: 2 days, 3:24:03 time: 0.9961 data_time: 0.0044 memory: 8457 loss: 0.0149 decode.loss_ce: 0.0065 decode.acc_seg: 99.6803 aux.loss_ce: 0.0083 aux.acc_seg: 98.6212 +04/19 01:19:30 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240419_004857 +04/19 01:19:30 - mmengine - INFO - Iter(train) [ 61000/160000] base_lr: 6.2461e-05 lr: 2.3093e-07 eta: 2 days, 2:10:35 time: 0.9963 data_time: 0.0042 memory: 8457 loss: 0.0124 decode.loss_ce: 0.0057 decode.acc_seg: 99.7381 aux.loss_ce: 0.0067 aux.acc_seg: 99.1875 +04/19 01:20:20 - mmengine - INFO - Iter(train) [ 61050/160000] base_lr: 6.2429e-05 lr: 2.3081e-07 eta: 2 days, 1:03:59 time: 0.9958 data_time: 0.0046 memory: 8457 loss: 0.0152 decode.loss_ce: 0.0071 decode.acc_seg: 99.5174 aux.loss_ce: 0.0081 aux.acc_seg: 98.8636 +04/19 01:21:10 - mmengine - INFO - Iter(train) [ 61100/160000] base_lr: 6.2398e-05 lr: 2.3070e-07 eta: 2 days, 0:03:25 time: 0.9957 data_time: 0.0043 memory: 8457 loss: 0.0155 decode.loss_ce: 0.0076 decode.acc_seg: 99.6033 aux.loss_ce: 0.0079 aux.acc_seg: 99.1104 +04/19 01:22:00 - mmengine - INFO - Iter(train) [ 61150/160000] base_lr: 6.2366e-05 lr: 2.3058e-07 eta: 1 day, 23:08:02 time: 0.9969 data_time: 0.0045 memory: 8457 loss: 0.0139 decode.loss_ce: 0.0061 decode.acc_seg: 99.7784 aux.loss_ce: 0.0079 aux.acc_seg: 99.1186 +04/19 01:22:50 - mmengine - INFO - Iter(train) [ 61200/160000] base_lr: 6.2335e-05 lr: 2.3046e-07 eta: 1 day, 22:17:13 time: 0.9957 data_time: 0.0048 memory: 8457 loss: 0.0144 decode.loss_ce: 0.0065 decode.acc_seg: 99.7452 aux.loss_ce: 0.0079 aux.acc_seg: 99.3444 +04/19 01:23:39 - mmengine - INFO - Iter(train) [ 61250/160000] base_lr: 6.2303e-05 lr: 2.3035e-07 eta: 1 day, 21:30:22 time: 0.9961 data_time: 0.0044 memory: 8457 loss: 0.0153 decode.loss_ce: 0.0069 decode.acc_seg: 99.7793 aux.loss_ce: 0.0085 aux.acc_seg: 99.4120 +04/19 01:24:29 - mmengine - INFO - Iter(train) [ 61300/160000] base_lr: 6.2272e-05 lr: 2.3023e-07 eta: 1 day, 20:47:05 time: 0.9964 data_time: 0.0050 memory: 8457 loss: 0.0133 decode.loss_ce: 0.0061 decode.acc_seg: 99.7265 aux.loss_ce: 0.0072 aux.acc_seg: 99.2550 +04/19 01:25:19 - mmengine - INFO - Iter(train) [ 61350/160000] base_lr: 6.2240e-05 lr: 2.3011e-07 eta: 1 day, 20:06:58 time: 0.9965 data_time: 0.0042 memory: 8457 loss: 0.0125 decode.loss_ce: 0.0057 decode.acc_seg: 99.7890 aux.loss_ce: 0.0067 aux.acc_seg: 99.2046 +04/19 01:26:09 - mmengine - INFO - Iter(train) [ 61400/160000] base_lr: 6.2209e-05 lr: 2.3000e-07 eta: 1 day, 19:29:38 time: 0.9963 data_time: 0.0046 memory: 8457 loss: 0.0133 decode.loss_ce: 0.0059 decode.acc_seg: 99.7929 aux.loss_ce: 0.0074 aux.acc_seg: 99.2052 +04/19 01:26:59 - mmengine - INFO - Iter(train) [ 61450/160000] base_lr: 6.2177e-05 lr: 2.2988e-07 eta: 1 day, 18:54:50 time: 0.9971 data_time: 0.0044 memory: 8457 loss: 0.0141 decode.loss_ce: 0.0066 decode.acc_seg: 99.7149 aux.loss_ce: 0.0075 aux.acc_seg: 99.1526 +04/19 01:27:49 - mmengine - INFO - Iter(train) [ 61500/160000] base_lr: 6.2146e-05 lr: 2.2976e-07 eta: 1 day, 18:22:20 time: 0.9974 data_time: 0.0043 memory: 8457 loss: 0.0145 decode.loss_ce: 0.0066 decode.acc_seg: 99.7095 aux.loss_ce: 0.0079 aux.acc_seg: 98.9267 +04/19 01:28:39 - mmengine - INFO - Iter(train) [ 61550/160000] base_lr: 6.2114e-05 lr: 2.2965e-07 eta: 1 day, 17:51:51 time: 0.9969 data_time: 0.0050 memory: 8457 loss: 0.0143 decode.loss_ce: 0.0063 decode.acc_seg: 99.7963 aux.loss_ce: 0.0079 aux.acc_seg: 99.4560 +04/19 01:29:28 - mmengine - INFO - Iter(train) [ 61600/160000] base_lr: 6.2082e-05 lr: 2.2953e-07 eta: 1 day, 17:23:13 time: 0.9973 data_time: 0.0046 memory: 8457 loss: 0.0139 decode.loss_ce: 0.0062 decode.acc_seg: 99.7114 aux.loss_ce: 0.0077 aux.acc_seg: 98.9637 +04/19 01:30:18 - mmengine - INFO - Iter(train) [ 61650/160000] base_lr: 6.2051e-05 lr: 2.2941e-07 eta: 1 day, 16:56:18 time: 0.9978 data_time: 0.0046 memory: 8457 loss: 0.0144 decode.loss_ce: 0.0068 decode.acc_seg: 99.7086 aux.loss_ce: 0.0076 aux.acc_seg: 98.9288 +04/19 01:31:08 - mmengine - INFO - Iter(train) [ 61700/160000] base_lr: 6.2019e-05 lr: 2.2930e-07 eta: 1 day, 16:30:53 time: 0.9968 data_time: 0.0045 memory: 8457 loss: 0.0149 decode.loss_ce: 0.0068 decode.acc_seg: 99.5613 aux.loss_ce: 0.0081 aux.acc_seg: 98.4583 +04/19 01:31:58 - mmengine - INFO - Iter(train) [ 61750/160000] base_lr: 6.1988e-05 lr: 2.2918e-07 eta: 1 day, 16:06:54 time: 0.9974 data_time: 0.0049 memory: 8457 loss: 0.0127 decode.loss_ce: 0.0056 decode.acc_seg: 99.7421 aux.loss_ce: 0.0071 aux.acc_seg: 99.1478 +04/19 01:32:48 - mmengine - INFO - Iter(train) [ 61800/160000] base_lr: 6.1956e-05 lr: 2.2906e-07 eta: 1 day, 15:44:10 time: 0.9975 data_time: 0.0045 memory: 8457 loss: 0.0129 decode.loss_ce: 0.0059 decode.acc_seg: 99.7810 aux.loss_ce: 0.0070 aux.acc_seg: 99.4858 +04/19 01:33:38 - mmengine - INFO - Iter(train) [ 61850/160000] base_lr: 6.1925e-05 lr: 2.2895e-07 eta: 1 day, 15:22:38 time: 0.9975 data_time: 0.0044 memory: 8457 loss: 0.0132 decode.loss_ce: 0.0060 decode.acc_seg: 99.8119 aux.loss_ce: 0.0072 aux.acc_seg: 99.4720 +04/19 01:34:28 - mmengine - INFO - Iter(train) [ 61900/160000] base_lr: 6.1893e-05 lr: 2.2883e-07 eta: 1 day, 15:02:13 time: 0.9977 data_time: 0.0044 memory: 8457 loss: 0.0134 decode.loss_ce: 0.0059 decode.acc_seg: 99.7007 aux.loss_ce: 0.0075 aux.acc_seg: 99.0156 +04/19 01:35:17 - mmengine - INFO - Iter(train) [ 61950/160000] base_lr: 6.1862e-05 lr: 2.2872e-07 eta: 1 day, 14:42:45 time: 0.9968 data_time: 0.0045 memory: 8457 loss: 0.0126 decode.loss_ce: 0.0060 decode.acc_seg: 99.7168 aux.loss_ce: 0.0066 aux.acc_seg: 99.2653 +04/19 01:36:07 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240419_004857 +04/19 01:36:07 - mmengine - INFO - Iter(train) [ 62000/160000] base_lr: 6.1830e-05 lr: 2.2860e-07 eta: 1 day, 14:24:15 time: 0.9968 data_time: 0.0047 memory: 8457 loss: 0.0138 decode.loss_ce: 0.0061 decode.acc_seg: 99.7591 aux.loss_ce: 0.0077 aux.acc_seg: 99.4114 +04/19 01:36:57 - mmengine - INFO - Iter(train) [ 62050/160000] base_lr: 6.1798e-05 lr: 2.2848e-07 eta: 1 day, 14:06:34 time: 0.9956 data_time: 0.0048 memory: 8457 loss: 0.0146 decode.loss_ce: 0.0066 decode.acc_seg: 99.6675 aux.loss_ce: 0.0079 aux.acc_seg: 98.9210 +04/19 01:37:47 - mmengine - INFO - Iter(train) [ 62100/160000] base_lr: 6.1767e-05 lr: 2.2837e-07 eta: 1 day, 13:49:42 time: 0.9973 data_time: 0.0043 memory: 8457 loss: 0.0137 decode.loss_ce: 0.0063 decode.acc_seg: 99.6946 aux.loss_ce: 0.0073 aux.acc_seg: 99.0738 +04/19 01:38:37 - mmengine - INFO - Iter(train) [ 62150/160000] base_lr: 6.1735e-05 lr: 2.2825e-07 eta: 1 day, 13:33:36 time: 0.9988 data_time: 0.0043 memory: 8457 loss: 0.0121 decode.loss_ce: 0.0057 decode.acc_seg: 99.7206 aux.loss_ce: 0.0064 aux.acc_seg: 99.2146 +04/19 01:39:27 - mmengine - INFO - Iter(train) [ 62200/160000] base_lr: 6.1704e-05 lr: 2.2813e-07 eta: 1 day, 13:18:14 time: 0.9995 data_time: 0.0047 memory: 8457 loss: 0.0125 decode.loss_ce: 0.0057 decode.acc_seg: 99.7284 aux.loss_ce: 0.0068 aux.acc_seg: 99.0816 +04/19 01:40:17 - mmengine - INFO - Iter(train) [ 62250/160000] base_lr: 6.1672e-05 lr: 2.2802e-07 eta: 1 day, 13:03:29 time: 0.9986 data_time: 0.0046 memory: 8457 loss: 0.0129 decode.loss_ce: 0.0056 decode.acc_seg: 99.7137 aux.loss_ce: 0.0073 aux.acc_seg: 99.1442 +04/19 01:41:07 - mmengine - INFO - Iter(train) [ 62300/160000] base_lr: 6.1641e-05 lr: 2.2790e-07 eta: 1 day, 12:49:20 time: 0.9961 data_time: 0.0049 memory: 8457 loss: 0.0120 decode.loss_ce: 0.0055 decode.acc_seg: 99.6380 aux.loss_ce: 0.0065 aux.acc_seg: 98.8453 +04/19 01:41:56 - mmengine - INFO - Iter(train) [ 62350/160000] base_lr: 6.1609e-05 lr: 2.2778e-07 eta: 1 day, 12:35:48 time: 0.9994 data_time: 0.0047 memory: 8457 loss: 0.0160 decode.loss_ce: 0.0076 decode.acc_seg: 99.7267 aux.loss_ce: 0.0084 aux.acc_seg: 99.2651 +04/19 01:42:46 - mmengine - INFO - Iter(train) [ 62400/160000] base_lr: 6.1578e-05 lr: 2.2767e-07 eta: 1 day, 12:22:46 time: 0.9967 data_time: 0.0049 memory: 8457 loss: 0.0145 decode.loss_ce: 0.0070 decode.acc_seg: 99.6550 aux.loss_ce: 0.0076 aux.acc_seg: 98.7083 +04/19 01:43:36 - mmengine - INFO - Iter(train) [ 62450/160000] base_lr: 6.1546e-05 lr: 2.2755e-07 eta: 1 day, 12:10:14 time: 0.9978 data_time: 0.0043 memory: 8457 loss: 0.0149 decode.loss_ce: 0.0072 decode.acc_seg: 99.7229 aux.loss_ce: 0.0076 aux.acc_seg: 99.4339 +04/19 01:44:26 - mmengine - INFO - Iter(train) [ 62500/160000] base_lr: 6.1515e-05 lr: 2.2743e-07 eta: 1 day, 11:58:09 time: 0.9980 data_time: 0.0042 memory: 8457 loss: 0.0160 decode.loss_ce: 0.0074 decode.acc_seg: 99.7686 aux.loss_ce: 0.0086 aux.acc_seg: 99.4291 +04/19 01:45:16 - mmengine - INFO - Iter(train) [ 62550/160000] base_lr: 6.1483e-05 lr: 2.2732e-07 eta: 1 day, 11:46:30 time: 0.9972 data_time: 0.0043 memory: 8457 loss: 0.0128 decode.loss_ce: 0.0059 decode.acc_seg: 99.7599 aux.loss_ce: 0.0069 aux.acc_seg: 99.1179 +04/19 01:46:06 - mmengine - INFO - Iter(train) [ 62600/160000] base_lr: 6.1451e-05 lr: 2.2720e-07 eta: 1 day, 11:35:17 time: 0.9975 data_time: 0.0045 memory: 8457 loss: 0.0128 decode.loss_ce: 0.0057 decode.acc_seg: 99.6805 aux.loss_ce: 0.0071 aux.acc_seg: 99.0713 +04/19 01:46:56 - mmengine - INFO - Iter(train) [ 62650/160000] base_lr: 6.1420e-05 lr: 2.2708e-07 eta: 1 day, 11:24:25 time: 0.9953 data_time: 0.0046 memory: 8457 loss: 0.0151 decode.loss_ce: 0.0067 decode.acc_seg: 99.6197 aux.loss_ce: 0.0084 aux.acc_seg: 99.1144 +04/19 01:47:46 - mmengine - INFO - Iter(train) [ 62700/160000] base_lr: 6.1388e-05 lr: 2.2697e-07 eta: 1 day, 11:13:55 time: 0.9955 data_time: 0.0043 memory: 8457 loss: 0.0145 decode.loss_ce: 0.0064 decode.acc_seg: 99.6607 aux.loss_ce: 0.0082 aux.acc_seg: 99.1154 +04/19 01:48:35 - mmengine - INFO - Iter(train) [ 62750/160000] base_lr: 6.1357e-05 lr: 2.2685e-07 eta: 1 day, 11:03:46 time: 0.9950 data_time: 0.0042 memory: 8457 loss: 0.0142 decode.loss_ce: 0.0065 decode.acc_seg: 99.7042 aux.loss_ce: 0.0077 aux.acc_seg: 99.1091 +04/19 01:49:25 - mmengine - INFO - Iter(train) [ 62800/160000] base_lr: 6.1325e-05 lr: 2.2673e-07 eta: 1 day, 10:53:58 time: 0.9973 data_time: 0.0043 memory: 8457 loss: 0.0143 decode.loss_ce: 0.0066 decode.acc_seg: 99.6748 aux.loss_ce: 0.0077 aux.acc_seg: 99.3170 +04/19 01:50:15 - mmengine - INFO - Iter(train) [ 62850/160000] base_lr: 6.1294e-05 lr: 2.2662e-07 eta: 1 day, 10:44:30 time: 0.9962 data_time: 0.0049 memory: 8457 loss: 0.0137 decode.loss_ce: 0.0060 decode.acc_seg: 99.7328 aux.loss_ce: 0.0077 aux.acc_seg: 99.1205 +04/19 01:51:05 - mmengine - INFO - Iter(train) [ 62900/160000] base_lr: 6.1262e-05 lr: 2.2650e-07 eta: 1 day, 10:35:19 time: 0.9975 data_time: 0.0046 memory: 8457 loss: 0.0149 decode.loss_ce: 0.0069 decode.acc_seg: 99.7749 aux.loss_ce: 0.0080 aux.acc_seg: 99.4528 +04/19 01:51:55 - mmengine - INFO - Iter(train) [ 62950/160000] base_lr: 6.1231e-05 lr: 2.2638e-07 eta: 1 day, 10:26:27 time: 0.9965 data_time: 0.0049 memory: 8457 loss: 0.0142 decode.loss_ce: 0.0068 decode.acc_seg: 99.7158 aux.loss_ce: 0.0074 aux.acc_seg: 99.1865 +04/19 01:52:45 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240419_004857 +04/19 01:52:45 - mmengine - INFO - Iter(train) [ 63000/160000] base_lr: 6.1199e-05 lr: 2.2627e-07 eta: 1 day, 10:17:50 time: 0.9970 data_time: 0.0044 memory: 8457 loss: 0.0127 decode.loss_ce: 0.0057 decode.acc_seg: 99.7782 aux.loss_ce: 0.0071 aux.acc_seg: 99.3174 +04/19 01:53:35 - mmengine - INFO - Iter(train) [ 63050/160000] base_lr: 6.1168e-05 lr: 2.2615e-07 eta: 1 day, 10:09:29 time: 0.9978 data_time: 0.0046 memory: 8457 loss: 0.0138 decode.loss_ce: 0.0063 decode.acc_seg: 99.6641 aux.loss_ce: 0.0075 aux.acc_seg: 98.9759 +04/19 01:54:24 - mmengine - INFO - Iter(train) [ 63100/160000] base_lr: 6.1136e-05 lr: 2.2603e-07 eta: 1 day, 10:01:24 time: 0.9978 data_time: 0.0045 memory: 8457 loss: 0.0148 decode.loss_ce: 0.0070 decode.acc_seg: 99.5815 aux.loss_ce: 0.0078 aux.acc_seg: 99.1236 +04/19 01:55:14 - mmengine - INFO - Iter(train) [ 63150/160000] base_lr: 6.1104e-05 lr: 2.2592e-07 eta: 1 day, 9:53:32 time: 0.9985 data_time: 0.0044 memory: 8457 loss: 0.0136 decode.loss_ce: 0.0061 decode.acc_seg: 99.7086 aux.loss_ce: 0.0075 aux.acc_seg: 98.6923 +04/19 01:56:04 - mmengine - INFO - Iter(train) [ 63200/160000] base_lr: 6.1073e-05 lr: 2.2580e-07 eta: 1 day, 9:45:53 time: 0.9989 data_time: 0.0046 memory: 8457 loss: 0.0123 decode.loss_ce: 0.0057 decode.acc_seg: 99.8119 aux.loss_ce: 0.0067 aux.acc_seg: 99.3879 +04/19 01:56:54 - mmengine - INFO - Iter(train) [ 63250/160000] base_lr: 6.1041e-05 lr: 2.2568e-07 eta: 1 day, 9:38:26 time: 0.9975 data_time: 0.0048 memory: 8457 loss: 0.0138 decode.loss_ce: 0.0061 decode.acc_seg: 99.6487 aux.loss_ce: 0.0077 aux.acc_seg: 99.1943 +04/19 01:57:44 - mmengine - INFO - Iter(train) [ 63300/160000] base_lr: 6.1010e-05 lr: 2.2557e-07 eta: 1 day, 9:31:13 time: 0.9984 data_time: 0.0041 memory: 8457 loss: 0.0128 decode.loss_ce: 0.0059 decode.acc_seg: 99.7293 aux.loss_ce: 0.0069 aux.acc_seg: 99.0902 +04/19 01:58:34 - mmengine - INFO - Iter(train) [ 63350/160000] base_lr: 6.0978e-05 lr: 2.2545e-07 eta: 1 day, 9:24:11 time: 0.9992 data_time: 0.0044 memory: 8457 loss: 0.0150 decode.loss_ce: 0.0067 decode.acc_seg: 99.6496 aux.loss_ce: 0.0083 aux.acc_seg: 98.8878 +04/19 01:59:24 - mmengine - INFO - Iter(train) [ 63400/160000] base_lr: 6.0947e-05 lr: 2.2533e-07 eta: 1 day, 9:17:21 time: 0.9982 data_time: 0.0042 memory: 8457 loss: 0.0150 decode.loss_ce: 0.0067 decode.acc_seg: 99.7051 aux.loss_ce: 0.0083 aux.acc_seg: 99.0618 +04/19 02:00:14 - mmengine - INFO - Iter(train) [ 63450/160000] base_lr: 6.0915e-05 lr: 2.2522e-07 eta: 1 day, 9:10:39 time: 0.9982 data_time: 0.0045 memory: 8457 loss: 0.0134 decode.loss_ce: 0.0063 decode.acc_seg: 99.7469 aux.loss_ce: 0.0071 aux.acc_seg: 99.3727 +04/19 02:01:04 - mmengine - INFO - Iter(train) [ 63500/160000] base_lr: 6.0884e-05 lr: 2.2510e-07 eta: 1 day, 9:04:09 time: 0.9986 data_time: 0.0045 memory: 8457 loss: 0.0134 decode.loss_ce: 0.0060 decode.acc_seg: 99.8259 aux.loss_ce: 0.0074 aux.acc_seg: 99.4972 +04/19 02:01:54 - mmengine - INFO - Iter(train) [ 63550/160000] base_lr: 6.0852e-05 lr: 2.2498e-07 eta: 1 day, 8:57:47 time: 0.9979 data_time: 0.0045 memory: 8457 loss: 0.0143 decode.loss_ce: 0.0062 decode.acc_seg: 99.8463 aux.loss_ce: 0.0081 aux.acc_seg: 99.3818 +04/19 02:02:44 - mmengine - INFO - Iter(train) [ 63600/160000] base_lr: 6.0821e-05 lr: 2.2487e-07 eta: 1 day, 8:51:36 time: 0.9979 data_time: 0.0043 memory: 8457 loss: 0.0138 decode.loss_ce: 0.0060 decode.acc_seg: 99.7198 aux.loss_ce: 0.0078 aux.acc_seg: 99.1198 +04/19 02:03:34 - mmengine - INFO - Iter(train) [ 63650/160000] base_lr: 6.0789e-05 lr: 2.2475e-07 eta: 1 day, 8:45:34 time: 0.9982 data_time: 0.0044 memory: 8457 loss: 0.0149 decode.loss_ce: 0.0068 decode.acc_seg: 99.7103 aux.loss_ce: 0.0081 aux.acc_seg: 99.2104 +04/19 02:04:24 - mmengine - INFO - Iter(train) [ 63700/160000] base_lr: 6.0757e-05 lr: 2.2463e-07 eta: 1 day, 8:39:39 time: 0.9981 data_time: 0.0045 memory: 8457 loss: 0.0150 decode.loss_ce: 0.0066 decode.acc_seg: 99.7835 aux.loss_ce: 0.0084 aux.acc_seg: 99.3408 +04/19 02:05:14 - mmengine - INFO - Iter(train) [ 63750/160000] base_lr: 6.0726e-05 lr: 2.2452e-07 eta: 1 day, 8:33:53 time: 0.9985 data_time: 0.0048 memory: 8457 loss: 0.0162 decode.loss_ce: 0.0075 decode.acc_seg: 99.7694 aux.loss_ce: 0.0087 aux.acc_seg: 99.3509 +04/19 02:06:04 - mmengine - INFO - Iter(train) [ 63800/160000] base_lr: 6.0694e-05 lr: 2.2440e-07 eta: 1 day, 8:28:14 time: 0.9988 data_time: 0.0045 memory: 8457 loss: 0.0140 decode.loss_ce: 0.0063 decode.acc_seg: 99.7492 aux.loss_ce: 0.0077 aux.acc_seg: 99.1364 +04/19 02:06:53 - mmengine - INFO - Iter(train) [ 63850/160000] base_lr: 6.0663e-05 lr: 2.2428e-07 eta: 1 day, 8:22:43 time: 0.9984 data_time: 0.0047 memory: 8457 loss: 0.0160 decode.loss_ce: 0.0080 decode.acc_seg: 99.5464 aux.loss_ce: 0.0080 aux.acc_seg: 98.6904 +04/19 02:07:43 - mmengine - INFO - Iter(train) [ 63900/160000] base_lr: 6.0631e-05 lr: 2.2417e-07 eta: 1 day, 8:17:20 time: 0.9994 data_time: 0.0049 memory: 8457 loss: 0.0138 decode.loss_ce: 0.0062 decode.acc_seg: 99.8302 aux.loss_ce: 0.0076 aux.acc_seg: 99.4253 +04/19 02:08:33 - mmengine - INFO - Iter(train) [ 63950/160000] base_lr: 6.0600e-05 lr: 2.2405e-07 eta: 1 day, 8:12:03 time: 0.9984 data_time: 0.0046 memory: 8457 loss: 0.0143 decode.loss_ce: 0.0064 decode.acc_seg: 99.6445 aux.loss_ce: 0.0080 aux.acc_seg: 98.8039 +04/19 02:09:23 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240419_004857 +04/19 02:09:23 - mmengine - INFO - Iter(train) [ 64000/160000] base_lr: 6.0568e-05 lr: 2.2393e-07 eta: 1 day, 8:06:53 time: 0.9994 data_time: 0.0049 memory: 8457 loss: 0.0136 decode.loss_ce: 0.0064 decode.acc_seg: 99.7782 aux.loss_ce: 0.0072 aux.acc_seg: 99.2029 +04/19 02:10:13 - mmengine - INFO - Iter(train) [ 64050/160000] base_lr: 6.0537e-05 lr: 2.2382e-07 eta: 1 day, 8:01:50 time: 0.9994 data_time: 0.0045 memory: 8457 loss: 0.0144 decode.loss_ce: 0.0069 decode.acc_seg: 99.7158 aux.loss_ce: 0.0075 aux.acc_seg: 99.2134 +04/19 02:11:03 - mmengine - INFO - Iter(train) [ 64100/160000] base_lr: 6.0505e-05 lr: 2.2370e-07 eta: 1 day, 7:56:52 time: 0.9975 data_time: 0.0044 memory: 8457 loss: 0.0134 decode.loss_ce: 0.0061 decode.acc_seg: 99.6708 aux.loss_ce: 0.0073 aux.acc_seg: 99.0332 +04/19 02:11:53 - mmengine - INFO - Iter(train) [ 64150/160000] base_lr: 6.0474e-05 lr: 2.2358e-07 eta: 1 day, 7:52:00 time: 0.9988 data_time: 0.0047 memory: 8457 loss: 0.0118 decode.loss_ce: 0.0056 decode.acc_seg: 99.7551 aux.loss_ce: 0.0063 aux.acc_seg: 99.2603 +04/19 02:12:43 - mmengine - INFO - Iter(train) [ 64200/160000] base_lr: 6.0442e-05 lr: 2.2347e-07 eta: 1 day, 7:47:14 time: 0.9996 data_time: 0.0045 memory: 8457 loss: 0.0159 decode.loss_ce: 0.0073 decode.acc_seg: 99.7343 aux.loss_ce: 0.0086 aux.acc_seg: 99.1961 +04/19 02:13:33 - mmengine - INFO - Iter(train) [ 64250/160000] base_lr: 6.0410e-05 lr: 2.2335e-07 eta: 1 day, 7:42:35 time: 0.9989 data_time: 0.0046 memory: 8457 loss: 0.0123 decode.loss_ce: 0.0052 decode.acc_seg: 99.7702 aux.loss_ce: 0.0071 aux.acc_seg: 99.3006 +04/19 02:14:23 - mmengine - INFO - Iter(train) [ 64300/160000] base_lr: 6.0379e-05 lr: 2.2323e-07 eta: 1 day, 7:38:00 time: 0.9994 data_time: 0.0043 memory: 8457 loss: 0.0139 decode.loss_ce: 0.0064 decode.acc_seg: 99.7328 aux.loss_ce: 0.0075 aux.acc_seg: 99.1760 +04/19 02:15:13 - mmengine - INFO - Iter(train) [ 64350/160000] base_lr: 6.0347e-05 lr: 2.2312e-07 eta: 1 day, 7:33:31 time: 0.9995 data_time: 0.0047 memory: 8457 loss: 0.0140 decode.loss_ce: 0.0065 decode.acc_seg: 99.6801 aux.loss_ce: 0.0075 aux.acc_seg: 99.0461 +04/19 02:16:03 - mmengine - INFO - Iter(train) [ 64400/160000] base_lr: 6.0316e-05 lr: 2.2300e-07 eta: 1 day, 7:29:07 time: 0.9992 data_time: 0.0043 memory: 8457 loss: 0.0144 decode.loss_ce: 0.0062 decode.acc_seg: 99.7795 aux.loss_ce: 0.0082 aux.acc_seg: 99.2189 +04/19 02:16:53 - mmengine - INFO - Iter(train) [ 64450/160000] base_lr: 6.0284e-05 lr: 2.2288e-07 eta: 1 day, 7:24:46 time: 0.9977 data_time: 0.0044 memory: 8457 loss: 0.0136 decode.loss_ce: 0.0063 decode.acc_seg: 99.7478 aux.loss_ce: 0.0073 aux.acc_seg: 99.2088 +04/19 02:17:43 - mmengine - INFO - Iter(train) [ 64500/160000] base_lr: 6.0253e-05 lr: 2.2277e-07 eta: 1 day, 7:20:31 time: 0.9998 data_time: 0.0046 memory: 8457 loss: 0.0140 decode.loss_ce: 0.0061 decode.acc_seg: 99.7173 aux.loss_ce: 0.0079 aux.acc_seg: 99.2582 +04/19 02:18:33 - mmengine - INFO - Iter(train) [ 64550/160000] base_lr: 6.0221e-05 lr: 2.2265e-07 eta: 1 day, 7:16:20 time: 0.9993 data_time: 0.0048 memory: 8457 loss: 0.0125 decode.loss_ce: 0.0059 decode.acc_seg: 99.8030 aux.loss_ce: 0.0066 aux.acc_seg: 99.3591 +04/19 02:19:23 - mmengine - INFO - Iter(train) [ 64600/160000] base_lr: 6.0190e-05 lr: 2.2253e-07 eta: 1 day, 7:12:15 time: 0.9984 data_time: 0.0043 memory: 8457 loss: 0.0142 decode.loss_ce: 0.0061 decode.acc_seg: 99.6122 aux.loss_ce: 0.0081 aux.acc_seg: 98.8277 +04/19 02:20:13 - mmengine - INFO - Iter(train) [ 64650/160000] base_lr: 6.0158e-05 lr: 2.2242e-07 eta: 1 day, 7:08:13 time: 0.9981 data_time: 0.0045 memory: 8457 loss: 0.0147 decode.loss_ce: 0.0067 decode.acc_seg: 99.6441 aux.loss_ce: 0.0080 aux.acc_seg: 99.1022 +04/19 02:21:03 - mmengine - INFO - Iter(train) [ 64700/160000] base_lr: 6.0127e-05 lr: 2.2230e-07 eta: 1 day, 7:04:16 time: 1.0002 data_time: 0.0048 memory: 8457 loss: 0.0139 decode.loss_ce: 0.0063 decode.acc_seg: 99.7303 aux.loss_ce: 0.0076 aux.acc_seg: 99.0334 +04/19 02:21:53 - mmengine - INFO - Iter(train) [ 64750/160000] base_lr: 6.0095e-05 lr: 2.2218e-07 eta: 1 day, 7:00:22 time: 1.0001 data_time: 0.0045 memory: 8457 loss: 0.0115 decode.loss_ce: 0.0051 decode.acc_seg: 99.8207 aux.loss_ce: 0.0063 aux.acc_seg: 99.4606 +04/19 02:22:43 - mmengine - INFO - Iter(train) [ 64800/160000] base_lr: 6.0063e-05 lr: 2.2207e-07 eta: 1 day, 6:56:33 time: 0.9993 data_time: 0.0043 memory: 8457 loss: 0.0145 decode.loss_ce: 0.0067 decode.acc_seg: 99.7196 aux.loss_ce: 0.0077 aux.acc_seg: 99.0515 +04/19 02:23:33 - mmengine - INFO - Iter(train) [ 64850/160000] base_lr: 6.0032e-05 lr: 2.2195e-07 eta: 1 day, 6:52:47 time: 1.0000 data_time: 0.0043 memory: 8457 loss: 0.0130 decode.loss_ce: 0.0058 decode.acc_seg: 99.7248 aux.loss_ce: 0.0072 aux.acc_seg: 98.9946 +04/19 02:24:23 - mmengine - INFO - Iter(train) [ 64900/160000] base_lr: 6.0000e-05 lr: 2.2183e-07 eta: 1 day, 6:49:05 time: 0.9998 data_time: 0.0045 memory: 8457 loss: 0.0133 decode.loss_ce: 0.0059 decode.acc_seg: 99.7585 aux.loss_ce: 0.0074 aux.acc_seg: 99.0746 +04/19 02:25:13 - mmengine - INFO - Iter(train) [ 64950/160000] base_lr: 5.9969e-05 lr: 2.2172e-07 eta: 1 day, 6:45:27 time: 1.0009 data_time: 0.0044 memory: 8457 loss: 0.0144 decode.loss_ce: 0.0064 decode.acc_seg: 99.8133 aux.loss_ce: 0.0080 aux.acc_seg: 99.3851 +04/19 02:26:03 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240419_004857 +04/19 02:26:03 - mmengine - INFO - Iter(train) [ 65000/160000] base_lr: 5.9937e-05 lr: 2.2160e-07 eta: 1 day, 6:41:51 time: 0.9989 data_time: 0.0046 memory: 8457 loss: 0.0119 decode.loss_ce: 0.0057 decode.acc_seg: 99.7786 aux.loss_ce: 0.0062 aux.acc_seg: 99.2167 +04/19 02:26:53 - mmengine - INFO - Iter(train) [ 65050/160000] base_lr: 5.9906e-05 lr: 2.2148e-07 eta: 1 day, 6:38:19 time: 0.9993 data_time: 0.0048 memory: 8457 loss: 0.0127 decode.loss_ce: 0.0057 decode.acc_seg: 99.7431 aux.loss_ce: 0.0069 aux.acc_seg: 98.9317 +04/19 02:27:43 - mmengine - INFO - Iter(train) [ 65100/160000] base_lr: 5.9874e-05 lr: 2.2137e-07 eta: 1 day, 6:34:50 time: 0.9985 data_time: 0.0048 memory: 8457 loss: 0.0146 decode.loss_ce: 0.0067 decode.acc_seg: 99.7698 aux.loss_ce: 0.0080 aux.acc_seg: 99.2094 +04/19 02:28:33 - mmengine - INFO - Iter(train) [ 65150/160000] base_lr: 5.9843e-05 lr: 2.2125e-07 eta: 1 day, 6:31:24 time: 0.9998 data_time: 0.0048 memory: 8457 loss: 0.0146 decode.loss_ce: 0.0075 decode.acc_seg: 99.7412 aux.loss_ce: 0.0071 aux.acc_seg: 99.1184 +04/19 02:29:23 - mmengine - INFO - Iter(train) [ 65200/160000] base_lr: 5.9811e-05 lr: 2.2113e-07 eta: 1 day, 6:28:01 time: 0.9999 data_time: 0.0043 memory: 8457 loss: 0.0144 decode.loss_ce: 0.0066 decode.acc_seg: 99.7622 aux.loss_ce: 0.0077 aux.acc_seg: 99.2657 +04/19 02:30:12 - mmengine - INFO - Iter(train) [ 65250/160000] base_lr: 5.9780e-05 lr: 2.2102e-07 eta: 1 day, 6:24:41 time: 0.9994 data_time: 0.0046 memory: 8457 loss: 0.0138 decode.loss_ce: 0.0063 decode.acc_seg: 99.7765 aux.loss_ce: 0.0075 aux.acc_seg: 99.2792 +04/19 02:31:02 - mmengine - INFO - Iter(train) [ 65300/160000] base_lr: 5.9748e-05 lr: 2.2090e-07 eta: 1 day, 6:21:23 time: 0.9990 data_time: 0.0046 memory: 8457 loss: 0.0142 decode.loss_ce: 0.0065 decode.acc_seg: 99.7976 aux.loss_ce: 0.0078 aux.acc_seg: 99.2073 +04/19 02:31:52 - mmengine - INFO - Iter(train) [ 65350/160000] base_lr: 5.9716e-05 lr: 2.2078e-07 eta: 1 day, 6:18:08 time: 0.9993 data_time: 0.0045 memory: 8457 loss: 0.0148 decode.loss_ce: 0.0065 decode.acc_seg: 99.7171 aux.loss_ce: 0.0083 aux.acc_seg: 98.8476 +04/19 02:32:42 - mmengine - INFO - Iter(train) [ 65400/160000] base_lr: 5.9685e-05 lr: 2.2067e-07 eta: 1 day, 6:14:56 time: 0.9988 data_time: 0.0045 memory: 8457 loss: 0.0132 decode.loss_ce: 0.0060 decode.acc_seg: 99.7591 aux.loss_ce: 0.0072 aux.acc_seg: 99.3172 +04/19 02:33:32 - mmengine - INFO - Iter(train) [ 65450/160000] base_lr: 5.9653e-05 lr: 2.2055e-07 eta: 1 day, 6:11:47 time: 1.0007 data_time: 0.0048 memory: 8457 loss: 0.0137 decode.loss_ce: 0.0064 decode.acc_seg: 99.7240 aux.loss_ce: 0.0073 aux.acc_seg: 99.0656 +04/19 02:34:22 - mmengine - INFO - Iter(train) [ 65500/160000] base_lr: 5.9622e-05 lr: 2.2043e-07 eta: 1 day, 6:08:40 time: 0.9988 data_time: 0.0043 memory: 8457 loss: 0.0148 decode.loss_ce: 0.0071 decode.acc_seg: 99.4360 aux.loss_ce: 0.0077 aux.acc_seg: 99.0091 +04/19 02:35:12 - mmengine - INFO - Iter(train) [ 65550/160000] base_lr: 5.9590e-05 lr: 2.2032e-07 eta: 1 day, 6:05:36 time: 1.0011 data_time: 0.0048 memory: 8457 loss: 0.0130 decode.loss_ce: 0.0058 decode.acc_seg: 99.7591 aux.loss_ce: 0.0072 aux.acc_seg: 99.2933 +04/19 02:36:02 - mmengine - INFO - Iter(train) [ 65600/160000] base_lr: 5.9559e-05 lr: 2.2020e-07 eta: 1 day, 6:02:34 time: 0.9995 data_time: 0.0045 memory: 8457 loss: 0.0130 decode.loss_ce: 0.0057 decode.acc_seg: 99.7826 aux.loss_ce: 0.0073 aux.acc_seg: 99.1121 +04/19 02:36:52 - mmengine - INFO - Iter(train) [ 65650/160000] base_lr: 5.9527e-05 lr: 2.2008e-07 eta: 1 day, 5:59:34 time: 0.9988 data_time: 0.0048 memory: 8457 loss: 0.0110 decode.loss_ce: 0.0050 decode.acc_seg: 99.7784 aux.loss_ce: 0.0060 aux.acc_seg: 99.5205 +04/19 02:37:42 - mmengine - INFO - Iter(train) [ 65700/160000] base_lr: 5.9496e-05 lr: 2.1997e-07 eta: 1 day, 5:56:37 time: 1.0004 data_time: 0.0047 memory: 8457 loss: 0.0142 decode.loss_ce: 0.0063 decode.acc_seg: 99.6353 aux.loss_ce: 0.0080 aux.acc_seg: 98.9258 +04/19 02:38:32 - mmengine - INFO - Iter(train) [ 65750/160000] base_lr: 5.9464e-05 lr: 2.1985e-07 eta: 1 day, 5:53:42 time: 0.9989 data_time: 0.0045 memory: 8457 loss: 0.0135 decode.loss_ce: 0.0060 decode.acc_seg: 99.7807 aux.loss_ce: 0.0075 aux.acc_seg: 99.2249 +04/19 02:39:22 - mmengine - INFO - Iter(train) [ 65800/160000] base_lr: 5.9433e-05 lr: 2.1973e-07 eta: 1 day, 5:50:50 time: 0.9999 data_time: 0.0051 memory: 8457 loss: 0.0131 decode.loss_ce: 0.0057 decode.acc_seg: 99.7437 aux.loss_ce: 0.0074 aux.acc_seg: 99.3151 +04/19 02:40:12 - mmengine - INFO - Iter(train) [ 65850/160000] base_lr: 5.9401e-05 lr: 2.1962e-07 eta: 1 day, 5:47:59 time: 1.0005 data_time: 0.0052 memory: 8457 loss: 0.0145 decode.loss_ce: 0.0064 decode.acc_seg: 99.7002 aux.loss_ce: 0.0080 aux.acc_seg: 99.0841 +04/19 02:41:02 - mmengine - INFO - Iter(train) [ 65900/160000] base_lr: 5.9369e-05 lr: 2.1950e-07 eta: 1 day, 5:45:11 time: 0.9999 data_time: 0.0043 memory: 8457 loss: 0.0155 decode.loss_ce: 0.0071 decode.acc_seg: 99.7105 aux.loss_ce: 0.0084 aux.acc_seg: 98.7299 +04/19 02:41:52 - mmengine - INFO - Iter(train) [ 65950/160000] base_lr: 5.9338e-05 lr: 2.1938e-07 eta: 1 day, 5:42:24 time: 0.9989 data_time: 0.0045 memory: 8457 loss: 0.0139 decode.loss_ce: 0.0067 decode.acc_seg: 99.7787 aux.loss_ce: 0.0073 aux.acc_seg: 99.1877 +04/19 02:42:42 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240419_004857 +04/19 02:42:42 - mmengine - INFO - Iter(train) [ 66000/160000] base_lr: 5.9306e-05 lr: 2.1927e-07 eta: 1 day, 5:39:40 time: 0.9984 data_time: 0.0042 memory: 8457 loss: 0.0155 decode.loss_ce: 0.0070 decode.acc_seg: 99.7824 aux.loss_ce: 0.0085 aux.acc_seg: 99.2033 +04/19 02:43:32 - mmengine - INFO - Iter(train) [ 66050/160000] base_lr: 5.9275e-05 lr: 2.1915e-07 eta: 1 day, 5:36:58 time: 1.0001 data_time: 0.0046 memory: 8457 loss: 0.0147 decode.loss_ce: 0.0067 decode.acc_seg: 99.6286 aux.loss_ce: 0.0080 aux.acc_seg: 98.9256 +04/19 02:44:22 - mmengine - INFO - Iter(train) [ 66100/160000] base_lr: 5.9243e-05 lr: 2.1903e-07 eta: 1 day, 5:34:17 time: 1.0003 data_time: 0.0044 memory: 8457 loss: 0.0135 decode.loss_ce: 0.0061 decode.acc_seg: 99.7688 aux.loss_ce: 0.0074 aux.acc_seg: 99.0057 +04/19 02:45:12 - mmengine - INFO - Iter(train) [ 66150/160000] base_lr: 5.9212e-05 lr: 2.1892e-07 eta: 1 day, 5:31:39 time: 1.0015 data_time: 0.0043 memory: 8457 loss: 0.0143 decode.loss_ce: 0.0062 decode.acc_seg: 99.6891 aux.loss_ce: 0.0081 aux.acc_seg: 98.8300 +04/19 02:46:02 - mmengine - INFO - Iter(train) [ 66200/160000] base_lr: 5.9180e-05 lr: 2.1880e-07 eta: 1 day, 5:29:02 time: 1.0000 data_time: 0.0047 memory: 8457 loss: 0.0124 decode.loss_ce: 0.0056 decode.acc_seg: 99.8066 aux.loss_ce: 0.0067 aux.acc_seg: 99.4806 +04/19 02:46:52 - mmengine - INFO - Iter(train) [ 66250/160000] base_lr: 5.9149e-05 lr: 2.1868e-07 eta: 1 day, 5:26:26 time: 0.9992 data_time: 0.0045 memory: 8457 loss: 0.0153 decode.loss_ce: 0.0071 decode.acc_seg: 99.7150 aux.loss_ce: 0.0081 aux.acc_seg: 99.1611 +04/19 02:47:42 - mmengine - INFO - Iter(train) [ 66300/160000] base_lr: 5.9117e-05 lr: 2.1857e-07 eta: 1 day, 5:23:52 time: 1.0002 data_time: 0.0047 memory: 8457 loss: 0.0141 decode.loss_ce: 0.0063 decode.acc_seg: 99.7278 aux.loss_ce: 0.0078 aux.acc_seg: 99.1610 +04/19 02:48:32 - mmengine - INFO - Iter(train) [ 66350/160000] base_lr: 5.9086e-05 lr: 2.1845e-07 eta: 1 day, 5:21:20 time: 0.9987 data_time: 0.0044 memory: 8457 loss: 0.0110 decode.loss_ce: 0.0052 decode.acc_seg: 99.8354 aux.loss_ce: 0.0058 aux.acc_seg: 99.5626 +04/19 02:49:22 - mmengine - INFO - Iter(train) [ 66400/160000] base_lr: 5.9054e-05 lr: 2.1833e-07 eta: 1 day, 5:18:49 time: 0.9990 data_time: 0.0042 memory: 8457 loss: 0.0130 decode.loss_ce: 0.0061 decode.acc_seg: 99.8220 aux.loss_ce: 0.0069 aux.acc_seg: 99.3271 +04/19 02:50:12 - mmengine - INFO - Iter(train) [ 66450/160000] base_lr: 5.9022e-05 lr: 2.1822e-07 eta: 1 day, 5:16:19 time: 0.9993 data_time: 0.0046 memory: 8457 loss: 0.0134 decode.loss_ce: 0.0061 decode.acc_seg: 99.7028 aux.loss_ce: 0.0073 aux.acc_seg: 99.0662 +04/19 02:51:02 - mmengine - INFO - Iter(train) [ 66500/160000] base_lr: 5.8991e-05 lr: 2.1810e-07 eta: 1 day, 5:13:52 time: 0.9996 data_time: 0.0046 memory: 8457 loss: 0.0160 decode.loss_ce: 0.0074 decode.acc_seg: 99.6134 aux.loss_ce: 0.0086 aux.acc_seg: 98.6597 +04/19 02:51:52 - mmengine - INFO - Iter(train) [ 66550/160000] base_lr: 5.8959e-05 lr: 2.1798e-07 eta: 1 day, 5:11:25 time: 0.9987 data_time: 0.0048 memory: 8457 loss: 0.0124 decode.loss_ce: 0.0057 decode.acc_seg: 99.7723 aux.loss_ce: 0.0067 aux.acc_seg: 99.4894 +04/19 02:52:42 - mmengine - INFO - Iter(train) [ 66600/160000] base_lr: 5.8928e-05 lr: 2.1787e-07 eta: 1 day, 5:09:01 time: 0.9987 data_time: 0.0044 memory: 8457 loss: 0.0144 decode.loss_ce: 0.0067 decode.acc_seg: 99.5272 aux.loss_ce: 0.0077 aux.acc_seg: 98.9498 +04/19 02:53:32 - mmengine - INFO - Iter(train) [ 66650/160000] base_lr: 5.8896e-05 lr: 2.1775e-07 eta: 1 day, 5:06:38 time: 0.9995 data_time: 0.0049 memory: 8457 loss: 0.0132 decode.loss_ce: 0.0058 decode.acc_seg: 99.7869 aux.loss_ce: 0.0074 aux.acc_seg: 99.3690 +04/19 02:54:22 - mmengine - INFO - Iter(train) [ 66700/160000] base_lr: 5.8865e-05 lr: 2.1764e-07 eta: 1 day, 5:04:17 time: 0.9997 data_time: 0.0046 memory: 8457 loss: 0.0143 decode.loss_ce: 0.0064 decode.acc_seg: 99.7763 aux.loss_ce: 0.0080 aux.acc_seg: 99.3626 +04/19 02:55:12 - mmengine - INFO - Iter(train) [ 66750/160000] base_lr: 5.8833e-05 lr: 2.1752e-07 eta: 1 day, 5:01:57 time: 1.0009 data_time: 0.0045 memory: 8457 loss: 0.0141 decode.loss_ce: 0.0061 decode.acc_seg: 99.7335 aux.loss_ce: 0.0080 aux.acc_seg: 99.1220 +04/19 02:56:02 - mmengine - INFO - Iter(train) [ 66800/160000] base_lr: 5.8802e-05 lr: 2.1740e-07 eta: 1 day, 4:59:38 time: 0.9996 data_time: 0.0049 memory: 8457 loss: 0.0137 decode.loss_ce: 0.0062 decode.acc_seg: 99.6586 aux.loss_ce: 0.0075 aux.acc_seg: 98.9151 +04/19 02:56:52 - mmengine - INFO - Iter(train) [ 66850/160000] base_lr: 5.8770e-05 lr: 2.1729e-07 eta: 1 day, 4:57:21 time: 0.9998 data_time: 0.0045 memory: 8457 loss: 0.0126 decode.loss_ce: 0.0056 decode.acc_seg: 99.7807 aux.loss_ce: 0.0070 aux.acc_seg: 99.2561 +04/19 02:57:42 - mmengine - INFO - Iter(train) [ 66900/160000] base_lr: 5.8739e-05 lr: 2.1717e-07 eta: 1 day, 4:55:04 time: 1.0002 data_time: 0.0046 memory: 8457 loss: 0.0158 decode.loss_ce: 0.0075 decode.acc_seg: 99.8150 aux.loss_ce: 0.0083 aux.acc_seg: 99.2662 +04/19 02:58:32 - mmengine - INFO - Iter(train) [ 66950/160000] base_lr: 5.8707e-05 lr: 2.1705e-07 eta: 1 day, 4:52:50 time: 0.9996 data_time: 0.0049 memory: 8457 loss: 0.0109 decode.loss_ce: 0.0050 decode.acc_seg: 99.7541 aux.loss_ce: 0.0058 aux.acc_seg: 99.3149 +04/19 02:59:22 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240419_004857 +04/19 02:59:22 - mmengine - INFO - Iter(train) [ 67000/160000] base_lr: 5.8675e-05 lr: 2.1694e-07 eta: 1 day, 4:50:36 time: 1.0016 data_time: 0.0050 memory: 8457 loss: 0.0149 decode.loss_ce: 0.0066 decode.acc_seg: 99.7412 aux.loss_ce: 0.0083 aux.acc_seg: 99.3046 +04/19 03:00:12 - mmengine - INFO - Iter(train) [ 67050/160000] base_lr: 5.8644e-05 lr: 2.1682e-07 eta: 1 day, 4:48:23 time: 1.0002 data_time: 0.0048 memory: 8457 loss: 0.0147 decode.loss_ce: 0.0064 decode.acc_seg: 99.8278 aux.loss_ce: 0.0083 aux.acc_seg: 99.4446 +04/19 03:01:02 - mmengine - INFO - Iter(train) [ 67100/160000] base_lr: 5.8612e-05 lr: 2.1670e-07 eta: 1 day, 4:46:12 time: 0.9996 data_time: 0.0050 memory: 8457 loss: 0.0127 decode.loss_ce: 0.0055 decode.acc_seg: 99.8041 aux.loss_ce: 0.0072 aux.acc_seg: 99.2178 +04/19 03:01:52 - mmengine - INFO - Iter(train) [ 67150/160000] base_lr: 5.8581e-05 lr: 2.1659e-07 eta: 1 day, 4:44:01 time: 1.0000 data_time: 0.0048 memory: 8457 loss: 0.0125 decode.loss_ce: 0.0057 decode.acc_seg: 99.7541 aux.loss_ce: 0.0068 aux.acc_seg: 99.2014 +04/19 03:02:42 - mmengine - INFO - Iter(train) [ 67200/160000] base_lr: 5.8549e-05 lr: 2.1647e-07 eta: 1 day, 4:41:52 time: 0.9990 data_time: 0.0045 memory: 8457 loss: 0.0138 decode.loss_ce: 0.0062 decode.acc_seg: 99.6986 aux.loss_ce: 0.0076 aux.acc_seg: 99.0858 +04/19 03:03:32 - mmengine - INFO - Iter(train) [ 67250/160000] base_lr: 5.8518e-05 lr: 2.1635e-07 eta: 1 day, 4:39:44 time: 1.0010 data_time: 0.0044 memory: 8457 loss: 0.0129 decode.loss_ce: 0.0058 decode.acc_seg: 99.7366 aux.loss_ce: 0.0070 aux.acc_seg: 99.3134 +04/19 03:04:22 - mmengine - INFO - Iter(train) [ 67300/160000] base_lr: 5.8486e-05 lr: 2.1624e-07 eta: 1 day, 4:37:37 time: 1.0003 data_time: 0.0044 memory: 8457 loss: 0.0145 decode.loss_ce: 0.0067 decode.acc_seg: 99.6817 aux.loss_ce: 0.0078 aux.acc_seg: 99.2720 +04/19 03:05:12 - mmengine - INFO - Iter(train) [ 67350/160000] base_lr: 5.8455e-05 lr: 2.1612e-07 eta: 1 day, 4:35:31 time: 1.0010 data_time: 0.0049 memory: 8457 loss: 0.0138 decode.loss_ce: 0.0061 decode.acc_seg: 99.7419 aux.loss_ce: 0.0077 aux.acc_seg: 99.3265 +04/19 03:06:02 - mmengine - INFO - Iter(train) [ 67400/160000] base_lr: 5.8423e-05 lr: 2.1600e-07 eta: 1 day, 4:33:26 time: 0.9983 data_time: 0.0043 memory: 8457 loss: 0.0129 decode.loss_ce: 0.0057 decode.acc_seg: 99.7553 aux.loss_ce: 0.0072 aux.acc_seg: 99.2546 +04/19 03:06:52 - mmengine - INFO - Iter(train) [ 67450/160000] base_lr: 5.8392e-05 lr: 2.1589e-07 eta: 1 day, 4:31:22 time: 1.0013 data_time: 0.0047 memory: 8457 loss: 0.0127 decode.loss_ce: 0.0059 decode.acc_seg: 99.8526 aux.loss_ce: 0.0067 aux.acc_seg: 99.5403 +04/19 03:07:42 - mmengine - INFO - Iter(train) [ 67500/160000] base_lr: 5.8360e-05 lr: 2.1577e-07 eta: 1 day, 4:29:19 time: 0.9999 data_time: 0.0049 memory: 8457 loss: 0.0141 decode.loss_ce: 0.0063 decode.acc_seg: 99.7936 aux.loss_ce: 0.0078 aux.acc_seg: 99.2247 +04/19 03:08:32 - mmengine - INFO - Iter(train) [ 67550/160000] base_lr: 5.8328e-05 lr: 2.1565e-07 eta: 1 day, 4:27:17 time: 0.9990 data_time: 0.0042 memory: 8457 loss: 0.0122 decode.loss_ce: 0.0055 decode.acc_seg: 99.7091 aux.loss_ce: 0.0067 aux.acc_seg: 99.1322 +04/19 03:09:22 - mmengine - INFO - Iter(train) [ 67600/160000] base_lr: 5.8297e-05 lr: 2.1554e-07 eta: 1 day, 4:25:16 time: 1.0010 data_time: 0.0051 memory: 8457 loss: 0.0147 decode.loss_ce: 0.0066 decode.acc_seg: 99.6870 aux.loss_ce: 0.0081 aux.acc_seg: 98.9845 +04/19 03:10:12 - mmengine - INFO - Iter(train) [ 67650/160000] base_lr: 5.8265e-05 lr: 2.1542e-07 eta: 1 day, 4:23:16 time: 0.9992 data_time: 0.0044 memory: 8457 loss: 0.0142 decode.loss_ce: 0.0062 decode.acc_seg: 99.6229 aux.loss_ce: 0.0080 aux.acc_seg: 99.0772 +04/19 03:11:02 - mmengine - INFO - Iter(train) [ 67700/160000] base_lr: 5.8234e-05 lr: 2.1530e-07 eta: 1 day, 4:21:16 time: 0.9984 data_time: 0.0047 memory: 8457 loss: 0.0130 decode.loss_ce: 0.0059 decode.acc_seg: 99.8013 aux.loss_ce: 0.0071 aux.acc_seg: 99.2416 +04/19 03:11:52 - mmengine - INFO - Iter(train) [ 67750/160000] base_lr: 5.8202e-05 lr: 2.1519e-07 eta: 1 day, 4:19:18 time: 0.9991 data_time: 0.0044 memory: 8457 loss: 0.0134 decode.loss_ce: 0.0059 decode.acc_seg: 99.6426 aux.loss_ce: 0.0075 aux.acc_seg: 99.1014 +04/19 03:12:42 - mmengine - INFO - Iter(train) [ 67800/160000] base_lr: 5.8171e-05 lr: 2.1507e-07 eta: 1 day, 4:17:20 time: 0.9993 data_time: 0.0050 memory: 8457 loss: 0.0123 decode.loss_ce: 0.0059 decode.acc_seg: 99.7988 aux.loss_ce: 0.0064 aux.acc_seg: 99.4812 +04/19 03:13:32 - mmengine - INFO - Iter(train) [ 67850/160000] base_lr: 5.8139e-05 lr: 2.1495e-07 eta: 1 day, 4:15:24 time: 0.9993 data_time: 0.0048 memory: 8457 loss: 0.0130 decode.loss_ce: 0.0061 decode.acc_seg: 99.7873 aux.loss_ce: 0.0069 aux.acc_seg: 99.4246 +04/19 03:14:22 - mmengine - INFO - Iter(train) [ 67900/160000] base_lr: 5.8108e-05 lr: 2.1484e-07 eta: 1 day, 4:13:28 time: 0.9989 data_time: 0.0044 memory: 8457 loss: 0.0144 decode.loss_ce: 0.0066 decode.acc_seg: 99.6967 aux.loss_ce: 0.0079 aux.acc_seg: 98.7911 +04/19 03:15:12 - mmengine - INFO - Iter(train) [ 67950/160000] base_lr: 5.8076e-05 lr: 2.1472e-07 eta: 1 day, 4:11:33 time: 1.0002 data_time: 0.0044 memory: 8457 loss: 0.0130 decode.loss_ce: 0.0059 decode.acc_seg: 99.8369 aux.loss_ce: 0.0071 aux.acc_seg: 99.5193 +04/19 03:16:02 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240419_004857 +04/19 03:16:02 - mmengine - INFO - Iter(train) [ 68000/160000] base_lr: 5.8045e-05 lr: 2.1460e-07 eta: 1 day, 4:09:39 time: 0.9999 data_time: 0.0047 memory: 8457 loss: 0.0141 decode.loss_ce: 0.0063 decode.acc_seg: 99.7166 aux.loss_ce: 0.0078 aux.acc_seg: 99.2054 +04/19 03:16:52 - mmengine - INFO - Iter(train) [ 68050/160000] base_lr: 5.8013e-05 lr: 2.1449e-07 eta: 1 day, 4:07:46 time: 1.0013 data_time: 0.0046 memory: 8457 loss: 0.0145 decode.loss_ce: 0.0063 decode.acc_seg: 99.6471 aux.loss_ce: 0.0081 aux.acc_seg: 98.9067 +04/19 03:17:42 - mmengine - INFO - Iter(train) [ 68100/160000] base_lr: 5.7981e-05 lr: 2.1437e-07 eta: 1 day, 4:05:54 time: 1.0010 data_time: 0.0044 memory: 8457 loss: 0.0147 decode.loss_ce: 0.0064 decode.acc_seg: 99.7122 aux.loss_ce: 0.0083 aux.acc_seg: 98.9893 +04/19 03:18:32 - mmengine - INFO - Iter(train) [ 68150/160000] base_lr: 5.7950e-05 lr: 2.1425e-07 eta: 1 day, 4:04:02 time: 0.9987 data_time: 0.0043 memory: 8457 loss: 0.0139 decode.loss_ce: 0.0059 decode.acc_seg: 99.7580 aux.loss_ce: 0.0079 aux.acc_seg: 99.1570 +04/19 03:19:22 - mmengine - INFO - Iter(train) [ 68200/160000] base_lr: 5.7918e-05 lr: 2.1414e-07 eta: 1 day, 4:02:11 time: 0.9984 data_time: 0.0043 memory: 8457 loss: 0.0149 decode.loss_ce: 0.0064 decode.acc_seg: 99.7087 aux.loss_ce: 0.0085 aux.acc_seg: 98.9267 +04/19 03:20:12 - mmengine - INFO - Iter(train) [ 68250/160000] base_lr: 5.7887e-05 lr: 2.1402e-07 eta: 1 day, 4:00:20 time: 1.0004 data_time: 0.0045 memory: 8457 loss: 0.0136 decode.loss_ce: 0.0061 decode.acc_seg: 99.6906 aux.loss_ce: 0.0075 aux.acc_seg: 99.1526 +04/19 03:21:02 - mmengine - INFO - Iter(train) [ 68300/160000] base_lr: 5.7855e-05 lr: 2.1390e-07 eta: 1 day, 3:58:31 time: 1.0000 data_time: 0.0043 memory: 8457 loss: 0.0130 decode.loss_ce: 0.0057 decode.acc_seg: 99.6550 aux.loss_ce: 0.0073 aux.acc_seg: 98.7543 +04/19 03:21:52 - mmengine - INFO - Iter(train) [ 68350/160000] base_lr: 5.7824e-05 lr: 2.1379e-07 eta: 1 day, 3:56:42 time: 1.0000 data_time: 0.0047 memory: 8457 loss: 0.0134 decode.loss_ce: 0.0060 decode.acc_seg: 99.8215 aux.loss_ce: 0.0074 aux.acc_seg: 99.4900 +04/19 03:22:42 - mmengine - INFO - Iter(train) [ 68400/160000] base_lr: 5.7792e-05 lr: 2.1367e-07 eta: 1 day, 3:54:53 time: 0.9983 data_time: 0.0047 memory: 8457 loss: 0.0153 decode.loss_ce: 0.0067 decode.acc_seg: 99.6662 aux.loss_ce: 0.0086 aux.acc_seg: 98.7791 +04/19 03:23:32 - mmengine - INFO - Iter(train) [ 68450/160000] base_lr: 5.7761e-05 lr: 2.1355e-07 eta: 1 day, 3:53:05 time: 1.0010 data_time: 0.0047 memory: 8457 loss: 0.0138 decode.loss_ce: 0.0062 decode.acc_seg: 99.8186 aux.loss_ce: 0.0076 aux.acc_seg: 99.4602 +04/19 03:24:22 - mmengine - INFO - Iter(train) [ 68500/160000] base_lr: 5.7729e-05 lr: 2.1344e-07 eta: 1 day, 3:51:18 time: 1.0002 data_time: 0.0050 memory: 8457 loss: 0.0127 decode.loss_ce: 0.0055 decode.acc_seg: 99.7313 aux.loss_ce: 0.0072 aux.acc_seg: 98.9658 +04/19 03:25:12 - mmengine - INFO - Iter(train) [ 68550/160000] base_lr: 5.7698e-05 lr: 2.1332e-07 eta: 1 day, 3:49:32 time: 0.9988 data_time: 0.0047 memory: 8457 loss: 0.0133 decode.loss_ce: 0.0062 decode.acc_seg: 99.7478 aux.loss_ce: 0.0071 aux.acc_seg: 99.2861 +04/19 03:26:02 - mmengine - INFO - Iter(train) [ 68600/160000] base_lr: 5.7666e-05 lr: 2.1320e-07 eta: 1 day, 3:47:46 time: 1.0024 data_time: 0.0043 memory: 8457 loss: 0.0126 decode.loss_ce: 0.0057 decode.acc_seg: 99.6540 aux.loss_ce: 0.0068 aux.acc_seg: 99.2451 +04/19 03:26:52 - mmengine - INFO - Iter(train) [ 68650/160000] base_lr: 5.7634e-05 lr: 2.1309e-07 eta: 1 day, 3:46:01 time: 0.9986 data_time: 0.0047 memory: 8457 loss: 0.0120 decode.loss_ce: 0.0054 decode.acc_seg: 99.6864 aux.loss_ce: 0.0066 aux.acc_seg: 99.1941 +04/19 03:27:42 - mmengine - INFO - Iter(train) [ 68700/160000] base_lr: 5.7603e-05 lr: 2.1297e-07 eta: 1 day, 3:44:17 time: 0.9990 data_time: 0.0047 memory: 8457 loss: 0.0128 decode.loss_ce: 0.0056 decode.acc_seg: 99.7318 aux.loss_ce: 0.0072 aux.acc_seg: 99.0759 +04/19 03:28:32 - mmengine - INFO - Iter(train) [ 68750/160000] base_lr: 5.7571e-05 lr: 2.1285e-07 eta: 1 day, 3:42:33 time: 0.9991 data_time: 0.0043 memory: 8457 loss: 0.0126 decode.loss_ce: 0.0056 decode.acc_seg: 99.7000 aux.loss_ce: 0.0071 aux.acc_seg: 99.3675 +04/19 03:29:22 - mmengine - INFO - Iter(train) [ 68800/160000] base_lr: 5.7540e-05 lr: 2.1274e-07 eta: 1 day, 3:40:50 time: 1.0005 data_time: 0.0049 memory: 8457 loss: 0.0108 decode.loss_ce: 0.0049 decode.acc_seg: 99.7734 aux.loss_ce: 0.0059 aux.acc_seg: 99.4505 +04/19 03:30:12 - mmengine - INFO - Iter(train) [ 68850/160000] base_lr: 5.7508e-05 lr: 2.1262e-07 eta: 1 day, 3:39:08 time: 1.0019 data_time: 0.0044 memory: 8457 loss: 0.0150 decode.loss_ce: 0.0064 decode.acc_seg: 99.7847 aux.loss_ce: 0.0085 aux.acc_seg: 99.3843 +04/19 03:31:02 - mmengine - INFO - Iter(train) [ 68900/160000] base_lr: 5.7477e-05 lr: 2.1250e-07 eta: 1 day, 3:37:26 time: 0.9999 data_time: 0.0045 memory: 8457 loss: 0.0138 decode.loss_ce: 0.0061 decode.acc_seg: 99.7026 aux.loss_ce: 0.0077 aux.acc_seg: 98.9614 +04/19 03:31:52 - mmengine - INFO - Iter(train) [ 68950/160000] base_lr: 5.7445e-05 lr: 2.1239e-07 eta: 1 day, 3:35:45 time: 1.0016 data_time: 0.0043 memory: 8457 loss: 0.0198 decode.loss_ce: 0.0106 decode.acc_seg: 99.7808 aux.loss_ce: 0.0092 aux.acc_seg: 99.3315 +04/19 03:32:42 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240419_004857 +04/19 03:32:42 - mmengine - INFO - Iter(train) [ 69000/160000] base_lr: 5.7414e-05 lr: 2.1227e-07 eta: 1 day, 3:34:05 time: 1.0001 data_time: 0.0046 memory: 8457 loss: 0.0139 decode.loss_ce: 0.0062 decode.acc_seg: 99.7728 aux.loss_ce: 0.0077 aux.acc_seg: 99.4513 +04/19 03:33:32 - mmengine - INFO - Iter(train) [ 69050/160000] base_lr: 5.7382e-05 lr: 2.1215e-07 eta: 1 day, 3:32:25 time: 1.0010 data_time: 0.0044 memory: 8457 loss: 0.0149 decode.loss_ce: 0.0066 decode.acc_seg: 99.7217 aux.loss_ce: 0.0083 aux.acc_seg: 99.1955 +04/19 03:34:22 - mmengine - INFO - Iter(train) [ 69100/160000] base_lr: 5.7351e-05 lr: 2.1204e-07 eta: 1 day, 3:30:45 time: 0.9997 data_time: 0.0047 memory: 8457 loss: 0.0134 decode.loss_ce: 0.0060 decode.acc_seg: 99.7412 aux.loss_ce: 0.0075 aux.acc_seg: 99.0776 +04/19 03:35:12 - mmengine - INFO - Iter(train) [ 69150/160000] base_lr: 5.7319e-05 lr: 2.1192e-07 eta: 1 day, 3:29:06 time: 0.9995 data_time: 0.0046 memory: 8457 loss: 0.0136 decode.loss_ce: 0.0063 decode.acc_seg: 99.7248 aux.loss_ce: 0.0073 aux.acc_seg: 99.2785 +04/19 03:36:02 - mmengine - INFO - Iter(train) [ 69200/160000] base_lr: 5.7287e-05 lr: 2.1180e-07 eta: 1 day, 3:27:27 time: 0.9994 data_time: 0.0044 memory: 8457 loss: 0.0119 decode.loss_ce: 0.0052 decode.acc_seg: 99.8348 aux.loss_ce: 0.0067 aux.acc_seg: 99.4093 +04/19 03:36:52 - mmengine - INFO - Iter(train) [ 69250/160000] base_lr: 5.7256e-05 lr: 2.1169e-07 eta: 1 day, 3:25:49 time: 1.0005 data_time: 0.0049 memory: 8457 loss: 0.0140 decode.loss_ce: 0.0066 decode.acc_seg: 99.7133 aux.loss_ce: 0.0074 aux.acc_seg: 99.3174 +04/19 03:37:42 - mmengine - INFO - Iter(train) [ 69300/160000] base_lr: 5.7224e-05 lr: 2.1157e-07 eta: 1 day, 3:24:12 time: 1.0003 data_time: 0.0046 memory: 8457 loss: 0.0153 decode.loss_ce: 0.0068 decode.acc_seg: 99.5737 aux.loss_ce: 0.0084 aux.acc_seg: 98.8388 +04/19 03:38:32 - mmengine - INFO - Iter(train) [ 69350/160000] base_lr: 5.7193e-05 lr: 2.1145e-07 eta: 1 day, 3:22:34 time: 0.9991 data_time: 0.0046 memory: 8457 loss: 0.0139 decode.loss_ce: 0.0063 decode.acc_seg: 99.5775 aux.loss_ce: 0.0076 aux.acc_seg: 98.9248 +04/19 03:39:22 - mmengine - INFO - Iter(train) [ 69400/160000] base_lr: 5.7161e-05 lr: 2.1134e-07 eta: 1 day, 3:20:57 time: 0.9998 data_time: 0.0048 memory: 8457 loss: 0.0133 decode.loss_ce: 0.0058 decode.acc_seg: 99.7053 aux.loss_ce: 0.0075 aux.acc_seg: 98.9986 +04/19 03:40:12 - mmengine - INFO - Iter(train) [ 69450/160000] base_lr: 5.7130e-05 lr: 2.1122e-07 eta: 1 day, 3:19:21 time: 0.9990 data_time: 0.0045 memory: 8457 loss: 0.0125 decode.loss_ce: 0.0057 decode.acc_seg: 99.7320 aux.loss_ce: 0.0067 aux.acc_seg: 99.3029 +04/19 03:41:02 - mmengine - INFO - Iter(train) [ 69500/160000] base_lr: 5.7098e-05 lr: 2.1110e-07 eta: 1 day, 3:17:45 time: 0.9993 data_time: 0.0045 memory: 8457 loss: 0.0139 decode.loss_ce: 0.0066 decode.acc_seg: 99.7351 aux.loss_ce: 0.0073 aux.acc_seg: 99.2998 +04/19 03:41:52 - mmengine - INFO - Iter(train) [ 69550/160000] base_lr: 5.7067e-05 lr: 2.1099e-07 eta: 1 day, 3:16:10 time: 0.9989 data_time: 0.0048 memory: 8457 loss: 0.0128 decode.loss_ce: 0.0058 decode.acc_seg: 99.8100 aux.loss_ce: 0.0071 aux.acc_seg: 99.4490 +04/19 03:42:42 - mmengine - INFO - Iter(train) [ 69600/160000] base_lr: 5.7035e-05 lr: 2.1087e-07 eta: 1 day, 3:14:35 time: 0.9978 data_time: 0.0046 memory: 8457 loss: 0.0134 decode.loss_ce: 0.0060 decode.acc_seg: 99.7519 aux.loss_ce: 0.0074 aux.acc_seg: 99.4026 +04/19 03:43:32 - mmengine - INFO - Iter(train) [ 69650/160000] base_lr: 5.7004e-05 lr: 2.1075e-07 eta: 1 day, 3:13:00 time: 0.9979 data_time: 0.0044 memory: 8457 loss: 0.0124 decode.loss_ce: 0.0055 decode.acc_seg: 99.7826 aux.loss_ce: 0.0069 aux.acc_seg: 99.2519 +04/19 03:44:22 - mmengine - INFO - Iter(train) [ 69700/160000] base_lr: 5.6972e-05 lr: 2.1064e-07 eta: 1 day, 3:11:25 time: 0.9978 data_time: 0.0045 memory: 8457 loss: 0.0121 decode.loss_ce: 0.0054 decode.acc_seg: 99.7658 aux.loss_ce: 0.0066 aux.acc_seg: 99.1362 +04/19 03:45:12 - mmengine - INFO - Iter(train) [ 69750/160000] base_lr: 5.6940e-05 lr: 2.1052e-07 eta: 1 day, 3:09:52 time: 0.9974 data_time: 0.0046 memory: 8457 loss: 0.0111 decode.loss_ce: 0.0051 decode.acc_seg: 99.7854 aux.loss_ce: 0.0060 aux.acc_seg: 99.4383 +04/19 03:46:02 - mmengine - INFO - Iter(train) [ 69800/160000] base_lr: 5.6909e-05 lr: 2.1040e-07 eta: 1 day, 3:08:18 time: 1.0000 data_time: 0.0047 memory: 8457 loss: 0.0139 decode.loss_ce: 0.0065 decode.acc_seg: 99.7082 aux.loss_ce: 0.0073 aux.acc_seg: 99.2994 +04/19 03:46:51 - mmengine - INFO - Iter(train) [ 69850/160000] base_lr: 5.6877e-05 lr: 2.1029e-07 eta: 1 day, 3:06:45 time: 0.9984 data_time: 0.0047 memory: 8457 loss: 0.0141 decode.loss_ce: 0.0063 decode.acc_seg: 99.6773 aux.loss_ce: 0.0078 aux.acc_seg: 99.0072 +04/19 03:47:41 - mmengine - INFO - Iter(train) [ 69900/160000] base_lr: 5.6846e-05 lr: 2.1017e-07 eta: 1 day, 3:05:12 time: 0.9976 data_time: 0.0046 memory: 8457 loss: 0.0146 decode.loss_ce: 0.0063 decode.acc_seg: 99.7482 aux.loss_ce: 0.0083 aux.acc_seg: 99.1983 +04/19 03:48:31 - mmengine - INFO - Iter(train) [ 69950/160000] base_lr: 5.6814e-05 lr: 2.1005e-07 eta: 1 day, 3:03:40 time: 0.9979 data_time: 0.0051 memory: 8457 loss: 0.0130 decode.loss_ce: 0.0059 decode.acc_seg: 99.7807 aux.loss_ce: 0.0072 aux.acc_seg: 99.3937 +04/19 03:49:21 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240419_004857 +04/19 03:49:21 - mmengine - INFO - Iter(train) [ 70000/160000] base_lr: 5.6783e-05 lr: 2.0994e-07 eta: 1 day, 3:02:07 time: 0.9969 data_time: 0.0045 memory: 8457 loss: 0.0134 decode.loss_ce: 0.0059 decode.acc_seg: 99.7444 aux.loss_ce: 0.0075 aux.acc_seg: 99.2407 +04/19 03:49:21 - mmengine - INFO - Saving checkpoint at 70000 iterations +04/19 03:49:32 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:21 time: 0.1162 data_time: 0.0015 memory: 7120 +04/19 03:49:38 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:12 time: 0.1160 data_time: 0.0015 memory: 3999 +04/19 03:49:44 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:06 time: 0.1168 data_time: 0.0015 memory: 3999 +04/19 03:49:50 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.1156 data_time: 0.0013 memory: 3999 +04/19 03:49:50 - mmengine - INFO - per class results: +04/19 03:49:50 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 99.17 | 99.56 | 99.58 | 99.61 | 99.56 | +| contrast | 81.94 | 90.71 | 90.07 | 89.45 | 90.71 | ++------------+-------+-------+--------+-----------+--------+ +04/19 03:49:50 - mmengine - INFO - Iter(val) [200/200] aAcc: 99.2000 mIoU: 90.5600 mAcc: 95.1300 mFscore: 94.8300 mPrecision: 94.5300 mRecall: 95.1300 data_time: 0.0024 time: 0.1227 +04/19 03:50:40 - mmengine - INFO - Iter(train) [ 70050/160000] base_lr: 5.6751e-05 lr: 2.0982e-07 eta: 1 day, 3:00:37 time: 0.9962 data_time: 0.0044 memory: 8457 loss: 0.0136 decode.loss_ce: 0.0066 decode.acc_seg: 99.6668 aux.loss_ce: 0.0070 aux.acc_seg: 99.2702 +04/19 03:51:30 - mmengine - INFO - Iter(train) [ 70100/160000] base_lr: 5.6720e-05 lr: 2.0970e-07 eta: 1 day, 2:59:06 time: 0.9969 data_time: 0.0048 memory: 8457 loss: 0.0131 decode.loss_ce: 0.0059 decode.acc_seg: 99.7742 aux.loss_ce: 0.0071 aux.acc_seg: 99.4482 +04/19 03:52:20 - mmengine - INFO - Iter(train) [ 70150/160000] base_lr: 5.6688e-05 lr: 2.0959e-07 eta: 1 day, 2:57:35 time: 0.9974 data_time: 0.0047 memory: 8457 loss: 0.0112 decode.loss_ce: 0.0050 decode.acc_seg: 99.7908 aux.loss_ce: 0.0062 aux.acc_seg: 99.1558 +04/19 03:53:10 - mmengine - INFO - Iter(train) [ 70200/160000] base_lr: 5.6657e-05 lr: 2.0947e-07 eta: 1 day, 2:56:05 time: 0.9973 data_time: 0.0044 memory: 8457 loss: 0.0132 decode.loss_ce: 0.0060 decode.acc_seg: 99.7467 aux.loss_ce: 0.0072 aux.acc_seg: 99.3105 +04/19 03:53:59 - mmengine - INFO - Iter(train) [ 70250/160000] base_lr: 5.6625e-05 lr: 2.0935e-07 eta: 1 day, 2:54:35 time: 0.9976 data_time: 0.0047 memory: 8457 loss: 0.0145 decode.loss_ce: 0.0066 decode.acc_seg: 99.5974 aux.loss_ce: 0.0079 aux.acc_seg: 99.0288 +04/19 03:54:49 - mmengine - INFO - Iter(train) [ 70300/160000] base_lr: 5.6593e-05 lr: 2.0924e-07 eta: 1 day, 2:53:05 time: 0.9987 data_time: 0.0046 memory: 8457 loss: 0.0131 decode.loss_ce: 0.0058 decode.acc_seg: 99.7795 aux.loss_ce: 0.0073 aux.acc_seg: 99.4055 +04/19 03:55:39 - mmengine - INFO - Iter(train) [ 70350/160000] base_lr: 5.6562e-05 lr: 2.0912e-07 eta: 1 day, 2:51:36 time: 0.9970 data_time: 0.0048 memory: 8457 loss: 0.0127 decode.loss_ce: 0.0058 decode.acc_seg: 99.7595 aux.loss_ce: 0.0069 aux.acc_seg: 99.1049 +04/19 03:56:29 - mmengine - INFO - Iter(train) [ 70400/160000] base_lr: 5.6530e-05 lr: 2.0900e-07 eta: 1 day, 2:50:08 time: 0.9981 data_time: 0.0047 memory: 8457 loss: 0.0116 decode.loss_ce: 0.0051 decode.acc_seg: 99.7854 aux.loss_ce: 0.0064 aux.acc_seg: 99.1919 +04/19 03:57:19 - mmengine - INFO - Iter(train) [ 70450/160000] base_lr: 5.6499e-05 lr: 2.0889e-07 eta: 1 day, 2:48:39 time: 0.9968 data_time: 0.0046 memory: 8457 loss: 0.0157 decode.loss_ce: 0.0073 decode.acc_seg: 99.7282 aux.loss_ce: 0.0084 aux.acc_seg: 99.1796 +04/19 03:58:09 - mmengine - INFO - Iter(train) [ 70500/160000] base_lr: 5.6467e-05 lr: 2.0877e-07 eta: 1 day, 2:47:11 time: 0.9973 data_time: 0.0046 memory: 8457 loss: 0.0143 decode.loss_ce: 0.0064 decode.acc_seg: 99.5211 aux.loss_ce: 0.0079 aux.acc_seg: 98.7547 +04/19 03:58:59 - mmengine - INFO - Iter(train) [ 70550/160000] base_lr: 5.6436e-05 lr: 2.0865e-07 eta: 1 day, 2:45:43 time: 0.9971 data_time: 0.0048 memory: 8457 loss: 0.0152 decode.loss_ce: 0.0066 decode.acc_seg: 99.7805 aux.loss_ce: 0.0086 aux.acc_seg: 99.4310 +04/19 03:59:49 - mmengine - INFO - Iter(train) [ 70600/160000] base_lr: 5.6404e-05 lr: 2.0854e-07 eta: 1 day, 2:44:16 time: 0.9974 data_time: 0.0045 memory: 8457 loss: 0.0143 decode.loss_ce: 0.0064 decode.acc_seg: 99.6746 aux.loss_ce: 0.0079 aux.acc_seg: 98.9529 +04/19 04:00:39 - mmengine - INFO - Iter(train) [ 70650/160000] base_lr: 5.6373e-05 lr: 2.0842e-07 eta: 1 day, 2:42:49 time: 0.9985 data_time: 0.0046 memory: 8457 loss: 0.0120 decode.loss_ce: 0.0054 decode.acc_seg: 99.7784 aux.loss_ce: 0.0066 aux.acc_seg: 99.1545 +04/19 04:01:28 - mmengine - INFO - Iter(train) [ 70700/160000] base_lr: 5.6341e-05 lr: 2.0830e-07 eta: 1 day, 2:41:22 time: 0.9983 data_time: 0.0047 memory: 8457 loss: 0.0119 decode.loss_ce: 0.0054 decode.acc_seg: 99.7101 aux.loss_ce: 0.0064 aux.acc_seg: 99.1886 +04/19 04:02:18 - mmengine - INFO - Iter(train) [ 70750/160000] base_lr: 5.6310e-05 lr: 2.0819e-07 eta: 1 day, 2:39:56 time: 0.9975 data_time: 0.0045 memory: 8457 loss: 0.0129 decode.loss_ce: 0.0058 decode.acc_seg: 99.7566 aux.loss_ce: 0.0071 aux.acc_seg: 99.3814 +04/19 04:03:08 - mmengine - INFO - Iter(train) [ 70800/160000] base_lr: 5.6278e-05 lr: 2.0807e-07 eta: 1 day, 2:38:30 time: 0.9976 data_time: 0.0049 memory: 8457 loss: 0.0130 decode.loss_ce: 0.0056 decode.acc_seg: 99.7635 aux.loss_ce: 0.0074 aux.acc_seg: 99.0618 +04/19 04:03:58 - mmengine - INFO - Iter(train) [ 70850/160000] base_lr: 5.6246e-05 lr: 2.0795e-07 eta: 1 day, 2:37:05 time: 0.9962 data_time: 0.0045 memory: 8457 loss: 0.0129 decode.loss_ce: 0.0059 decode.acc_seg: 99.6508 aux.loss_ce: 0.0070 aux.acc_seg: 98.9620 +04/19 04:04:48 - mmengine - INFO - Iter(train) [ 70900/160000] base_lr: 5.6215e-05 lr: 2.0784e-07 eta: 1 day, 2:35:40 time: 0.9984 data_time: 0.0044 memory: 8457 loss: 0.0150 decode.loss_ce: 0.0068 decode.acc_seg: 99.5358 aux.loss_ce: 0.0082 aux.acc_seg: 98.5611 +04/19 04:05:38 - mmengine - INFO - Iter(train) [ 70950/160000] base_lr: 5.6183e-05 lr: 2.0772e-07 eta: 1 day, 2:34:15 time: 0.9982 data_time: 0.0047 memory: 8457 loss: 0.0120 decode.loss_ce: 0.0056 decode.acc_seg: 99.7366 aux.loss_ce: 0.0064 aux.acc_seg: 99.2506 +04/19 04:06:28 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240419_004857 +04/19 04:06:28 - mmengine - INFO - Iter(train) [ 71000/160000] base_lr: 5.6152e-05 lr: 2.0760e-07 eta: 1 day, 2:32:50 time: 0.9985 data_time: 0.0052 memory: 8457 loss: 0.0131 decode.loss_ce: 0.0059 decode.acc_seg: 99.7271 aux.loss_ce: 0.0072 aux.acc_seg: 99.3591 +04/19 04:07:18 - mmengine - INFO - Iter(train) [ 71050/160000] base_lr: 5.6120e-05 lr: 2.0749e-07 eta: 1 day, 2:31:26 time: 0.9993 data_time: 0.0045 memory: 8457 loss: 0.0138 decode.loss_ce: 0.0061 decode.acc_seg: 99.7278 aux.loss_ce: 0.0077 aux.acc_seg: 99.2346 +04/19 04:08:08 - mmengine - INFO - Iter(train) [ 71100/160000] base_lr: 5.6089e-05 lr: 2.0737e-07 eta: 1 day, 2:30:02 time: 0.9973 data_time: 0.0044 memory: 8457 loss: 0.0120 decode.loss_ce: 0.0054 decode.acc_seg: 99.7286 aux.loss_ce: 0.0066 aux.acc_seg: 99.2502 +04/19 04:08:58 - mmengine - INFO - Iter(train) [ 71150/160000] base_lr: 5.6057e-05 lr: 2.0725e-07 eta: 1 day, 2:28:38 time: 0.9982 data_time: 0.0048 memory: 8457 loss: 0.0145 decode.loss_ce: 0.0063 decode.acc_seg: 99.6756 aux.loss_ce: 0.0082 aux.acc_seg: 99.1711 +04/19 04:09:47 - mmengine - INFO - Iter(train) [ 71200/160000] base_lr: 5.6026e-05 lr: 2.0714e-07 eta: 1 day, 2:27:15 time: 0.9988 data_time: 0.0046 memory: 8457 loss: 0.0133 decode.loss_ce: 0.0058 decode.acc_seg: 99.7915 aux.loss_ce: 0.0074 aux.acc_seg: 99.2517 +04/19 04:10:37 - mmengine - INFO - Iter(train) [ 71250/160000] base_lr: 5.5994e-05 lr: 2.0702e-07 eta: 1 day, 2:25:52 time: 0.9971 data_time: 0.0051 memory: 8457 loss: 0.0117 decode.loss_ce: 0.0056 decode.acc_seg: 99.8142 aux.loss_ce: 0.0061 aux.acc_seg: 99.4389 +04/19 04:11:27 - mmengine - INFO - Iter(train) [ 71300/160000] base_lr: 5.5962e-05 lr: 2.0690e-07 eta: 1 day, 2:24:29 time: 0.9973 data_time: 0.0047 memory: 8457 loss: 0.0136 decode.loss_ce: 0.0062 decode.acc_seg: 99.6574 aux.loss_ce: 0.0074 aux.acc_seg: 99.0965 +04/19 04:12:17 - mmengine - INFO - Iter(train) [ 71350/160000] base_lr: 5.5931e-05 lr: 2.0679e-07 eta: 1 day, 2:23:07 time: 0.9980 data_time: 0.0053 memory: 8457 loss: 0.0137 decode.loss_ce: 0.0064 decode.acc_seg: 99.6717 aux.loss_ce: 0.0073 aux.acc_seg: 99.0242 +04/19 04:13:07 - mmengine - INFO - Iter(train) [ 71400/160000] base_lr: 5.5899e-05 lr: 2.0667e-07 eta: 1 day, 2:21:45 time: 0.9986 data_time: 0.0048 memory: 8457 loss: 0.0131 decode.loss_ce: 0.0060 decode.acc_seg: 99.8230 aux.loss_ce: 0.0071 aux.acc_seg: 99.4883 +04/19 04:13:57 - mmengine - INFO - Iter(train) [ 71450/160000] base_lr: 5.5868e-05 lr: 2.0655e-07 eta: 1 day, 2:20:23 time: 0.9985 data_time: 0.0050 memory: 8457 loss: 0.0138 decode.loss_ce: 0.0062 decode.acc_seg: 99.7934 aux.loss_ce: 0.0075 aux.acc_seg: 99.3322 +04/19 04:14:47 - mmengine - INFO - Iter(train) [ 71500/160000] base_lr: 5.5836e-05 lr: 2.0644e-07 eta: 1 day, 2:19:01 time: 0.9977 data_time: 0.0046 memory: 8457 loss: 0.0115 decode.loss_ce: 0.0050 decode.acc_seg: 99.7473 aux.loss_ce: 0.0066 aux.acc_seg: 99.1293 +04/19 04:15:37 - mmengine - INFO - Iter(train) [ 71550/160000] base_lr: 5.5805e-05 lr: 2.0632e-07 eta: 1 day, 2:17:39 time: 0.9969 data_time: 0.0049 memory: 8457 loss: 0.0155 decode.loss_ce: 0.0068 decode.acc_seg: 99.8236 aux.loss_ce: 0.0087 aux.acc_seg: 99.4001 +04/19 04:16:27 - mmengine - INFO - Iter(train) [ 71600/160000] base_lr: 5.5773e-05 lr: 2.0621e-07 eta: 1 day, 2:16:18 time: 0.9972 data_time: 0.0047 memory: 8457 loss: 0.0131 decode.loss_ce: 0.0060 decode.acc_seg: 99.8430 aux.loss_ce: 0.0071 aux.acc_seg: 99.5842 +04/19 04:17:16 - mmengine - INFO - Iter(train) [ 71650/160000] base_lr: 5.5742e-05 lr: 2.0609e-07 eta: 1 day, 2:14:57 time: 0.9974 data_time: 0.0046 memory: 8457 loss: 0.0131 decode.loss_ce: 0.0061 decode.acc_seg: 99.7597 aux.loss_ce: 0.0071 aux.acc_seg: 99.4097 +04/19 04:18:06 - mmengine - INFO - Iter(train) [ 71700/160000] base_lr: 5.5710e-05 lr: 2.0597e-07 eta: 1 day, 2:13:36 time: 0.9976 data_time: 0.0048 memory: 8457 loss: 0.0133 decode.loss_ce: 0.0059 decode.acc_seg: 99.7728 aux.loss_ce: 0.0075 aux.acc_seg: 99.3156 +04/19 04:18:56 - mmengine - INFO - Iter(train) [ 71750/160000] base_lr: 5.5679e-05 lr: 2.0586e-07 eta: 1 day, 2:12:16 time: 0.9975 data_time: 0.0047 memory: 8457 loss: 0.0113 decode.loss_ce: 0.0051 decode.acc_seg: 99.7702 aux.loss_ce: 0.0061 aux.acc_seg: 99.1955 +04/19 04:19:46 - mmengine - INFO - Iter(train) [ 71800/160000] base_lr: 5.5647e-05 lr: 2.0574e-07 eta: 1 day, 2:10:57 time: 0.9991 data_time: 0.0047 memory: 8457 loss: 0.0133 decode.loss_ce: 0.0056 decode.acc_seg: 99.7063 aux.loss_ce: 0.0076 aux.acc_seg: 99.0969 +04/19 04:20:36 - mmengine - INFO - Iter(train) [ 71850/160000] base_lr: 5.5615e-05 lr: 2.0562e-07 eta: 1 day, 2:09:37 time: 0.9976 data_time: 0.0045 memory: 8457 loss: 0.0144 decode.loss_ce: 0.0065 decode.acc_seg: 99.8339 aux.loss_ce: 0.0079 aux.acc_seg: 99.6168 +04/19 04:21:26 - mmengine - INFO - Iter(train) [ 71900/160000] base_lr: 5.5584e-05 lr: 2.0551e-07 eta: 1 day, 2:08:18 time: 0.9980 data_time: 0.0047 memory: 8457 loss: 0.0129 decode.loss_ce: 0.0055 decode.acc_seg: 99.7591 aux.loss_ce: 0.0074 aux.acc_seg: 99.0673 +04/19 04:22:16 - mmengine - INFO - Iter(train) [ 71950/160000] base_lr: 5.5552e-05 lr: 2.0539e-07 eta: 1 day, 2:06:59 time: 0.9987 data_time: 0.0049 memory: 8457 loss: 0.0137 decode.loss_ce: 0.0061 decode.acc_seg: 99.6096 aux.loss_ce: 0.0076 aux.acc_seg: 98.9359 +04/19 04:23:06 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240419_004857 +04/19 04:23:06 - mmengine - INFO - Iter(train) [ 72000/160000] base_lr: 5.5521e-05 lr: 2.0527e-07 eta: 1 day, 2:05:40 time: 0.9981 data_time: 0.0044 memory: 8457 loss: 0.0149 decode.loss_ce: 0.0067 decode.acc_seg: 99.6532 aux.loss_ce: 0.0082 aux.acc_seg: 98.9784 +04/19 04:23:56 - mmengine - INFO - Iter(train) [ 72050/160000] base_lr: 5.5489e-05 lr: 2.0516e-07 eta: 1 day, 2:04:22 time: 1.0004 data_time: 0.0043 memory: 8457 loss: 0.0120 decode.loss_ce: 0.0056 decode.acc_seg: 99.7738 aux.loss_ce: 0.0064 aux.acc_seg: 99.4066 +04/19 04:24:46 - mmengine - INFO - Iter(train) [ 72100/160000] base_lr: 5.5458e-05 lr: 2.0504e-07 eta: 1 day, 2:03:04 time: 0.9988 data_time: 0.0048 memory: 8457 loss: 0.0126 decode.loss_ce: 0.0053 decode.acc_seg: 99.7711 aux.loss_ce: 0.0073 aux.acc_seg: 99.2994 +04/19 04:25:36 - mmengine - INFO - Iter(train) [ 72150/160000] base_lr: 5.5426e-05 lr: 2.0492e-07 eta: 1 day, 2:01:47 time: 1.0004 data_time: 0.0052 memory: 8457 loss: 0.0118 decode.loss_ce: 0.0052 decode.acc_seg: 99.7536 aux.loss_ce: 0.0066 aux.acc_seg: 99.1631 +04/19 04:26:26 - mmengine - INFO - Iter(train) [ 72200/160000] base_lr: 5.5395e-05 lr: 2.0481e-07 eta: 1 day, 2:00:29 time: 1.0000 data_time: 0.0046 memory: 8457 loss: 0.0138 decode.loss_ce: 0.0061 decode.acc_seg: 99.7293 aux.loss_ce: 0.0078 aux.acc_seg: 98.9708 +04/19 04:27:16 - mmengine - INFO - Iter(train) [ 72250/160000] base_lr: 5.5363e-05 lr: 2.0469e-07 eta: 1 day, 1:59:12 time: 0.9989 data_time: 0.0044 memory: 8457 loss: 0.0121 decode.loss_ce: 0.0053 decode.acc_seg: 99.8394 aux.loss_ce: 0.0068 aux.acc_seg: 99.4120 +04/19 04:28:06 - mmengine - INFO - Iter(train) [ 72300/160000] base_lr: 5.5332e-05 lr: 2.0457e-07 eta: 1 day, 1:57:55 time: 0.9999 data_time: 0.0048 memory: 8457 loss: 0.0163 decode.loss_ce: 0.0072 decode.acc_seg: 99.6603 aux.loss_ce: 0.0091 aux.acc_seg: 99.1877 +04/19 04:28:56 - mmengine - INFO - Iter(train) [ 72350/160000] base_lr: 5.5300e-05 lr: 2.0446e-07 eta: 1 day, 1:56:38 time: 1.0006 data_time: 0.0048 memory: 8457 loss: 0.0140 decode.loss_ce: 0.0058 decode.acc_seg: 99.7231 aux.loss_ce: 0.0081 aux.acc_seg: 99.1499 +04/19 04:29:46 - mmengine - INFO - Iter(train) [ 72400/160000] base_lr: 5.5268e-05 lr: 2.0434e-07 eta: 1 day, 1:55:22 time: 1.0005 data_time: 0.0049 memory: 8457 loss: 0.0122 decode.loss_ce: 0.0054 decode.acc_seg: 99.6748 aux.loss_ce: 0.0067 aux.acc_seg: 99.1932 +04/19 04:30:36 - mmengine - INFO - Iter(train) [ 72450/160000] base_lr: 5.5237e-05 lr: 2.0422e-07 eta: 1 day, 1:54:06 time: 1.0009 data_time: 0.0045 memory: 8457 loss: 0.0129 decode.loss_ce: 0.0056 decode.acc_seg: 99.7463 aux.loss_ce: 0.0073 aux.acc_seg: 99.0011 +04/19 04:31:26 - mmengine - INFO - Iter(train) [ 72500/160000] base_lr: 5.5205e-05 lr: 2.0411e-07 eta: 1 day, 1:52:50 time: 1.0010 data_time: 0.0047 memory: 8457 loss: 0.0134 decode.loss_ce: 0.0064 decode.acc_seg: 99.8198 aux.loss_ce: 0.0071 aux.acc_seg: 99.5687 +04/19 04:32:16 - mmengine - INFO - Iter(train) [ 72550/160000] base_lr: 5.5174e-05 lr: 2.0399e-07 eta: 1 day, 1:51:35 time: 0.9999 data_time: 0.0048 memory: 8457 loss: 0.0133 decode.loss_ce: 0.0060 decode.acc_seg: 99.7398 aux.loss_ce: 0.0073 aux.acc_seg: 99.0044 +04/19 04:33:06 - mmengine - INFO - Iter(train) [ 72600/160000] base_lr: 5.5142e-05 lr: 2.0387e-07 eta: 1 day, 1:50:19 time: 0.9998 data_time: 0.0044 memory: 8457 loss: 0.0123 decode.loss_ce: 0.0053 decode.acc_seg: 99.7574 aux.loss_ce: 0.0070 aux.acc_seg: 99.3227 +04/19 04:33:56 - mmengine - INFO - Iter(train) [ 72650/160000] base_lr: 5.5111e-05 lr: 2.0376e-07 eta: 1 day, 1:49:04 time: 1.0023 data_time: 0.0045 memory: 8457 loss: 0.0139 decode.loss_ce: 0.0062 decode.acc_seg: 99.7295 aux.loss_ce: 0.0077 aux.acc_seg: 99.1224 +04/19 04:34:46 - mmengine - INFO - Iter(train) [ 72700/160000] base_lr: 5.5079e-05 lr: 2.0364e-07 eta: 1 day, 1:47:50 time: 1.0013 data_time: 0.0044 memory: 8457 loss: 0.0136 decode.loss_ce: 0.0061 decode.acc_seg: 99.7032 aux.loss_ce: 0.0075 aux.acc_seg: 98.8962 +04/19 04:35:36 - mmengine - INFO - Iter(train) [ 72750/160000] base_lr: 5.5048e-05 lr: 2.0352e-07 eta: 1 day, 1:46:35 time: 1.0004 data_time: 0.0046 memory: 8457 loss: 0.0117 decode.loss_ce: 0.0053 decode.acc_seg: 99.8070 aux.loss_ce: 0.0064 aux.acc_seg: 99.4915 +04/19 04:36:27 - mmengine - INFO - Iter(train) [ 72800/160000] base_lr: 5.5016e-05 lr: 2.0341e-07 eta: 1 day, 1:45:21 time: 1.0012 data_time: 0.0044 memory: 8457 loss: 0.0136 decode.loss_ce: 0.0062 decode.acc_seg: 99.8421 aux.loss_ce: 0.0073 aux.acc_seg: 99.3183 +04/19 04:37:17 - mmengine - INFO - Iter(train) [ 72850/160000] base_lr: 5.4985e-05 lr: 2.0329e-07 eta: 1 day, 1:44:07 time: 1.0010 data_time: 0.0044 memory: 8457 loss: 0.0131 decode.loss_ce: 0.0058 decode.acc_seg: 99.7499 aux.loss_ce: 0.0073 aux.acc_seg: 99.2165 +04/19 04:38:07 - mmengine - INFO - Iter(train) [ 72900/160000] base_lr: 5.4953e-05 lr: 2.0317e-07 eta: 1 day, 1:42:53 time: 0.9998 data_time: 0.0046 memory: 8457 loss: 0.0142 decode.loss_ce: 0.0063 decode.acc_seg: 99.7795 aux.loss_ce: 0.0078 aux.acc_seg: 99.2311 +04/19 04:38:57 - mmengine - INFO - Iter(train) [ 72950/160000] base_lr: 5.4921e-05 lr: 2.0306e-07 eta: 1 day, 1:41:39 time: 1.0011 data_time: 0.0045 memory: 8457 loss: 0.0132 decode.loss_ce: 0.0059 decode.acc_seg: 99.6662 aux.loss_ce: 0.0073 aux.acc_seg: 99.2420 +04/19 04:39:47 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240419_004857 +04/19 04:39:47 - mmengine - INFO - Iter(train) [ 73000/160000] base_lr: 5.4890e-05 lr: 2.0294e-07 eta: 1 day, 1:40:25 time: 1.0022 data_time: 0.0047 memory: 8457 loss: 0.0124 decode.loss_ce: 0.0056 decode.acc_seg: 99.7541 aux.loss_ce: 0.0067 aux.acc_seg: 99.3658 +04/19 04:40:37 - mmengine - INFO - Iter(train) [ 73050/160000] base_lr: 5.4858e-05 lr: 2.0282e-07 eta: 1 day, 1:39:11 time: 0.9994 data_time: 0.0046 memory: 8457 loss: 0.0121 decode.loss_ce: 0.0056 decode.acc_seg: 99.7995 aux.loss_ce: 0.0065 aux.acc_seg: 99.4841 +04/19 04:41:27 - mmengine - INFO - Iter(train) [ 73100/160000] base_lr: 5.4827e-05 lr: 2.0271e-07 eta: 1 day, 1:37:58 time: 1.0006 data_time: 0.0050 memory: 8457 loss: 0.0127 decode.loss_ce: 0.0058 decode.acc_seg: 99.7210 aux.loss_ce: 0.0069 aux.acc_seg: 99.0213 +04/19 04:42:17 - mmengine - INFO - Iter(train) [ 73150/160000] base_lr: 5.4795e-05 lr: 2.0259e-07 eta: 1 day, 1:36:45 time: 1.0021 data_time: 0.0050 memory: 8457 loss: 0.0127 decode.loss_ce: 0.0055 decode.acc_seg: 99.6876 aux.loss_ce: 0.0073 aux.acc_seg: 98.7301 +04/19 04:43:07 - mmengine - INFO - Iter(train) [ 73200/160000] base_lr: 5.4764e-05 lr: 2.0247e-07 eta: 1 day, 1:35:32 time: 0.9996 data_time: 0.0047 memory: 8457 loss: 0.0130 decode.loss_ce: 0.0058 decode.acc_seg: 99.6916 aux.loss_ce: 0.0072 aux.acc_seg: 99.1751 +04/19 04:43:57 - mmengine - INFO - Iter(train) [ 73250/160000] base_lr: 5.4732e-05 lr: 2.0236e-07 eta: 1 day, 1:34:19 time: 0.9997 data_time: 0.0045 memory: 8457 loss: 0.0146 decode.loss_ce: 0.0063 decode.acc_seg: 99.7709 aux.loss_ce: 0.0083 aux.acc_seg: 99.2876 +04/19 04:44:47 - mmengine - INFO - Iter(train) [ 73300/160000] base_lr: 5.4701e-05 lr: 2.0224e-07 eta: 1 day, 1:33:06 time: 1.0012 data_time: 0.0053 memory: 8457 loss: 0.0130 decode.loss_ce: 0.0057 decode.acc_seg: 99.7135 aux.loss_ce: 0.0073 aux.acc_seg: 98.9508 +04/19 04:45:37 - mmengine - INFO - Iter(train) [ 73350/160000] base_lr: 5.4669e-05 lr: 2.0212e-07 eta: 1 day, 1:31:53 time: 1.0002 data_time: 0.0046 memory: 8457 loss: 0.0122 decode.loss_ce: 0.0059 decode.acc_seg: 99.7562 aux.loss_ce: 0.0063 aux.acc_seg: 99.3999 +04/19 04:46:27 - mmengine - INFO - Iter(train) [ 73400/160000] base_lr: 5.4638e-05 lr: 2.0201e-07 eta: 1 day, 1:30:41 time: 1.0008 data_time: 0.0045 memory: 8457 loss: 0.0137 decode.loss_ce: 0.0061 decode.acc_seg: 99.7080 aux.loss_ce: 0.0076 aux.acc_seg: 99.0358 +04/19 04:47:17 - mmengine - INFO - Iter(train) [ 73450/160000] base_lr: 5.4606e-05 lr: 2.0189e-07 eta: 1 day, 1:29:28 time: 1.0004 data_time: 0.0049 memory: 8457 loss: 0.0145 decode.loss_ce: 0.0065 decode.acc_seg: 99.7345 aux.loss_ce: 0.0080 aux.acc_seg: 99.1789 +04/19 04:48:07 - mmengine - INFO - Iter(train) [ 73500/160000] base_lr: 5.4574e-05 lr: 2.0177e-07 eta: 1 day, 1:28:16 time: 1.0005 data_time: 0.0044 memory: 8457 loss: 0.0132 decode.loss_ce: 0.0062 decode.acc_seg: 99.7120 aux.loss_ce: 0.0070 aux.acc_seg: 99.2508 +04/19 04:48:57 - mmengine - INFO - Iter(train) [ 73550/160000] base_lr: 5.4543e-05 lr: 2.0166e-07 eta: 1 day, 1:27:04 time: 1.0006 data_time: 0.0044 memory: 8457 loss: 0.0152 decode.loss_ce: 0.0068 decode.acc_seg: 99.7654 aux.loss_ce: 0.0084 aux.acc_seg: 99.1514 +04/19 04:49:47 - mmengine - INFO - Iter(train) [ 73600/160000] base_lr: 5.4511e-05 lr: 2.0154e-07 eta: 1 day, 1:25:52 time: 1.0010 data_time: 0.0045 memory: 8457 loss: 0.0118 decode.loss_ce: 0.0052 decode.acc_seg: 99.7936 aux.loss_ce: 0.0066 aux.acc_seg: 99.4471 +04/19 04:50:37 - mmengine - INFO - Iter(train) [ 73650/160000] base_lr: 5.4480e-05 lr: 2.0142e-07 eta: 1 day, 1:24:41 time: 1.0006 data_time: 0.0045 memory: 8457 loss: 0.0133 decode.loss_ce: 0.0060 decode.acc_seg: 99.6895 aux.loss_ce: 0.0073 aux.acc_seg: 98.9532 +04/19 04:51:27 - mmengine - INFO - Iter(train) [ 73700/160000] base_lr: 5.4448e-05 lr: 2.0131e-07 eta: 1 day, 1:23:29 time: 0.9983 data_time: 0.0045 memory: 8457 loss: 0.0129 decode.loss_ce: 0.0058 decode.acc_seg: 99.7168 aux.loss_ce: 0.0071 aux.acc_seg: 99.0805 +04/19 04:52:17 - mmengine - INFO - Iter(train) [ 73750/160000] base_lr: 5.4417e-05 lr: 2.0119e-07 eta: 1 day, 1:22:17 time: 0.9983 data_time: 0.0044 memory: 8457 loss: 0.0139 decode.loss_ce: 0.0063 decode.acc_seg: 99.8234 aux.loss_ce: 0.0076 aux.acc_seg: 99.5314 +04/19 04:53:07 - mmengine - INFO - Iter(train) [ 73800/160000] base_lr: 5.4385e-05 lr: 2.0107e-07 eta: 1 day, 1:21:06 time: 1.0010 data_time: 0.0045 memory: 8457 loss: 0.0122 decode.loss_ce: 0.0056 decode.acc_seg: 99.7301 aux.loss_ce: 0.0066 aux.acc_seg: 99.3900 +04/19 04:53:57 - mmengine - INFO - Iter(train) [ 73850/160000] base_lr: 5.4354e-05 lr: 2.0096e-07 eta: 1 day, 1:19:55 time: 1.0005 data_time: 0.0043 memory: 8457 loss: 0.0134 decode.loss_ce: 0.0060 decode.acc_seg: 99.7335 aux.loss_ce: 0.0073 aux.acc_seg: 99.1028 +04/19 04:54:47 - mmengine - INFO - Iter(train) [ 73900/160000] base_lr: 5.4322e-05 lr: 2.0084e-07 eta: 1 day, 1:18:43 time: 1.0005 data_time: 0.0044 memory: 8457 loss: 0.0127 decode.loss_ce: 0.0058 decode.acc_seg: 99.8169 aux.loss_ce: 0.0069 aux.acc_seg: 99.5962 +04/19 04:55:37 - mmengine - INFO - Iter(train) [ 73950/160000] base_lr: 5.4291e-05 lr: 2.0072e-07 eta: 1 day, 1:17:32 time: 1.0002 data_time: 0.0050 memory: 8457 loss: 0.0137 decode.loss_ce: 0.0063 decode.acc_seg: 99.7305 aux.loss_ce: 0.0073 aux.acc_seg: 99.1680 +04/19 04:56:27 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240419_004857 +04/19 04:56:27 - mmengine - INFO - Iter(train) [ 74000/160000] base_lr: 5.4259e-05 lr: 2.0061e-07 eta: 1 day, 1:16:21 time: 0.9996 data_time: 0.0045 memory: 8457 loss: 0.0130 decode.loss_ce: 0.0056 decode.acc_seg: 99.7553 aux.loss_ce: 0.0073 aux.acc_seg: 99.1318 +04/19 04:57:17 - mmengine - INFO - Iter(train) [ 74050/160000] base_lr: 5.4227e-05 lr: 2.0049e-07 eta: 1 day, 1:15:11 time: 1.0020 data_time: 0.0044 memory: 8457 loss: 0.0099 decode.loss_ce: 0.0046 decode.acc_seg: 99.7950 aux.loss_ce: 0.0053 aux.acc_seg: 99.3742 +04/19 04:58:07 - mmengine - INFO - Iter(train) [ 74100/160000] base_lr: 5.4196e-05 lr: 2.0037e-07 eta: 1 day, 1:14:00 time: 1.0002 data_time: 0.0046 memory: 8457 loss: 0.0122 decode.loss_ce: 0.0054 decode.acc_seg: 99.7715 aux.loss_ce: 0.0067 aux.acc_seg: 99.3795 +04/19 04:58:57 - mmengine - INFO - Iter(train) [ 74150/160000] base_lr: 5.4164e-05 lr: 2.0026e-07 eta: 1 day, 1:12:50 time: 1.0006 data_time: 0.0045 memory: 8457 loss: 0.0130 decode.loss_ce: 0.0054 decode.acc_seg: 99.7684 aux.loss_ce: 0.0077 aux.acc_seg: 99.3389 +04/19 04:59:47 - mmengine - INFO - Iter(train) [ 74200/160000] base_lr: 5.4133e-05 lr: 2.0014e-07 eta: 1 day, 1:11:40 time: 1.0006 data_time: 0.0046 memory: 8457 loss: 0.0142 decode.loss_ce: 0.0061 decode.acc_seg: 99.7543 aux.loss_ce: 0.0081 aux.acc_seg: 99.1320 +04/19 05:00:37 - mmengine - INFO - Iter(train) [ 74250/160000] base_lr: 5.4101e-05 lr: 2.0002e-07 eta: 1 day, 1:10:30 time: 1.0003 data_time: 0.0051 memory: 8457 loss: 0.0124 decode.loss_ce: 0.0056 decode.acc_seg: 99.8207 aux.loss_ce: 0.0068 aux.acc_seg: 99.3891 +04/19 05:01:27 - mmengine - INFO - Iter(train) [ 74300/160000] base_lr: 5.4070e-05 lr: 1.9991e-07 eta: 1 day, 1:09:20 time: 1.0001 data_time: 0.0048 memory: 8457 loss: 0.0124 decode.loss_ce: 0.0058 decode.acc_seg: 99.7780 aux.loss_ce: 0.0066 aux.acc_seg: 99.2271 +04/19 05:02:17 - mmengine - INFO - Iter(train) [ 74350/160000] base_lr: 5.4038e-05 lr: 1.9979e-07 eta: 1 day, 1:08:10 time: 0.9995 data_time: 0.0049 memory: 8457 loss: 0.0142 decode.loss_ce: 0.0062 decode.acc_seg: 99.6273 aux.loss_ce: 0.0080 aux.acc_seg: 98.3192 +04/19 05:03:07 - mmengine - INFO - Iter(train) [ 74400/160000] base_lr: 5.4007e-05 lr: 1.9967e-07 eta: 1 day, 1:07:00 time: 0.9985 data_time: 0.0044 memory: 8457 loss: 0.0127 decode.loss_ce: 0.0055 decode.acc_seg: 99.7429 aux.loss_ce: 0.0072 aux.acc_seg: 99.1068 +04/19 05:03:57 - mmengine - INFO - Iter(train) [ 74450/160000] base_lr: 5.3975e-05 lr: 1.9956e-07 eta: 1 day, 1:05:51 time: 0.9995 data_time: 0.0046 memory: 8457 loss: 0.0130 decode.loss_ce: 0.0059 decode.acc_seg: 99.7587 aux.loss_ce: 0.0071 aux.acc_seg: 99.3235 +04/19 05:04:47 - mmengine - INFO - Iter(train) [ 74500/160000] base_lr: 5.3944e-05 lr: 1.9944e-07 eta: 1 day, 1:04:41 time: 0.9986 data_time: 0.0050 memory: 8457 loss: 0.0146 decode.loss_ce: 0.0066 decode.acc_seg: 99.7036 aux.loss_ce: 0.0080 aux.acc_seg: 99.2405 +04/19 05:05:37 - mmengine - INFO - Iter(train) [ 74550/160000] base_lr: 5.3912e-05 lr: 1.9932e-07 eta: 1 day, 1:03:32 time: 0.9985 data_time: 0.0047 memory: 8457 loss: 0.0143 decode.loss_ce: 0.0061 decode.acc_seg: 99.7038 aux.loss_ce: 0.0082 aux.acc_seg: 99.0374 +04/19 05:06:27 - mmengine - INFO - Iter(train) [ 74600/160000] base_lr: 5.3880e-05 lr: 1.9921e-07 eta: 1 day, 1:02:22 time: 0.9999 data_time: 0.0049 memory: 8457 loss: 0.0133 decode.loss_ce: 0.0061 decode.acc_seg: 99.7276 aux.loss_ce: 0.0072 aux.acc_seg: 98.9836 +04/19 05:07:17 - mmengine - INFO - Iter(train) [ 74650/160000] base_lr: 5.3849e-05 lr: 1.9909e-07 eta: 1 day, 1:01:13 time: 0.9986 data_time: 0.0047 memory: 8457 loss: 0.0144 decode.loss_ce: 0.0064 decode.acc_seg: 99.7622 aux.loss_ce: 0.0080 aux.acc_seg: 99.2432 +04/19 05:08:07 - mmengine - INFO - Iter(train) [ 74700/160000] base_lr: 5.3817e-05 lr: 1.9897e-07 eta: 1 day, 1:00:04 time: 0.9990 data_time: 0.0046 memory: 8457 loss: 0.0142 decode.loss_ce: 0.0070 decode.acc_seg: 99.7416 aux.loss_ce: 0.0072 aux.acc_seg: 99.4238 +04/19 05:08:57 - mmengine - INFO - Iter(train) [ 74750/160000] base_lr: 5.3786e-05 lr: 1.9886e-07 eta: 1 day, 0:58:55 time: 0.9984 data_time: 0.0046 memory: 8457 loss: 0.0120 decode.loss_ce: 0.0055 decode.acc_seg: 99.7988 aux.loss_ce: 0.0065 aux.acc_seg: 99.2226 +04/19 05:09:47 - mmengine - INFO - Iter(train) [ 74800/160000] base_lr: 5.3754e-05 lr: 1.9874e-07 eta: 1 day, 0:57:46 time: 1.0000 data_time: 0.0048 memory: 8457 loss: 0.0132 decode.loss_ce: 0.0057 decode.acc_seg: 99.7255 aux.loss_ce: 0.0075 aux.acc_seg: 99.1173 +04/19 05:10:37 - mmengine - INFO - Iter(train) [ 74850/160000] base_lr: 5.3723e-05 lr: 1.9862e-07 eta: 1 day, 0:56:38 time: 0.9993 data_time: 0.0044 memory: 8457 loss: 0.0120 decode.loss_ce: 0.0054 decode.acc_seg: 99.8716 aux.loss_ce: 0.0066 aux.acc_seg: 99.4553 +04/19 05:11:27 - mmengine - INFO - Iter(train) [ 74900/160000] base_lr: 5.3691e-05 lr: 1.9851e-07 eta: 1 day, 0:55:29 time: 0.9976 data_time: 0.0047 memory: 8457 loss: 0.0132 decode.loss_ce: 0.0060 decode.acc_seg: 99.7499 aux.loss_ce: 0.0072 aux.acc_seg: 99.4543 +04/19 05:12:17 - mmengine - INFO - Iter(train) [ 74950/160000] base_lr: 5.3660e-05 lr: 1.9839e-07 eta: 1 day, 0:54:21 time: 0.9996 data_time: 0.0047 memory: 8457 loss: 0.0147 decode.loss_ce: 0.0063 decode.acc_seg: 99.7396 aux.loss_ce: 0.0084 aux.acc_seg: 99.0547 +04/19 05:13:07 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240419_004857 +04/19 05:13:07 - mmengine - INFO - Iter(train) [ 75000/160000] base_lr: 5.3628e-05 lr: 1.9827e-07 eta: 1 day, 0:53:12 time: 0.9981 data_time: 0.0047 memory: 8457 loss: 0.0138 decode.loss_ce: 0.0062 decode.acc_seg: 99.7417 aux.loss_ce: 0.0076 aux.acc_seg: 99.1919 +04/19 05:13:57 - mmengine - INFO - Iter(train) [ 75050/160000] base_lr: 5.3597e-05 lr: 1.9816e-07 eta: 1 day, 0:52:04 time: 1.0009 data_time: 0.0048 memory: 8457 loss: 0.0137 decode.loss_ce: 0.0060 decode.acc_seg: 99.7860 aux.loss_ce: 0.0077 aux.acc_seg: 99.1089 +04/19 05:14:47 - mmengine - INFO - Iter(train) [ 75100/160000] base_lr: 5.3565e-05 lr: 1.9804e-07 eta: 1 day, 0:50:56 time: 0.9992 data_time: 0.0044 memory: 8457 loss: 0.0135 decode.loss_ce: 0.0060 decode.acc_seg: 99.7814 aux.loss_ce: 0.0076 aux.acc_seg: 99.3277 +04/19 05:15:37 - mmengine - INFO - Iter(train) [ 75150/160000] base_lr: 5.3533e-05 lr: 1.9792e-07 eta: 1 day, 0:49:49 time: 0.9985 data_time: 0.0048 memory: 8457 loss: 0.0143 decode.loss_ce: 0.0062 decode.acc_seg: 99.7602 aux.loss_ce: 0.0081 aux.acc_seg: 99.2458 +04/19 05:16:27 - mmengine - INFO - Iter(train) [ 75200/160000] base_lr: 5.3502e-05 lr: 1.9781e-07 eta: 1 day, 0:48:41 time: 0.9996 data_time: 0.0045 memory: 8457 loss: 0.0142 decode.loss_ce: 0.0066 decode.acc_seg: 99.7877 aux.loss_ce: 0.0077 aux.acc_seg: 99.3122 +04/19 05:17:17 - mmengine - INFO - Iter(train) [ 75250/160000] base_lr: 5.3470e-05 lr: 1.9769e-07 eta: 1 day, 0:47:33 time: 0.9993 data_time: 0.0046 memory: 8457 loss: 0.0141 decode.loss_ce: 0.0063 decode.acc_seg: 99.7557 aux.loss_ce: 0.0077 aux.acc_seg: 99.0278 +04/19 05:18:07 - mmengine - INFO - Iter(train) [ 75300/160000] base_lr: 5.3439e-05 lr: 1.9757e-07 eta: 1 day, 0:46:26 time: 0.9989 data_time: 0.0045 memory: 8457 loss: 0.0144 decode.loss_ce: 0.0064 decode.acc_seg: 99.7427 aux.loss_ce: 0.0080 aux.acc_seg: 98.9557 +04/19 05:18:57 - mmengine - INFO - Iter(train) [ 75350/160000] base_lr: 5.3407e-05 lr: 1.9746e-07 eta: 1 day, 0:45:18 time: 0.9993 data_time: 0.0045 memory: 8457 loss: 0.0144 decode.loss_ce: 0.0063 decode.acc_seg: 99.8064 aux.loss_ce: 0.0082 aux.acc_seg: 99.4492 +04/19 05:19:47 - mmengine - INFO - Iter(train) [ 75400/160000] base_lr: 5.3376e-05 lr: 1.9734e-07 eta: 1 day, 0:44:11 time: 0.9991 data_time: 0.0045 memory: 8457 loss: 0.0116 decode.loss_ce: 0.0050 decode.acc_seg: 99.7866 aux.loss_ce: 0.0066 aux.acc_seg: 99.4564 +04/19 05:20:37 - mmengine - INFO - Iter(train) [ 75450/160000] base_lr: 5.3344e-05 lr: 1.9722e-07 eta: 1 day, 0:43:04 time: 0.9992 data_time: 0.0046 memory: 8457 loss: 0.0141 decode.loss_ce: 0.0065 decode.acc_seg: 99.7326 aux.loss_ce: 0.0077 aux.acc_seg: 99.1964 +04/19 05:21:27 - mmengine - INFO - Iter(train) [ 75500/160000] base_lr: 5.3313e-05 lr: 1.9711e-07 eta: 1 day, 0:41:57 time: 0.9987 data_time: 0.0045 memory: 8457 loss: 0.0130 decode.loss_ce: 0.0055 decode.acc_seg: 99.7711 aux.loss_ce: 0.0074 aux.acc_seg: 99.3107 +04/19 05:22:17 - mmengine - INFO - Iter(train) [ 75550/160000] base_lr: 5.3281e-05 lr: 1.9699e-07 eta: 1 day, 0:40:50 time: 1.0003 data_time: 0.0048 memory: 8457 loss: 0.0144 decode.loss_ce: 0.0065 decode.acc_seg: 99.7040 aux.loss_ce: 0.0079 aux.acc_seg: 99.0627 +04/19 05:23:07 - mmengine - INFO - Iter(train) [ 75600/160000] base_lr: 5.3250e-05 lr: 1.9687e-07 eta: 1 day, 0:39:43 time: 0.9978 data_time: 0.0050 memory: 8457 loss: 0.0117 decode.loss_ce: 0.0052 decode.acc_seg: 99.7404 aux.loss_ce: 0.0065 aux.acc_seg: 99.1165 +04/19 05:23:57 - mmengine - INFO - Iter(train) [ 75650/160000] base_lr: 5.3218e-05 lr: 1.9676e-07 eta: 1 day, 0:38:36 time: 0.9983 data_time: 0.0049 memory: 8457 loss: 0.0135 decode.loss_ce: 0.0058 decode.acc_seg: 99.7952 aux.loss_ce: 0.0077 aux.acc_seg: 99.4871 +04/19 05:24:47 - mmengine - INFO - Iter(train) [ 75700/160000] base_lr: 5.3186e-05 lr: 1.9664e-07 eta: 1 day, 0:37:29 time: 1.0001 data_time: 0.0049 memory: 8457 loss: 0.0138 decode.loss_ce: 0.0063 decode.acc_seg: 99.6210 aux.loss_ce: 0.0075 aux.acc_seg: 98.9208 +04/19 05:25:36 - mmengine - INFO - Iter(train) [ 75750/160000] base_lr: 5.3155e-05 lr: 1.9652e-07 eta: 1 day, 0:36:22 time: 0.9981 data_time: 0.0049 memory: 8457 loss: 0.0143 decode.loss_ce: 0.0066 decode.acc_seg: 99.6876 aux.loss_ce: 0.0077 aux.acc_seg: 99.2889 +04/19 05:26:26 - mmengine - INFO - Iter(train) [ 75800/160000] base_lr: 5.3123e-05 lr: 1.9641e-07 eta: 1 day, 0:35:16 time: 0.9984 data_time: 0.0048 memory: 8457 loss: 0.0124 decode.loss_ce: 0.0056 decode.acc_seg: 99.7414 aux.loss_ce: 0.0067 aux.acc_seg: 99.2846 +04/19 05:27:16 - mmengine - INFO - Iter(train) [ 75850/160000] base_lr: 5.3092e-05 lr: 1.9629e-07 eta: 1 day, 0:34:09 time: 0.9990 data_time: 0.0048 memory: 8457 loss: 0.0163 decode.loss_ce: 0.0072 decode.acc_seg: 99.7507 aux.loss_ce: 0.0091 aux.acc_seg: 99.0150 +04/19 05:28:06 - mmengine - INFO - Iter(train) [ 75900/160000] base_lr: 5.3060e-05 lr: 1.9617e-07 eta: 1 day, 0:33:03 time: 0.9985 data_time: 0.0053 memory: 8457 loss: 0.0122 decode.loss_ce: 0.0053 decode.acc_seg: 99.7759 aux.loss_ce: 0.0069 aux.acc_seg: 99.4860 +04/19 05:28:56 - mmengine - INFO - Iter(train) [ 75950/160000] base_lr: 5.3029e-05 lr: 1.9606e-07 eta: 1 day, 0:31:56 time: 0.9987 data_time: 0.0044 memory: 8457 loss: 0.0128 decode.loss_ce: 0.0055 decode.acc_seg: 99.8390 aux.loss_ce: 0.0072 aux.acc_seg: 99.5131 +04/19 05:29:46 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240419_004857 +04/19 05:29:46 - mmengine - INFO - Iter(train) [ 76000/160000] base_lr: 5.2997e-05 lr: 1.9594e-07 eta: 1 day, 0:30:50 time: 0.9990 data_time: 0.0045 memory: 8457 loss: 0.0137 decode.loss_ce: 0.0063 decode.acc_seg: 99.6550 aux.loss_ce: 0.0074 aux.acc_seg: 98.8285 +04/19 05:30:36 - mmengine - INFO - Iter(train) [ 76050/160000] base_lr: 5.2966e-05 lr: 1.9582e-07 eta: 1 day, 0:29:44 time: 0.9975 data_time: 0.0047 memory: 8457 loss: 0.0134 decode.loss_ce: 0.0061 decode.acc_seg: 99.7934 aux.loss_ce: 0.0073 aux.acc_seg: 99.1962 +04/19 05:31:26 - mmengine - INFO - Iter(train) [ 76100/160000] base_lr: 5.2934e-05 lr: 1.9571e-07 eta: 1 day, 0:28:37 time: 0.9982 data_time: 0.0054 memory: 8457 loss: 0.0137 decode.loss_ce: 0.0061 decode.acc_seg: 99.7311 aux.loss_ce: 0.0076 aux.acc_seg: 99.2887 +04/19 05:32:16 - mmengine - INFO - Iter(train) [ 76150/160000] base_lr: 5.2903e-05 lr: 1.9559e-07 eta: 1 day, 0:27:31 time: 0.9984 data_time: 0.0049 memory: 8457 loss: 0.0136 decode.loss_ce: 0.0061 decode.acc_seg: 99.7318 aux.loss_ce: 0.0075 aux.acc_seg: 99.2664 +04/19 05:33:06 - mmengine - INFO - Iter(train) [ 76200/160000] base_lr: 5.2871e-05 lr: 1.9547e-07 eta: 1 day, 0:26:26 time: 0.9985 data_time: 0.0050 memory: 8457 loss: 0.0140 decode.loss_ce: 0.0060 decode.acc_seg: 99.7332 aux.loss_ce: 0.0080 aux.acc_seg: 99.0215 +04/19 05:33:56 - mmengine - INFO - Iter(train) [ 76250/160000] base_lr: 5.2839e-05 lr: 1.9536e-07 eta: 1 day, 0:25:20 time: 0.9986 data_time: 0.0050 memory: 8457 loss: 0.0127 decode.loss_ce: 0.0058 decode.acc_seg: 99.6675 aux.loss_ce: 0.0070 aux.acc_seg: 98.7019 +04/19 05:34:46 - mmengine - INFO - Iter(train) [ 76300/160000] base_lr: 5.2808e-05 lr: 1.9524e-07 eta: 1 day, 0:24:14 time: 0.9992 data_time: 0.0050 memory: 8457 loss: 0.0146 decode.loss_ce: 0.0069 decode.acc_seg: 99.7238 aux.loss_ce: 0.0077 aux.acc_seg: 99.3649 +04/19 05:35:36 - mmengine - INFO - Iter(train) [ 76350/160000] base_lr: 5.2776e-05 lr: 1.9513e-07 eta: 1 day, 0:23:08 time: 0.9979 data_time: 0.0050 memory: 8457 loss: 0.0134 decode.loss_ce: 0.0059 decode.acc_seg: 99.7236 aux.loss_ce: 0.0075 aux.acc_seg: 99.2241 +04/19 05:36:25 - mmengine - INFO - Iter(train) [ 76400/160000] base_lr: 5.2745e-05 lr: 1.9501e-07 eta: 1 day, 0:22:03 time: 0.9979 data_time: 0.0050 memory: 8457 loss: 0.0127 decode.loss_ce: 0.0060 decode.acc_seg: 99.7807 aux.loss_ce: 0.0067 aux.acc_seg: 99.5777 +04/19 05:37:15 - mmengine - INFO - Iter(train) [ 76450/160000] base_lr: 5.2713e-05 lr: 1.9489e-07 eta: 1 day, 0:20:57 time: 0.9987 data_time: 0.0051 memory: 8457 loss: 0.0138 decode.loss_ce: 0.0061 decode.acc_seg: 99.6969 aux.loss_ce: 0.0077 aux.acc_seg: 99.1611 +04/19 05:38:05 - mmengine - INFO - Iter(train) [ 76500/160000] base_lr: 5.2682e-05 lr: 1.9478e-07 eta: 1 day, 0:19:52 time: 0.9978 data_time: 0.0046 memory: 8457 loss: 0.0143 decode.loss_ce: 0.0060 decode.acc_seg: 99.6164 aux.loss_ce: 0.0082 aux.acc_seg: 98.6273 +04/19 05:38:55 - mmengine - INFO - Iter(train) [ 76550/160000] base_lr: 5.2650e-05 lr: 1.9466e-07 eta: 1 day, 0:18:47 time: 0.9967 data_time: 0.0046 memory: 8457 loss: 0.0124 decode.loss_ce: 0.0057 decode.acc_seg: 99.6601 aux.loss_ce: 0.0067 aux.acc_seg: 99.0395 +04/19 05:39:45 - mmengine - INFO - Iter(train) [ 76600/160000] base_lr: 5.2619e-05 lr: 1.9454e-07 eta: 1 day, 0:17:41 time: 0.9970 data_time: 0.0046 memory: 8457 loss: 0.0132 decode.loss_ce: 0.0059 decode.acc_seg: 99.7316 aux.loss_ce: 0.0073 aux.acc_seg: 99.1491 +04/19 05:40:35 - mmengine - INFO - Iter(train) [ 76650/160000] base_lr: 5.2587e-05 lr: 1.9443e-07 eta: 1 day, 0:16:36 time: 0.9980 data_time: 0.0050 memory: 8457 loss: 0.0133 decode.loss_ce: 0.0060 decode.acc_seg: 99.6912 aux.loss_ce: 0.0074 aux.acc_seg: 99.0866 +04/19 05:41:25 - mmengine - INFO - Iter(train) [ 76700/160000] base_lr: 5.2556e-05 lr: 1.9431e-07 eta: 1 day, 0:15:31 time: 0.9988 data_time: 0.0048 memory: 8457 loss: 0.0125 decode.loss_ce: 0.0054 decode.acc_seg: 99.7957 aux.loss_ce: 0.0071 aux.acc_seg: 99.3572 +04/19 05:42:15 - mmengine - INFO - Iter(train) [ 76750/160000] base_lr: 5.2524e-05 lr: 1.9419e-07 eta: 1 day, 0:14:26 time: 0.9984 data_time: 0.0047 memory: 8457 loss: 0.0130 decode.loss_ce: 0.0057 decode.acc_seg: 99.7433 aux.loss_ce: 0.0072 aux.acc_seg: 99.1327 +04/19 05:43:05 - mmengine - INFO - Iter(train) [ 76800/160000] base_lr: 5.2492e-05 lr: 1.9408e-07 eta: 1 day, 0:13:21 time: 0.9982 data_time: 0.0046 memory: 8457 loss: 0.0134 decode.loss_ce: 0.0057 decode.acc_seg: 99.7021 aux.loss_ce: 0.0077 aux.acc_seg: 99.0967 +04/19 05:43:55 - mmengine - INFO - Iter(train) [ 76850/160000] base_lr: 5.2461e-05 lr: 1.9396e-07 eta: 1 day, 0:12:17 time: 0.9969 data_time: 0.0045 memory: 8457 loss: 0.0140 decode.loss_ce: 0.0063 decode.acc_seg: 99.7480 aux.loss_ce: 0.0077 aux.acc_seg: 99.3870 +04/19 05:44:44 - mmengine - INFO - Iter(train) [ 76900/160000] base_lr: 5.2429e-05 lr: 1.9384e-07 eta: 1 day, 0:11:12 time: 0.9968 data_time: 0.0045 memory: 8457 loss: 0.0129 decode.loss_ce: 0.0058 decode.acc_seg: 99.6981 aux.loss_ce: 0.0070 aux.acc_seg: 99.1398 +04/19 05:45:34 - mmengine - INFO - Iter(train) [ 76950/160000] base_lr: 5.2398e-05 lr: 1.9373e-07 eta: 1 day, 0:10:07 time: 0.9973 data_time: 0.0045 memory: 8457 loss: 0.0150 decode.loss_ce: 0.0065 decode.acc_seg: 99.7515 aux.loss_ce: 0.0085 aux.acc_seg: 99.0440 +04/19 05:46:24 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240419_004857 +04/19 05:46:24 - mmengine - INFO - Iter(train) [ 77000/160000] base_lr: 5.2366e-05 lr: 1.9361e-07 eta: 1 day, 0:09:02 time: 0.9980 data_time: 0.0046 memory: 8457 loss: 0.0129 decode.loss_ce: 0.0057 decode.acc_seg: 99.7608 aux.loss_ce: 0.0072 aux.acc_seg: 99.3055 +04/19 05:47:14 - mmengine - INFO - Iter(train) [ 77050/160000] base_lr: 5.2335e-05 lr: 1.9349e-07 eta: 1 day, 0:07:58 time: 0.9988 data_time: 0.0049 memory: 8457 loss: 0.0138 decode.loss_ce: 0.0062 decode.acc_seg: 99.7597 aux.loss_ce: 0.0076 aux.acc_seg: 99.3515 +04/19 05:48:04 - mmengine - INFO - Iter(train) [ 77100/160000] base_lr: 5.2303e-05 lr: 1.9338e-07 eta: 1 day, 0:06:53 time: 0.9975 data_time: 0.0045 memory: 8457 loss: 0.0137 decode.loss_ce: 0.0060 decode.acc_seg: 99.7835 aux.loss_ce: 0.0077 aux.acc_seg: 98.8512 +04/19 05:48:54 - mmengine - INFO - Iter(train) [ 77150/160000] base_lr: 5.2272e-05 lr: 1.9326e-07 eta: 1 day, 0:05:49 time: 0.9970 data_time: 0.0045 memory: 8457 loss: 0.0135 decode.loss_ce: 0.0060 decode.acc_seg: 99.7192 aux.loss_ce: 0.0074 aux.acc_seg: 99.2281 +04/19 05:49:44 - mmengine - INFO - Iter(train) [ 77200/160000] base_lr: 5.2240e-05 lr: 1.9314e-07 eta: 1 day, 0:04:45 time: 0.9968 data_time: 0.0047 memory: 8457 loss: 0.0132 decode.loss_ce: 0.0059 decode.acc_seg: 99.6908 aux.loss_ce: 0.0073 aux.acc_seg: 98.7049 +04/19 05:50:34 - mmengine - INFO - Iter(train) [ 77250/160000] base_lr: 5.2209e-05 lr: 1.9303e-07 eta: 1 day, 0:03:41 time: 0.9989 data_time: 0.0054 memory: 8457 loss: 0.0125 decode.loss_ce: 0.0055 decode.acc_seg: 99.7864 aux.loss_ce: 0.0070 aux.acc_seg: 99.1493 +04/19 05:51:24 - mmengine - INFO - Iter(train) [ 77300/160000] base_lr: 5.2177e-05 lr: 1.9291e-07 eta: 1 day, 0:02:37 time: 0.9981 data_time: 0.0045 memory: 8457 loss: 0.0131 decode.loss_ce: 0.0058 decode.acc_seg: 99.7105 aux.loss_ce: 0.0073 aux.acc_seg: 99.0391 +04/19 05:52:14 - mmengine - INFO - Iter(train) [ 77350/160000] base_lr: 5.2145e-05 lr: 1.9279e-07 eta: 1 day, 0:01:34 time: 0.9991 data_time: 0.0050 memory: 8457 loss: 0.0123 decode.loss_ce: 0.0056 decode.acc_seg: 99.7997 aux.loss_ce: 0.0067 aux.acc_seg: 99.4102 +04/19 05:53:04 - mmengine - INFO - Iter(train) [ 77400/160000] base_lr: 5.2114e-05 lr: 1.9268e-07 eta: 1 day, 0:00:30 time: 0.9982 data_time: 0.0048 memory: 8457 loss: 0.0110 decode.loss_ce: 0.0051 decode.acc_seg: 99.8072 aux.loss_ce: 0.0059 aux.acc_seg: 99.2752 +04/19 05:53:54 - mmengine - INFO - Iter(train) [ 77450/160000] base_lr: 5.2082e-05 lr: 1.9256e-07 eta: 23:59:26 time: 0.9987 data_time: 0.0050 memory: 8457 loss: 0.0129 decode.loss_ce: 0.0060 decode.acc_seg: 99.6710 aux.loss_ce: 0.0069 aux.acc_seg: 98.9635 +04/19 05:54:43 - mmengine - INFO - Iter(train) [ 77500/160000] base_lr: 5.2051e-05 lr: 1.9244e-07 eta: 23:58:23 time: 0.9988 data_time: 0.0046 memory: 8457 loss: 0.0132 decode.loss_ce: 0.0059 decode.acc_seg: 99.7751 aux.loss_ce: 0.0073 aux.acc_seg: 99.2783 +04/19 05:55:33 - mmengine - INFO - Iter(train) [ 77550/160000] base_lr: 5.2019e-05 lr: 1.9233e-07 eta: 23:57:19 time: 0.9999 data_time: 0.0046 memory: 8457 loss: 0.0122 decode.loss_ce: 0.0054 decode.acc_seg: 99.7700 aux.loss_ce: 0.0068 aux.acc_seg: 99.3414 +04/19 05:56:23 - mmengine - INFO - Iter(train) [ 77600/160000] base_lr: 5.1988e-05 lr: 1.9221e-07 eta: 23:56:16 time: 0.9991 data_time: 0.0045 memory: 8457 loss: 0.0126 decode.loss_ce: 0.0055 decode.acc_seg: 99.7419 aux.loss_ce: 0.0071 aux.acc_seg: 99.2914 +04/19 05:57:13 - mmengine - INFO - Iter(train) [ 77650/160000] base_lr: 5.1956e-05 lr: 1.9209e-07 eta: 23:55:13 time: 1.0000 data_time: 0.0042 memory: 8457 loss: 0.0116 decode.loss_ce: 0.0049 decode.acc_seg: 99.8108 aux.loss_ce: 0.0066 aux.acc_seg: 99.4274 +04/19 05:58:03 - mmengine - INFO - Iter(train) [ 77700/160000] base_lr: 5.1925e-05 lr: 1.9198e-07 eta: 23:54:09 time: 0.9986 data_time: 0.0045 memory: 8457 loss: 0.0127 decode.loss_ce: 0.0056 decode.acc_seg: 99.7299 aux.loss_ce: 0.0071 aux.acc_seg: 99.0665 +04/19 05:58:53 - mmengine - INFO - Iter(train) [ 77750/160000] base_lr: 5.1893e-05 lr: 1.9186e-07 eta: 23:53:06 time: 0.9981 data_time: 0.0050 memory: 8457 loss: 0.0130 decode.loss_ce: 0.0059 decode.acc_seg: 99.6883 aux.loss_ce: 0.0071 aux.acc_seg: 99.0023 +04/19 05:59:43 - mmengine - INFO - Iter(train) [ 77800/160000] base_lr: 5.1862e-05 lr: 1.9174e-07 eta: 23:52:03 time: 0.9973 data_time: 0.0047 memory: 8457 loss: 0.0137 decode.loss_ce: 0.0064 decode.acc_seg: 99.7408 aux.loss_ce: 0.0073 aux.acc_seg: 99.2992 +04/19 06:00:33 - mmengine - INFO - Iter(train) [ 77850/160000] base_lr: 5.1830e-05 lr: 1.9163e-07 eta: 23:51:00 time: 1.0011 data_time: 0.0049 memory: 8457 loss: 0.0123 decode.loss_ce: 0.0057 decode.acc_seg: 99.7818 aux.loss_ce: 0.0066 aux.acc_seg: 99.3597 +04/19 06:01:23 - mmengine - INFO - Iter(train) [ 77900/160000] base_lr: 5.1798e-05 lr: 1.9151e-07 eta: 23:49:57 time: 0.9979 data_time: 0.0047 memory: 8457 loss: 0.0139 decode.loss_ce: 0.0060 decode.acc_seg: 99.7086 aux.loss_ce: 0.0079 aux.acc_seg: 99.0963 +04/19 06:02:13 - mmengine - INFO - Iter(train) [ 77950/160000] base_lr: 5.1767e-05 lr: 1.9139e-07 eta: 23:48:54 time: 0.9972 data_time: 0.0045 memory: 8457 loss: 0.0116 decode.loss_ce: 0.0052 decode.acc_seg: 99.7627 aux.loss_ce: 0.0064 aux.acc_seg: 99.4083 +04/19 06:03:03 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240419_004857 +04/19 06:03:03 - mmengine - INFO - Iter(train) [ 78000/160000] base_lr: 5.1735e-05 lr: 1.9128e-07 eta: 23:47:51 time: 0.9979 data_time: 0.0048 memory: 8457 loss: 0.0130 decode.loss_ce: 0.0056 decode.acc_seg: 99.7499 aux.loss_ce: 0.0075 aux.acc_seg: 99.2113 +04/19 06:03:53 - mmengine - INFO - Iter(train) [ 78050/160000] base_lr: 5.1704e-05 lr: 1.9116e-07 eta: 23:46:48 time: 0.9980 data_time: 0.0050 memory: 8457 loss: 0.0121 decode.loss_ce: 0.0054 decode.acc_seg: 99.6878 aux.loss_ce: 0.0067 aux.acc_seg: 99.0467 +04/19 06:04:43 - mmengine - INFO - Iter(train) [ 78100/160000] base_lr: 5.1672e-05 lr: 1.9104e-07 eta: 23:45:45 time: 0.9996 data_time: 0.0046 memory: 8457 loss: 0.0129 decode.loss_ce: 0.0056 decode.acc_seg: 99.7698 aux.loss_ce: 0.0073 aux.acc_seg: 99.2567 +04/19 06:05:33 - mmengine - INFO - Iter(train) [ 78150/160000] base_lr: 5.1641e-05 lr: 1.9093e-07 eta: 23:44:43 time: 0.9991 data_time: 0.0046 memory: 8457 loss: 0.0120 decode.loss_ce: 0.0054 decode.acc_seg: 99.7957 aux.loss_ce: 0.0066 aux.acc_seg: 99.4730 +04/19 06:06:23 - mmengine - INFO - Iter(train) [ 78200/160000] base_lr: 5.1609e-05 lr: 1.9081e-07 eta: 23:43:41 time: 0.9976 data_time: 0.0045 memory: 8457 loss: 0.0120 decode.loss_ce: 0.0055 decode.acc_seg: 99.7253 aux.loss_ce: 0.0065 aux.acc_seg: 99.3870 +04/19 06:07:13 - mmengine - INFO - Iter(train) [ 78250/160000] base_lr: 5.1578e-05 lr: 1.9069e-07 eta: 23:42:39 time: 0.9995 data_time: 0.0048 memory: 8457 loss: 0.0118 decode.loss_ce: 0.0051 decode.acc_seg: 99.7387 aux.loss_ce: 0.0067 aux.acc_seg: 99.0322 +04/19 06:08:03 - mmengine - INFO - Iter(train) [ 78300/160000] base_lr: 5.1546e-05 lr: 1.9058e-07 eta: 23:41:36 time: 0.9991 data_time: 0.0044 memory: 8457 loss: 0.0138 decode.loss_ce: 0.0062 decode.acc_seg: 99.7467 aux.loss_ce: 0.0077 aux.acc_seg: 99.2725 +04/19 06:08:53 - mmengine - INFO - Iter(train) [ 78350/160000] base_lr: 5.1515e-05 lr: 1.9046e-07 eta: 23:40:34 time: 0.9999 data_time: 0.0051 memory: 8457 loss: 0.0144 decode.loss_ce: 0.0066 decode.acc_seg: 99.7419 aux.loss_ce: 0.0078 aux.acc_seg: 99.2062 +04/19 06:09:43 - mmengine - INFO - Iter(train) [ 78400/160000] base_lr: 5.1483e-05 lr: 1.9034e-07 eta: 23:39:32 time: 0.9978 data_time: 0.0046 memory: 8457 loss: 0.0132 decode.loss_ce: 0.0057 decode.acc_seg: 99.6658 aux.loss_ce: 0.0075 aux.acc_seg: 99.2020 +04/19 06:10:32 - mmengine - INFO - Iter(train) [ 78450/160000] base_lr: 5.1451e-05 lr: 1.9023e-07 eta: 23:38:30 time: 0.9990 data_time: 0.0046 memory: 8457 loss: 0.0134 decode.loss_ce: 0.0059 decode.acc_seg: 99.7873 aux.loss_ce: 0.0075 aux.acc_seg: 99.1709 +04/19 06:11:22 - mmengine - INFO - Iter(train) [ 78500/160000] base_lr: 5.1420e-05 lr: 1.9011e-07 eta: 23:37:28 time: 0.9997 data_time: 0.0048 memory: 8457 loss: 0.0143 decode.loss_ce: 0.0064 decode.acc_seg: 99.7080 aux.loss_ce: 0.0079 aux.acc_seg: 99.2508 +04/19 06:12:12 - mmengine - INFO - Iter(train) [ 78550/160000] base_lr: 5.1388e-05 lr: 1.8999e-07 eta: 23:36:26 time: 0.9989 data_time: 0.0047 memory: 8457 loss: 0.0128 decode.loss_ce: 0.0054 decode.acc_seg: 99.7654 aux.loss_ce: 0.0074 aux.acc_seg: 99.3689 +04/19 06:13:02 - mmengine - INFO - Iter(train) [ 78600/160000] base_lr: 5.1357e-05 lr: 1.8988e-07 eta: 23:35:24 time: 0.9972 data_time: 0.0046 memory: 8457 loss: 0.0155 decode.loss_ce: 0.0067 decode.acc_seg: 99.7011 aux.loss_ce: 0.0087 aux.acc_seg: 98.7347 +04/19 06:13:52 - mmengine - INFO - Iter(train) [ 78650/160000] base_lr: 5.1325e-05 lr: 1.8976e-07 eta: 23:34:22 time: 1.0000 data_time: 0.0047 memory: 8457 loss: 0.0128 decode.loss_ce: 0.0059 decode.acc_seg: 99.6748 aux.loss_ce: 0.0069 aux.acc_seg: 99.0702 +04/19 06:14:42 - mmengine - INFO - Iter(train) [ 78700/160000] base_lr: 5.1294e-05 lr: 1.8964e-07 eta: 23:33:20 time: 0.9996 data_time: 0.0047 memory: 8457 loss: 0.0125 decode.loss_ce: 0.0058 decode.acc_seg: 99.7826 aux.loss_ce: 0.0066 aux.acc_seg: 99.4392 +04/19 06:15:32 - mmengine - INFO - Iter(train) [ 78750/160000] base_lr: 5.1262e-05 lr: 1.8953e-07 eta: 23:32:18 time: 0.9989 data_time: 0.0047 memory: 8457 loss: 0.0127 decode.loss_ce: 0.0056 decode.acc_seg: 99.7278 aux.loss_ce: 0.0070 aux.acc_seg: 99.1493 +04/19 06:16:22 - mmengine - INFO - Iter(train) [ 78800/160000] base_lr: 5.1231e-05 lr: 1.8941e-07 eta: 23:31:17 time: 0.9985 data_time: 0.0046 memory: 8457 loss: 0.0124 decode.loss_ce: 0.0057 decode.acc_seg: 99.7591 aux.loss_ce: 0.0067 aux.acc_seg: 99.1268 +04/19 06:17:12 - mmengine - INFO - Iter(train) [ 78850/160000] base_lr: 5.1199e-05 lr: 1.8929e-07 eta: 23:30:15 time: 0.9996 data_time: 0.0048 memory: 8457 loss: 0.0113 decode.loss_ce: 0.0050 decode.acc_seg: 99.6885 aux.loss_ce: 0.0063 aux.acc_seg: 98.7377 +04/19 06:18:02 - mmengine - INFO - Iter(train) [ 78900/160000] base_lr: 5.1168e-05 lr: 1.8918e-07 eta: 23:29:14 time: 0.9994 data_time: 0.0045 memory: 8457 loss: 0.0124 decode.loss_ce: 0.0056 decode.acc_seg: 99.6813 aux.loss_ce: 0.0068 aux.acc_seg: 99.1846 +04/19 06:18:52 - mmengine - INFO - Iter(train) [ 78950/160000] base_lr: 5.1136e-05 lr: 1.8906e-07 eta: 23:28:13 time: 1.0016 data_time: 0.0043 memory: 8457 loss: 0.0157 decode.loss_ce: 0.0069 decode.acc_seg: 99.7581 aux.loss_ce: 0.0088 aux.acc_seg: 99.3343 +04/19 06:19:42 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240419_004857 +04/19 06:19:42 - mmengine - INFO - Iter(train) [ 79000/160000] base_lr: 5.1104e-05 lr: 1.8894e-07 eta: 23:27:12 time: 1.0016 data_time: 0.0051 memory: 8457 loss: 0.0110 decode.loss_ce: 0.0051 decode.acc_seg: 99.8230 aux.loss_ce: 0.0059 aux.acc_seg: 99.5138 +04/19 06:20:32 - mmengine - INFO - Iter(train) [ 79050/160000] base_lr: 5.1073e-05 lr: 1.8883e-07 eta: 23:26:11 time: 1.0001 data_time: 0.0048 memory: 8457 loss: 0.0136 decode.loss_ce: 0.0061 decode.acc_seg: 99.7364 aux.loss_ce: 0.0075 aux.acc_seg: 98.9046 +04/19 06:21:22 - mmengine - INFO - Iter(train) [ 79100/160000] base_lr: 5.1041e-05 lr: 1.8871e-07 eta: 23:25:10 time: 1.0014 data_time: 0.0050 memory: 8457 loss: 0.0133 decode.loss_ce: 0.0058 decode.acc_seg: 99.6956 aux.loss_ce: 0.0075 aux.acc_seg: 99.1325 +04/19 06:22:12 - mmengine - INFO - Iter(train) [ 79150/160000] base_lr: 5.1010e-05 lr: 1.8859e-07 eta: 23:24:09 time: 1.0004 data_time: 0.0058 memory: 8457 loss: 0.0131 decode.loss_ce: 0.0055 decode.acc_seg: 99.7379 aux.loss_ce: 0.0076 aux.acc_seg: 98.9176 +04/19 06:23:02 - mmengine - INFO - Iter(train) [ 79200/160000] base_lr: 5.0978e-05 lr: 1.8848e-07 eta: 23:23:08 time: 1.0009 data_time: 0.0047 memory: 8457 loss: 0.0134 decode.loss_ce: 0.0059 decode.acc_seg: 99.6513 aux.loss_ce: 0.0075 aux.acc_seg: 98.9153 +04/19 06:23:52 - mmengine - INFO - Iter(train) [ 79250/160000] base_lr: 5.0947e-05 lr: 1.8836e-07 eta: 23:22:07 time: 0.9978 data_time: 0.0049 memory: 8457 loss: 0.0131 decode.loss_ce: 0.0058 decode.acc_seg: 99.7013 aux.loss_ce: 0.0073 aux.acc_seg: 99.3008 +04/19 06:24:42 - mmengine - INFO - Iter(train) [ 79300/160000] base_lr: 5.0915e-05 lr: 1.8824e-07 eta: 23:21:06 time: 0.9985 data_time: 0.0046 memory: 8457 loss: 0.0127 decode.loss_ce: 0.0061 decode.acc_seg: 99.7015 aux.loss_ce: 0.0066 aux.acc_seg: 99.2834 +04/19 06:25:32 - mmengine - INFO - Iter(train) [ 79350/160000] base_lr: 5.0884e-05 lr: 1.8813e-07 eta: 23:20:05 time: 0.9995 data_time: 0.0046 memory: 8457 loss: 0.0131 decode.loss_ce: 0.0058 decode.acc_seg: 99.7406 aux.loss_ce: 0.0072 aux.acc_seg: 99.1667 +04/19 06:26:22 - mmengine - INFO - Iter(train) [ 79400/160000] base_lr: 5.0852e-05 lr: 1.8801e-07 eta: 23:19:04 time: 0.9991 data_time: 0.0047 memory: 8457 loss: 0.0122 decode.loss_ce: 0.0054 decode.acc_seg: 99.8453 aux.loss_ce: 0.0069 aux.acc_seg: 99.5214 +04/19 06:27:12 - mmengine - INFO - Iter(train) [ 79450/160000] base_lr: 5.0821e-05 lr: 1.8789e-07 eta: 23:18:03 time: 0.9974 data_time: 0.0045 memory: 8457 loss: 0.0116 decode.loss_ce: 0.0052 decode.acc_seg: 99.7797 aux.loss_ce: 0.0064 aux.acc_seg: 99.2872 +04/19 06:28:02 - mmengine - INFO - Iter(train) [ 79500/160000] base_lr: 5.0789e-05 lr: 1.8778e-07 eta: 23:17:02 time: 0.9990 data_time: 0.0047 memory: 8457 loss: 0.0127 decode.loss_ce: 0.0054 decode.acc_seg: 99.7972 aux.loss_ce: 0.0073 aux.acc_seg: 99.3031 +04/19 06:28:52 - mmengine - INFO - Iter(train) [ 79550/160000] base_lr: 5.0757e-05 lr: 1.8766e-07 eta: 23:16:01 time: 0.9999 data_time: 0.0048 memory: 8457 loss: 0.0116 decode.loss_ce: 0.0052 decode.acc_seg: 99.8238 aux.loss_ce: 0.0064 aux.acc_seg: 99.3664 +04/19 06:29:42 - mmengine - INFO - Iter(train) [ 79600/160000] base_lr: 5.0726e-05 lr: 1.8754e-07 eta: 23:15:01 time: 0.9977 data_time: 0.0046 memory: 8457 loss: 0.0122 decode.loss_ce: 0.0052 decode.acc_seg: 99.8024 aux.loss_ce: 0.0070 aux.acc_seg: 99.2453 +04/19 06:30:32 - mmengine - INFO - Iter(train) [ 79650/160000] base_lr: 5.0694e-05 lr: 1.8743e-07 eta: 23:14:00 time: 0.9991 data_time: 0.0047 memory: 8457 loss: 0.0141 decode.loss_ce: 0.0059 decode.acc_seg: 99.7225 aux.loss_ce: 0.0082 aux.acc_seg: 98.9305 +04/19 06:31:22 - mmengine - INFO - Iter(train) [ 79700/160000] base_lr: 5.0663e-05 lr: 1.8731e-07 eta: 23:12:59 time: 0.9990 data_time: 0.0048 memory: 8457 loss: 0.0126 decode.loss_ce: 0.0058 decode.acc_seg: 99.6677 aux.loss_ce: 0.0069 aux.acc_seg: 99.2506 +04/19 06:32:12 - mmengine - INFO - Iter(train) [ 79750/160000] base_lr: 5.0631e-05 lr: 1.8719e-07 eta: 23:11:59 time: 1.0000 data_time: 0.0048 memory: 8457 loss: 0.0139 decode.loss_ce: 0.0062 decode.acc_seg: 99.7698 aux.loss_ce: 0.0077 aux.acc_seg: 99.1795 +04/19 06:33:02 - mmengine - INFO - Iter(train) [ 79800/160000] base_lr: 5.0600e-05 lr: 1.8708e-07 eta: 23:10:59 time: 0.9991 data_time: 0.0052 memory: 8457 loss: 0.0120 decode.loss_ce: 0.0056 decode.acc_seg: 99.6943 aux.loss_ce: 0.0063 aux.acc_seg: 99.2014 +04/19 06:33:52 - mmengine - INFO - Iter(train) [ 79850/160000] base_lr: 5.0568e-05 lr: 1.8696e-07 eta: 23:09:58 time: 0.9993 data_time: 0.0048 memory: 8457 loss: 0.0135 decode.loss_ce: 0.0060 decode.acc_seg: 99.6765 aux.loss_ce: 0.0076 aux.acc_seg: 98.8350 +04/19 06:34:42 - mmengine - INFO - Iter(train) [ 79900/160000] base_lr: 5.0537e-05 lr: 1.8684e-07 eta: 23:08:58 time: 0.9978 data_time: 0.0048 memory: 8457 loss: 0.0131 decode.loss_ce: 0.0058 decode.acc_seg: 99.7200 aux.loss_ce: 0.0073 aux.acc_seg: 99.0034 +04/19 06:35:32 - mmengine - INFO - Iter(train) [ 79950/160000] base_lr: 5.0505e-05 lr: 1.8673e-07 eta: 23:07:57 time: 0.9982 data_time: 0.0043 memory: 8457 loss: 0.0135 decode.loss_ce: 0.0058 decode.acc_seg: 99.7614 aux.loss_ce: 0.0077 aux.acc_seg: 99.4556 +04/19 06:36:22 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_cag-512x512_20240419_004857 +04/19 06:36:22 - mmengine - INFO - Iter(train) [ 80000/160000] base_lr: 5.0474e-05 lr: 1.8661e-07 eta: 23:06:57 time: 0.9971 data_time: 0.0045 memory: 8457 loss: 0.0134 decode.loss_ce: 0.0060 decode.acc_seg: 99.7063 aux.loss_ce: 0.0074 aux.acc_seg: 99.2573 +04/19 06:36:22 - mmengine - INFO - Saving checkpoint at 80000 iterations +04/19 06:36:32 - mmengine - INFO - Iter(val) [ 50/200] eta: 0:00:17 time: 0.1156 data_time: 0.0014 memory: 3999 +04/19 06:36:37 - mmengine - INFO - Iter(val) [100/200] eta: 0:00:11 time: 0.1158 data_time: 0.0015 memory: 3999 +04/19 06:36:43 - mmengine - INFO - Iter(val) [150/200] eta: 0:00:05 time: 0.1158 data_time: 0.0014 memory: 3999 +04/19 06:36:49 - mmengine - INFO - Iter(val) [200/200] eta: 0:00:00 time: 0.1160 data_time: 0.0016 memory: 3999 +04/19 06:36:49 - mmengine - INFO - per class results: +04/19 06:36:49 - mmengine - INFO - ++------------+-------+-------+--------+-----------+--------+ +| Class | IoU | Acc | Fscore | Precision | Recall | ++------------+-------+-------+--------+-----------+--------+ +| background | 99.16 | 99.51 | 99.58 | 99.64 | 99.51 | +| contrast | 81.8 | 91.37 | 89.99 | 88.65 | 91.37 | ++------------+-------+-------+--------+-----------+--------+ +04/19 06:36:49 - mmengine - INFO - Iter(val) [200/200] aAcc: 99.1900 mIoU: 90.4800 mAcc: 95.4400 mFscore: 94.7800 mPrecision: 94.1500 mRecall: 95.4400 data_time: 0.0015 time: 0.1160 +WARNING:torch.distributed.elastic.agent.server.api:Received 2 death signal, shutting down workers +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 64757 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 64758 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 64759 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 64760 closing signal SIGINT +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + model = self.train_loop.run() # type: ignore + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + self.run_iter(data_batch) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + outputs = self.runner.model.train_step( + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + optim_wrapper.update_params(parsed_loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 196, in update_params + self.backward(loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 220, in backward + loss.backward(**kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/_tensor.py", line 363, in backward + torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py", line 173, in backward + Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass +KeyboardInterrupt +Traceback (most recent call last): + File "tools/train.py", line 104, in +Traceback (most recent call last): + File "tools/train.py", line 104, in +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train +main()runner.train() + + File "tools/train.py", line 100, in main + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + model = self.train_loop.run() # type: ignore + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + self.run_iter(data_batch) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + outputs = self.runner.model.train_step( + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + model = self.train_loop.run() # type: ignoreoptim_wrapper.update_params(parsed_loss) + + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 196, in update_params + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + self.backward(loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 220, in backward + model = self.train_loop.run() # type: ignore + loss.backward(**kwargs) File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + + File "/opt/conda/lib/python3.8/site-packages/torch/_tensor.py", line 363, in backward + torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py", line 173, in backward + Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass +KeyboardInterrupt + self.run_iter(data_batch) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter +self.run_iter(data_batch) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + outputs = self.runner.model.train_step( + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + outputs = self.runner.model.train_step( + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + optim_wrapper.update_params(parsed_loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 196, in update_params + optim_wrapper.update_params(parsed_loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 196, in update_params + self.backward(loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 220, in backward + self.backward(loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 220, in backward +loss.backward(**kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/_tensor.py", line 363, in backward + loss.backward(**kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/_tensor.py", line 363, in backward + torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py", line 173, in backward + Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass +KeyboardInterrupt + torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py", line 173, in backward + Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass +KeyboardInterrupt +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 64757 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 64758 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 64759 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 64760 closing signal SIGTERM +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 709, in run + result = self._invoke_run(role) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 850, in _invoke_run + time.sleep(monitor_interval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 64727 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 716, in run + self._shutdown(e.sigval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 64727 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 236, in launch_agent + result = agent.run() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/metrics/api.py", line 125, in wrapper + result = f(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 721, in run + self._shutdown() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 64727 got signal: 2 +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/19 06:46:48 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 1862928832 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 1862928832 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/19 06:46:49 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag_10000' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +fp16 = dict(loss_scale='dynamic') +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = None +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=768, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + num_convs=1, + type='FCNHead'), + backbone=dict( + act_cfg=dict(type='GELU'), + attn_drop_rate=0.0, + drop_path_rate=0.1, + embed_dims=768, + img_size=( + 512, + 512, + ), + in_channels=3, + init_values=1.0, + mlp_ratio=4, + norm_cfg=dict(eps=1e-06, type='LN'), + norm_eval=False, + num_heads=12, + num_layers=12, + out_indices=[ + 3, + 5, + 7, + 11, + ], + patch_size=16, + type='MAE'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=768, + dropout_ratio=0.1, + in_channels=[ + 768, + 768, + 768, + 768, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + neck=dict( + embed_dim=768, rescales=[ + 4, + 2, + 1, + 0.5, + ], type='Feature2Pyramid'), + pretrained='/workspaces/mmsegmentation-1/converted_model.pth', + test_cfg=dict(crop_size=( + 512, + 512, + ), mode='slide', stride=( + 341, + 341, + )), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + constructor='LayerDecayOptimizerConstructor', + optimizer=dict( + betas=( + 0.9, + 0.999, + ), lr=0.0001, type='AdamW', weight_decay=0.05), + paramwise_cfg=dict(layer_decay_rate=0.65, num_layers=12), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, by_epoch=False, end=1500, start_factor=1e-06, + type='LinearLR'), + dict( + begin=1500, + by_epoch=False, + end=160000, + eta_min=0.0, + power=1.0, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag_10000', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=2, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag_10000', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag_10000', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = './work_dirs/mae-base_upernet_8xb2-amp-160k_cag-512x512' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/19 06:46:51 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +Param groups = { + "layer_0_decay": { + "param_names": [ + "backbone.cls_token", + "backbone.pos_embed", + "backbone.patch_embed.projection.weight" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.05 + }, + "layer_0_no_decay": { + "param_names": [ + "backbone.patch_embed.projection.bias" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.0 + }, + "layer_1_no_decay": { + "param_names": [ + "backbone.layers.0.gamma_1", + "backbone.layers.0.gamma_2", + "backbone.layers.0.ln1.weight", + "backbone.layers.0.ln1.bias", + "backbone.layers.0.attn.qkv.bias", + "backbone.layers.0.attn.proj.bias", + "backbone.layers.0.ln2.weight", + "backbone.layers.0.ln2.bias", + "backbone.layers.0.ffn.layers.0.0.bias", + "backbone.layers.0.ffn.layers.1.bias" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.0 + }, + "layer_1_decay": { + "param_names": [ + "backbone.layers.0.attn.relative_position_bias_table", + "backbone.layers.0.attn.qkv.weight", + "backbone.layers.0.attn.proj.weight", + "backbone.layers.0.ffn.layers.0.0.weight", + "backbone.layers.0.ffn.layers.1.weight" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.05 + }, + "layer_2_no_decay": { + "param_names": [ + "backbone.layers.1.gamma_1", + "backbone.layers.1.gamma_2", + "backbone.layers.1.ln1.weight", + "backbone.layers.1.ln1.bias", + "backbone.layers.1.attn.qkv.bias", + "backbone.layers.1.attn.proj.bias", + "backbone.layers.1.ln2.weight", + "backbone.layers.1.ln2.bias", + "backbone.layers.1.ffn.layers.0.0.bias", + "backbone.layers.1.ffn.layers.1.bias" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.0 + }, + "layer_2_decay": { + "param_names": [ + "backbone.layers.1.attn.relative_position_bias_table", + "backbone.layers.1.attn.qkv.weight", + "backbone.layers.1.attn.proj.weight", + "backbone.layers.1.ffn.layers.0.0.weight", + "backbone.layers.1.ffn.layers.1.weight" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.05 + }, + "layer_3_no_decay": { + "param_names": [ + "backbone.layers.2.gamma_1", + "backbone.layers.2.gamma_2", + "backbone.layers.2.ln1.weight", + "backbone.layers.2.ln1.bias", + "backbone.layers.2.attn.qkv.bias", + "backbone.layers.2.attn.proj.bias", + "backbone.layers.2.ln2.weight", + "backbone.layers.2.ln2.bias", + "backbone.layers.2.ffn.layers.0.0.bias", + "backbone.layers.2.ffn.layers.1.bias" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.0 + }, + "layer_3_decay": { + "param_names": [ + "backbone.layers.2.attn.relative_position_bias_table", + "backbone.layers.2.attn.qkv.weight", + "backbone.layers.2.attn.proj.weight", + "backbone.layers.2.ffn.layers.0.0.weight", + "backbone.layers.2.ffn.layers.1.weight" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.05 + }, + "layer_4_no_decay": { + "param_names": [ + "backbone.layers.3.gamma_1", + "backbone.layers.3.gamma_2", + "backbone.layers.3.ln1.weight", + "backbone.layers.3.ln1.bias", + "backbone.layers.3.attn.qkv.bias", + "backbone.layers.3.attn.proj.bias", + "backbone.layers.3.ln2.weight", + "backbone.layers.3.ln2.bias", + "backbone.layers.3.ffn.layers.0.0.bias", + "backbone.layers.3.ffn.layers.1.bias" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.0 + }, + "layer_4_decay": { + "param_names": [ + "backbone.layers.3.attn.relative_position_bias_table", + "backbone.layers.3.attn.qkv.weight", + "backbone.layers.3.attn.proj.weight", + "backbone.layers.3.ffn.layers.0.0.weight", + "backbone.layers.3.ffn.layers.1.weight" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.05 + }, + "layer_5_no_decay": { + "param_names": [ + "backbone.layers.4.gamma_1", + "backbone.layers.4.gamma_2", + "backbone.layers.4.ln1.weight", + "backbone.layers.4.ln1.bias", + "backbone.layers.4.attn.qkv.bias", + "backbone.layers.4.attn.proj.bias", + "backbone.layers.4.ln2.weight", + "backbone.layers.4.ln2.bias", + "backbone.layers.4.ffn.layers.0.0.bias", + "backbone.layers.4.ffn.layers.1.bias" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.0 + }, + "layer_5_decay": { + "param_names": [ + "backbone.layers.4.attn.relative_position_bias_table", + "backbone.layers.4.attn.qkv.weight", + "backbone.layers.4.attn.proj.weight", + "backbone.layers.4.ffn.layers.0.0.weight", + "backbone.layers.4.ffn.layers.1.weight" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.05 + }, + "layer_6_no_decay": { + "param_names": [ + "backbone.layers.5.gamma_1", + "backbone.layers.5.gamma_2", + "backbone.layers.5.ln1.weight", + "backbone.layers.5.ln1.bias", + "backbone.layers.5.attn.qkv.bias", + "backbone.layers.5.attn.proj.bias", + "backbone.layers.5.ln2.weight", + "backbone.layers.5.ln2.bias", + "backbone.layers.5.ffn.layers.0.0.bias", + "backbone.layers.5.ffn.layers.1.bias" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.0 + }, + "layer_6_decay": { + "param_names": [ + "backbone.layers.5.attn.relative_position_bias_table", + "backbone.layers.5.attn.qkv.weight", + "backbone.layers.5.attn.proj.weight", + "backbone.layers.5.ffn.layers.0.0.weight", + "backbone.layers.5.ffn.layers.1.weight" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.05 + }, + "layer_7_no_decay": { + "param_names": [ + "backbone.layers.6.gamma_1", + "backbone.layers.6.gamma_2", + "backbone.layers.6.ln1.weight", + "backbone.layers.6.ln1.bias", + "backbone.layers.6.attn.qkv.bias", + "backbone.layers.6.attn.proj.bias", + "backbone.layers.6.ln2.weight", + "backbone.layers.6.ln2.bias", + "backbone.layers.6.ffn.layers.0.0.bias", + "backbone.layers.6.ffn.layers.1.bias" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.0 + }, + "layer_7_decay": { + "param_names": [ + "backbone.layers.6.attn.relative_position_bias_table", + "backbone.layers.6.attn.qkv.weight", + "backbone.layers.6.attn.proj.weight", + "backbone.layers.6.ffn.layers.0.0.weight", + "backbone.layers.6.ffn.layers.1.weight" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.05 + }, + "layer_8_no_decay": { + "param_names": [ + "backbone.layers.7.gamma_1", + "backbone.layers.7.gamma_2", + "backbone.layers.7.ln1.weight", + "backbone.layers.7.ln1.bias", + "backbone.layers.7.attn.qkv.bias", + "backbone.layers.7.attn.proj.bias", + "backbone.layers.7.ln2.weight", + "backbone.layers.7.ln2.bias", + "backbone.layers.7.ffn.layers.0.0.bias", + "backbone.layers.7.ffn.layers.1.bias" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.0 + }, + "layer_8_decay": { + "param_names": [ + "backbone.layers.7.attn.relative_position_bias_table", + "backbone.layers.7.attn.qkv.weight", + "backbone.layers.7.attn.proj.weight", + "backbone.layers.7.ffn.layers.0.0.weight", + "backbone.layers.7.ffn.layers.1.weight" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.05 + }, + "layer_9_no_decay": { + "param_names": [ + "backbone.layers.8.gamma_1", + "backbone.layers.8.gamma_2", + "backbone.layers.8.ln1.weight", + "backbone.layers.8.ln1.bias", + "backbone.layers.8.attn.qkv.bias", + "backbone.layers.8.attn.proj.bias", + "backbone.layers.8.ln2.weight", + "backbone.layers.8.ln2.bias", + "backbone.layers.8.ffn.layers.0.0.bias", + "backbone.layers.8.ffn.layers.1.bias" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.0 + }, + "layer_9_decay": { + "param_names": [ + "backbone.layers.8.attn.relative_position_bias_table", + "backbone.layers.8.attn.qkv.weight", + "backbone.layers.8.attn.proj.weight", + "backbone.layers.8.ffn.layers.0.0.weight", + "backbone.layers.8.ffn.layers.1.weight" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.05 + }, + "layer_10_no_decay": { + "param_names": [ + "backbone.layers.9.gamma_1", + "backbone.layers.9.gamma_2", + "backbone.layers.9.ln1.weight", + "backbone.layers.9.ln1.bias", + "backbone.layers.9.attn.qkv.bias", + "backbone.layers.9.attn.proj.bias", + "backbone.layers.9.ln2.weight", + "backbone.layers.9.ln2.bias", + "backbone.layers.9.ffn.layers.0.0.bias", + "backbone.layers.9.ffn.layers.1.bias" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.0 + }, + "layer_10_decay": { + "param_names": [ + "backbone.layers.9.attn.relative_position_bias_table", + "backbone.layers.9.attn.qkv.weight", + "backbone.layers.9.attn.proj.weight", + "backbone.layers.9.ffn.layers.0.0.weight", + "backbone.layers.9.ffn.layers.1.weight" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.05 + }, + "layer_11_no_decay": { + "param_names": [ + "backbone.layers.10.gamma_1", + "backbone.layers.10.gamma_2", + "backbone.layers.10.ln1.weight", + "backbone.layers.10.ln1.bias", + "backbone.layers.10.attn.qkv.bias", + "backbone.layers.10.attn.proj.bias", + "backbone.layers.10.ln2.weight", + "backbone.layers.10.ln2.bias", + "backbone.layers.10.ffn.layers.0.0.bias", + "backbone.layers.10.ffn.layers.1.bias" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.0 + }, + "layer_11_decay": { + "param_names": [ + "backbone.layers.10.attn.relative_position_bias_table", + "backbone.layers.10.attn.qkv.weight", + "backbone.layers.10.attn.proj.weight", + "backbone.layers.10.ffn.layers.0.0.weight", + "backbone.layers.10.ffn.layers.1.weight" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.05 + }, + "layer_12_no_decay": { + "param_names": [ + "backbone.layers.11.gamma_1", + "backbone.layers.11.gamma_2", + "backbone.layers.11.ln1.weight", + "backbone.layers.11.ln1.bias", + "backbone.layers.11.attn.qkv.bias", + "backbone.layers.11.attn.proj.bias", + "backbone.layers.11.ln2.weight", + "backbone.layers.11.ln2.bias", + "backbone.layers.11.ffn.layers.0.0.bias", + "backbone.layers.11.ffn.layers.1.bias" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.0 + }, + "layer_12_decay": { + "param_names": [ + "backbone.layers.11.attn.relative_position_bias_table", + "backbone.layers.11.attn.qkv.weight", + "backbone.layers.11.attn.proj.weight", + "backbone.layers.11.ffn.layers.0.0.weight", + "backbone.layers.11.ffn.layers.1.weight" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.05 + }, + "layer_13_decay": { + "param_names": [ + "neck.upsample_4x.0.weight", + "neck.upsample_4x.3.weight", + "neck.upsample_2x.0.weight", + "decode_head.conv_seg.weight", + "decode_head.psp_modules.0.1.conv.weight", + "decode_head.psp_modules.1.1.conv.weight", + "decode_head.psp_modules.2.1.conv.weight", + "decode_head.psp_modules.3.1.conv.weight", + "decode_head.bottleneck.conv.weight", + "decode_head.lateral_convs.0.conv.weight", + "decode_head.lateral_convs.1.conv.weight", + "decode_head.lateral_convs.2.conv.weight", + "decode_head.fpn_convs.0.conv.weight", + "decode_head.fpn_convs.1.conv.weight", + "decode_head.fpn_convs.2.conv.weight", + "decode_head.fpn_bottleneck.conv.weight", + "auxiliary_head.conv_seg.weight", + "auxiliary_head.convs.0.conv.weight" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.05 + }, + "layer_13_no_decay": { + "param_names": [ + "neck.upsample_4x.0.bias", + "neck.upsample_4x.1.weight", + "neck.upsample_4x.1.bias", + "neck.upsample_4x.3.bias", + "neck.upsample_2x.0.bias", + "decode_head.conv_seg.bias", + "decode_head.psp_modules.0.1.bn.weight", + "decode_head.psp_modules.0.1.bn.bias", + "decode_head.psp_modules.1.1.bn.weight", + "decode_head.psp_modules.1.1.bn.bias", + "decode_head.psp_modules.2.1.bn.weight", + "decode_head.psp_modules.2.1.bn.bias", + "decode_head.psp_modules.3.1.bn.weight", + "decode_head.psp_modules.3.1.bn.bias", + "decode_head.bottleneck.bn.weight", + "decode_head.bottleneck.bn.bias", + "decode_head.lateral_convs.0.bn.weight", + "decode_head.lateral_convs.0.bn.bias", + "decode_head.lateral_convs.1.bn.weight", + "decode_head.lateral_convs.1.bn.bias", + "decode_head.lateral_convs.2.bn.weight", + "decode_head.lateral_convs.2.bn.bias", + "decode_head.fpn_convs.0.bn.weight", + "decode_head.fpn_convs.0.bn.bias", + "decode_head.fpn_convs.1.bn.weight", + "decode_head.fpn_convs.1.bn.bias", + "decode_head.fpn_convs.2.bn.weight", + "decode_head.fpn_convs.2.bn.bias", + "decode_head.fpn_bottleneck.bn.weight", + "decode_head.fpn_bottleneck.bn.bias", + "auxiliary_head.conv_seg.bias", + "auxiliary_head.convs.0.bn.weight", + "auxiliary_head.convs.0.bn.bias" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.0 + } +} +04/19 06:46:52 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/converted_model.pth +04/19 06:46:53 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +04/19 06:46:53 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +04/19 06:46:53 - mmengine - INFO - Checkpoints will be saved to /workspaces/mmsegmentation-1/work_dirs/mae-base_upernet_8xb2-amp-160k_cag-512x512. +04/19 06:47:50 - mmengine - INFO - Iter(train) [ 50/160000] base_lr: 3.2689e-06 lr: 1.2086e-08 eta: 2 days, 2:31:49 time: 1.0291 data_time: 0.0041 memory: 8935 loss: 6.8105 decode.loss_ce: 4.8183 decode.acc_seg: 19.2308 aux.loss_ce: 1.9922 aux.acc_seg: 0.0448 +04/19 06:48:41 - mmengine - INFO - Iter(train) [ 100/160000] base_lr: 6.6045e-06 lr: 2.4418e-08 eta: 2 days, 0:07:14 time: 1.0283 data_time: 0.0045 memory: 8462 loss: 6.4806 decode.loss_ce: 4.5186 decode.acc_seg: 27.7260 aux.loss_ce: 1.9619 aux.acc_seg: 15.1159 +WARNING:torch.distributed.elastic.agent.server.api:Received 2 death signal, shutting down workers +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 303901 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 303902 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 303903 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 303904 closing signal SIGINT +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + model = self.train_loop.run() # type: ignore + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + self.run_iter(data_batch) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + outputs = self.runner.model.train_step( + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + optim_wrapper.update_params(parsed_loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 196, in update_params + self.backward(loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 220, in backward + loss.backward(**kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/_tensor.py", line 363, in backward + torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py", line 173, in backward + Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass +KeyboardInterrupt +Traceback (most recent call last): + File "tools/train.py", line 104, in +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train +Traceback (most recent call last): + File "tools/train.py", line 104, in + model = self.train_loop.run() # type: ignore + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + self.run_iter(data_batch) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + outputs = self.runner.model.train_step( + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + main()optim_wrapper.update_params(parsed_loss) + + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 196, in update_params + File "tools/train.py", line 100, in main + main()self.backward(loss) + + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 220, in backward + File "tools/train.py", line 100, in main + loss.backward(**kwargs)runner.train() + + File "/opt/conda/lib/python3.8/site-packages/torch/_tensor.py", line 363, in backward + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + runner.train() + torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1777, in train + + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py", line 173, in backward + Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass +KeyboardInterrupt + model = self.train_loop.run() # type: ignore + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + model = self.train_loop.run() # type: ignore + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 286, in run + self.run_iter(data_batch) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + outputs = self.runner.model.train_step( + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step +self.run_iter(data_batch) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/loops.py", line 309, in run_iter + optim_wrapper.update_params(parsed_loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 196, in update_params + outputs = self.runner.model.train_step( + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/wrappers/distributed.py", line 123, in train_step + self.backward(loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 220, in backward + optim_wrapper.update_params(parsed_loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 196, in update_params + loss.backward(**kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/_tensor.py", line 363, in backward + self.backward(loss) + File "/opt/conda/lib/python3.8/site-packages/mmengine/optim/optimizer/optimizer_wrapper.py", line 220, in backward + torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) +loss.backward(**kwargs) File "/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py", line 173, in backward + + File "/opt/conda/lib/python3.8/site-packages/torch/_tensor.py", line 363, in backward + Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass +KeyboardInterrupt + torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) + File "/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py", line 173, in backward + Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass +KeyboardInterrupt +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 303901 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 303902 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 303903 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 303904 closing signal SIGTERM +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 709, in run + result = self._invoke_run(role) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 850, in _invoke_run + time.sleep(monitor_interval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 303871 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 716, in run + self._shutdown(e.sigval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 303871 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 236, in launch_agent + result = agent.run() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/metrics/api.py", line 125, in wrapper + result = f(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 721, in run + self._shutdown() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 303871 got signal: 2 +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/19 06:52:22 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 1322765965 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 1322765965 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/19 06:52:23 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag_10000' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +fp16 = dict(loss_scale='dynamic') +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = None +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=768, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + num_convs=1, + type='FCNHead'), + backbone=dict( + act_cfg=dict(type='GELU'), + attn_drop_rate=0.0, + drop_path_rate=0.1, + embed_dims=768, + img_size=( + 512, + 512, + ), + in_channels=3, + init_values=1.0, + mlp_ratio=4, + norm_cfg=dict(eps=1e-06, type='LN'), + norm_eval=False, + num_heads=12, + num_layers=12, + out_indices=[ + 3, + 5, + 7, + 11, + ], + patch_size=16, + type='MAE'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=768, + dropout_ratio=0.1, + in_channels=[ + 768, + 768, + 768, + 768, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + neck=dict( + embed_dim=768, rescales=[ + 4, + 2, + 1, + 0.5, + ], type='Feature2Pyramid'), + pretrained='./pretrain/mae_pretrain_vit_base_mmcls.pth', + test_cfg=dict(crop_size=( + 512, + 512, + ), mode='slide', stride=( + 341, + 341, + )), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + constructor='LayerDecayOptimizerConstructor', + optimizer=dict( + betas=( + 0.9, + 0.999, + ), lr=0.0001, type='AdamW', weight_decay=0.05), + paramwise_cfg=dict(layer_decay_rate=0.65, num_layers=12), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, by_epoch=False, end=1500, start_factor=1e-06, + type='LinearLR'), + dict( + begin=1500, + by_epoch=False, + end=160000, + eta_min=0.0, + power=1.0, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag_10000', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=2, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag_10000', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag_10000', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = './work_dirs/mae-base_upernet_8xb2-amp-160k_ade20k-512x512' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/19 06:52:25 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} + +set param backbone.layers.0.gamma_2 as id 1 +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1set param backbone.cls_token as id 0 + +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.pos_embed as id 0 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1set param backbone.patch_embed.projection.bias as id 0 + +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2set param backbone.layers.0.attn.qkv.weight as id 1 + +set param backbone.layers.1.attn.qkv.bias as id 2set param backbone.layers.0.attn.qkv.bias as id 1 + +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.0.attn.proj.weight as id 1set param backbone.layers.1.attn.proj.bias as id 2 + +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2set param backbone.layers.0.ln2.weight as id 1 + +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1set param backbone.layers.1.ffn.layers.1.weight as id 2 + +set param backbone.layers.0.ffn.layers.0.0.bias as id 1set param backbone.layers.1.ffn.layers.1.bias as id 2 + +set param backbone.layers.2.gamma_1 as id 3set param backbone.layers.0.ffn.layers.1.weight as id 1 + +set param backbone.layers.2.gamma_2 as id 3set param backbone.layers.0.ffn.layers.1.bias as id 1 + +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.1.gamma_1 as id 2set param backbone.layers.2.attn.relative_position_bias_table as id 3 + +set param backbone.layers.1.gamma_2 as id 2set param backbone.layers.2.attn.qkv.weight as id 3 + +set param backbone.layers.1.ln1.weight as id 2set param backbone.layers.2.attn.qkv.bias as id 3 + +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.1.attn.qkv.weight as id 2set param backbone.layers.2.ln2.weight as id 3 + +set param backbone.layers.1.attn.qkv.bias as id 2set param backbone.layers.2.ln2.bias as id 3 + +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2set param backbone.layers.3.gamma_1 as id 4 + +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.1.ffn.layers.1.weight as id 2set param backbone.layers.3.ln1.weight as id 4 + +set param backbone.layers.1.ffn.layers.1.bias as id 2set param backbone.layers.3.ln1.bias as id 4 + +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.3.attn.proj.weight as id 4set param backbone.layers.2.ln1.bias as id 3 + +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4set param backbone.layers.2.attn.proj.weight as id 3 + +set param backbone.layers.3.ffn.layers.0.0.bias as id 4set param backbone.layers.2.attn.proj.bias as id 3 + +set param backbone.layers.2.ln2.weight as id 3set param backbone.layers.3.ffn.layers.1.weight as id 4 + +set param backbone.layers.2.ln2.bias as id 3set param backbone.layers.3.ffn.layers.1.bias as id 4 + +set param backbone.layers.2.ffn.layers.0.0.weight as id 3set param backbone.layers.4.gamma_1 as id 5 + +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.2.ffn.layers.1.weight as id 3set param backbone.layers.4.ln1.weight as id 5 + +set param backbone.layers.2.ffn.layers.1.bias as id 3set param backbone.layers.4.ln1.bias as id 5 + +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.4.attn.qkv.weight as id 5set param backbone.layers.3.gamma_2 as id 4 + +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4set param backbone.layers.4.attn.proj.weight as id 5 + +set param backbone.layers.4.attn.proj.bias as id 5set param backbone.layers.3.attn.relative_position_bias_table as id 4 + +set param backbone.layers.4.ln2.weight as id 5set param backbone.layers.3.attn.qkv.weight as id 4 + +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4set param backbone.layers.4.ffn.layers.0.0.weight as id 5 + +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.4.ffn.layers.1.weight as id 5set param backbone.layers.3.ln2.bias as id 4 + +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.5.ln1.weight as id 6set param backbone.layers.3.ffn.layers.1.bias as id 4 + +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.5.attn.qkv.bias as id 6set param backbone.layers.4.ln1.bias as id 5 + +set param backbone.layers.4.attn.relative_position_bias_table as id 5set param backbone.layers.5.attn.proj.weight as id 6 + +set param backbone.layers.5.attn.proj.bias as id 6set param backbone.layers.4.attn.qkv.weight as id 5 + +set param backbone.layers.4.attn.qkv.bias as id 5set param backbone.layers.5.ln2.weight as id 6 + +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.4.ln2.weight as id 5set param backbone.layers.5.ffn.layers.0.0.bias as id 6 + +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5set param backbone.layers.6.gamma_1 as id 7 + +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7set param backbone.layers.5.gamma_1 as id 6 + +set param backbone.layers.6.attn.qkv.weight as id 7set param backbone.layers.5.gamma_2 as id 6 + +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7set param backbone.layers.6.gamma_2 as id 7 + +set param backbone.layers.6.ffn.layers.0.0.bias as id 7set param backbone.layers.6.ln1.weight as id 7 + +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.6.attn.proj.weight as id 7set param backbone.layers.7.gamma_2 as id 8 + +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9set param backbone.layers.6.ffn.layers.0.0.bias as id 7 + +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.7.gamma_1 as id 8set param backbone.layers.8.attn.proj.weight as id 9 + +set param backbone.layers.8.attn.proj.bias as id 9set param backbone.layers.7.gamma_2 as id 8 + +set param backbone.layers.7.ln1.weight as id 8set param backbone.layers.8.ln2.weight as id 9 + +set param backbone.layers.7.ln1.bias as id 8set param backbone.layers.8.ln2.bias as id 9 + +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.7.attn.qkv.weight as id 8set param backbone.layers.8.ffn.layers.0.0.bias as id 9 + +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.9.gamma_1 as id 10set param backbone.layers.7.ln2.weight as id 8 + +set param backbone.layers.9.gamma_2 as id 10set param backbone.layers.7.ln2.bias as id 8 + +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8set param backbone.layers.9.ln1.bias as id 10 + +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10set param backbone.layers.8.attn.relative_position_bias_table as id 9 + +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.9.ffn.layers.1.weight as id 10set param backbone.layers.8.attn.qkv.bias as id 9 + +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9set param backbone.layers.10.gamma_1 as id 11 + +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.10.attn.qkv.weight as id 11set param backbone.layers.8.ffn.layers.0.0.bias as id 9 + +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.9.gamma_1 as id 10set param backbone.layers.10.ln2.weight as id 11 + +set param backbone.layers.9.gamma_2 as id 10set param backbone.layers.10.ln2.bias as id 11 + +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10set param backbone.layers.10.ffn.layers.0.0.weight as id 11 + +set param backbone.layers.9.attn.relative_position_bias_table as id 10set param backbone.layers.10.ffn.layers.0.0.bias as id 11 + +set param backbone.layers.10.ffn.layers.1.weight as id 11set param backbone.layers.9.attn.qkv.weight as id 10 + +set param backbone.layers.10.ffn.layers.1.bias as id 11set param backbone.layers.9.attn.qkv.bias as id 10 + +set param backbone.layers.9.attn.proj.weight as id 10set param backbone.layers.11.gamma_1 as id 12 + +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.9.ln2.weight as id 10set param backbone.layers.11.ln1.weight as id 12 + +set param backbone.layers.9.ln2.bias as id 10set param backbone.layers.11.ln1.bias as id 12 + +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.11.attn.qkv.weight as id 12set param backbone.layers.9.ffn.layers.0.0.bias as id 10 + +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.11.attn.proj.weight as id 12set param backbone.layers.9.ffn.layers.1.bias as id 10 + +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.10.gamma_1 as id 11set param backbone.layers.11.ln2.weight as id 12 + +set param backbone.layers.10.gamma_2 as id 11set param backbone.layers.11.ln2.bias as id 12 + +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12set param backbone.layers.10.ln1.bias as id 11 + +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param neck.upsample_4x.0.weight as id 13set param backbone.layers.10.ln2.weight as id 11 + +set param neck.upsample_4x.0.bias as id 13set param backbone.layers.10.ln2.bias as id 11 + +set param neck.upsample_4x.1.weight as id 13set param backbone.layers.10.ffn.layers.0.0.weight as id 11 + +set param neck.upsample_4x.1.bias as id 13set param backbone.layers.10.ffn.layers.0.0.bias as id 11 + +set param neck.upsample_4x.3.weight as id 13 +set param backbone.layers.10.ffn.layers.1.weight as id 11set param neck.upsample_4x.3.bias as id 13 + +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param decode_head.conv_seg.weight as id 13 +set param backbone.layers.11.attn.relative_position_bias_table as id 12set param decode_head.conv_seg.bias as id 13 + +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param backbone.layers.11.attn.proj.weight as id 12 +set param decode_head.psp_modules.0.1.bn.weight as id 13set param backbone.layers.11.attn.proj.bias as id 12 + +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param decode_head.psp_modules.2.1.conv.weight as id 13set param backbone.layers.11.ffn.layers.1.bias as id 12 + +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param neck.upsample_4x.0.weight as id 13set param decode_head.psp_modules.3.1.conv.weight as id 13 + +set param neck.upsample_4x.0.bias as id 13set param decode_head.psp_modules.3.1.bn.weight as id 13 + +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param decode_head.bottleneck.conv.weight as id 13set param neck.upsample_4x.1.bias as id 13 + +set param decode_head.bottleneck.bn.weight as id 13set param neck.upsample_4x.3.weight as id 13 + +set param decode_head.bottleneck.bn.bias as id 13set param neck.upsample_4x.3.bias as id 13 + +set param neck.upsample_2x.0.weight as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13set param decode_head.psp_modules.0.1.bn.weight as id 13 + +set param decode_head.fpn_convs.0.bn.weight as id 13set param decode_head.psp_modules.0.1.bn.bias as id 13 + +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13set param decode_head.fpn_convs.2.bn.weight as id 13 + +set param decode_head.psp_modules.2.1.bn.weight as id 13set param decode_head.fpn_convs.2.bn.bias as id 13 + +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13set param decode_head.fpn_bottleneck.bn.weight as id 13 + +set param decode_head.psp_modules.3.1.bn.weight as id 13set param decode_head.fpn_bottleneck.bn.bias as id 13 + +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13set param auxiliary_head.conv_seg.bias as id 13 + +set param decode_head.bottleneck.bn.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13set param decode_head.lateral_convs.0.conv.weight as id 13 + +set param auxiliary_head.convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +Param groups = { + "layer_0_decay": { + "param_names": [ + "backbone.cls_token", + "backbone.pos_embed", + "backbone.patch_embed.projection.weight" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.05 + }, + "layer_0_no_decay": { + "param_names": [ + "backbone.patch_embed.projection.bias" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.0 + }, + "layer_1_no_decay": { + "param_names": [ + "backbone.layers.0.gamma_1", + "backbone.layers.0.gamma_2", + "backbone.layers.0.ln1.weight", + "backbone.layers.0.ln1.bias", + "backbone.layers.0.attn.qkv.bias", + "backbone.layers.0.attn.proj.bias", + "backbone.layers.0.ln2.weight", + "backbone.layers.0.ln2.bias", + "backbone.layers.0.ffn.layers.0.0.bias", + "backbone.layers.0.ffn.layers.1.bias" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.0 + }, + "layer_1_decay": { + "param_names": [ + "backbone.layers.0.attn.relative_position_bias_table", + "backbone.layers.0.attn.qkv.weight", + "backbone.layers.0.attn.proj.weight", + "backbone.layers.0.ffn.layers.0.0.weight", + "backbone.layers.0.ffn.layers.1.weight" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.05 + }, + "layer_2_no_decay": { + "param_names": [ + "backbone.layers.1.gamma_1", + "backbone.layers.1.gamma_2", + "backbone.layers.1.ln1.weight", + "backbone.layers.1.ln1.bias", + "backbone.layers.1.attn.qkv.bias", + "backbone.layers.1.attn.proj.bias", + "backbone.layers.1.ln2.weight", + "backbone.layers.1.ln2.bias", + "backbone.layers.1.ffn.layers.0.0.bias", + "backbone.layers.1.ffn.layers.1.bias" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.0 + }, + "layer_2_decay": { + "param_names": [ + "backbone.layers.1.attn.relative_position_bias_table", + "backbone.layers.1.attn.qkv.weight", + "backbone.layers.1.attn.proj.weight", + "backbone.layers.1.ffn.layers.0.0.weight", + "backbone.layers.1.ffn.layers.1.weight" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.05 + }, + "layer_3_no_decay": { + "param_names": [ + "backbone.layers.2.gamma_1", + "backbone.layers.2.gamma_2", + "backbone.layers.2.ln1.weight", + "backbone.layers.2.ln1.bias", + "backbone.layers.2.attn.qkv.bias", + "backbone.layers.2.attn.proj.bias", + "backbone.layers.2.ln2.weight", + "backbone.layers.2.ln2.bias", + "backbone.layers.2.ffn.layers.0.0.bias", + "backbone.layers.2.ffn.layers.1.bias" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.0 + }, + "layer_3_decay": { + "param_names": [ + "backbone.layers.2.attn.relative_position_bias_table", + "backbone.layers.2.attn.qkv.weight", + "backbone.layers.2.attn.proj.weight", + "backbone.layers.2.ffn.layers.0.0.weight", + "backbone.layers.2.ffn.layers.1.weight" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.05 + }, + "layer_4_no_decay": { + "param_names": [ + "backbone.layers.3.gamma_1", + "backbone.layers.3.gamma_2", + "backbone.layers.3.ln1.weight", + "backbone.layers.3.ln1.bias", + "backbone.layers.3.attn.qkv.bias", + "backbone.layers.3.attn.proj.bias", + "backbone.layers.3.ln2.weight", + "backbone.layers.3.ln2.bias", + "backbone.layers.3.ffn.layers.0.0.bias", + "backbone.layers.3.ffn.layers.1.bias" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.0 + }, + "layer_4_decay": { + "param_names": [ + "backbone.layers.3.attn.relative_position_bias_table", + "backbone.layers.3.attn.qkv.weight", + "backbone.layers.3.attn.proj.weight", + "backbone.layers.3.ffn.layers.0.0.weight", + "backbone.layers.3.ffn.layers.1.weight" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.05 + }, + "layer_5_no_decay": { + "param_names": [ + "backbone.layers.4.gamma_1", + "backbone.layers.4.gamma_2", + "backbone.layers.4.ln1.weight", + "backbone.layers.4.ln1.bias", + "backbone.layers.4.attn.qkv.bias", + "backbone.layers.4.attn.proj.bias", + "backbone.layers.4.ln2.weight", + "backbone.layers.4.ln2.bias", + "backbone.layers.4.ffn.layers.0.0.bias", + "backbone.layers.4.ffn.layers.1.bias" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.0 + }, + "layer_5_decay": { + "param_names": [ + "backbone.layers.4.attn.relative_position_bias_table", + "backbone.layers.4.attn.qkv.weight", + "backbone.layers.4.attn.proj.weight", + "backbone.layers.4.ffn.layers.0.0.weight", + "backbone.layers.4.ffn.layers.1.weight" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.05 + }, + "layer_6_no_decay": { + "param_names": [ + "backbone.layers.5.gamma_1", + "backbone.layers.5.gamma_2", + "backbone.layers.5.ln1.weight", + "backbone.layers.5.ln1.bias", + "backbone.layers.5.attn.qkv.bias", + "backbone.layers.5.attn.proj.bias", + "backbone.layers.5.ln2.weight", + "backbone.layers.5.ln2.bias", + "backbone.layers.5.ffn.layers.0.0.bias", + "backbone.layers.5.ffn.layers.1.bias" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.0 + }, + "layer_6_decay": { + "param_names": [ + "backbone.layers.5.attn.relative_position_bias_table", + "backbone.layers.5.attn.qkv.weight", + "backbone.layers.5.attn.proj.weight", + "backbone.layers.5.ffn.layers.0.0.weight", + "backbone.layers.5.ffn.layers.1.weight" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.05 + }, + "layer_7_no_decay": { + "param_names": [ + "backbone.layers.6.gamma_1", + "backbone.layers.6.gamma_2", + "backbone.layers.6.ln1.weight", + "backbone.layers.6.ln1.bias", + "backbone.layers.6.attn.qkv.bias", + "backbone.layers.6.attn.proj.bias", + "backbone.layers.6.ln2.weight", + "backbone.layers.6.ln2.bias", + "backbone.layers.6.ffn.layers.0.0.bias", + "backbone.layers.6.ffn.layers.1.bias" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.0 + }, + "layer_7_decay": { + "param_names": [ + "backbone.layers.6.attn.relative_position_bias_table", + "backbone.layers.6.attn.qkv.weight", + "backbone.layers.6.attn.proj.weight", + "backbone.layers.6.ffn.layers.0.0.weight", + "backbone.layers.6.ffn.layers.1.weight" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.05 + }, + "layer_8_no_decay": { + "param_names": [ + "backbone.layers.7.gamma_1", + "backbone.layers.7.gamma_2", + "backbone.layers.7.ln1.weight", + "backbone.layers.7.ln1.bias", + "backbone.layers.7.attn.qkv.bias", + "backbone.layers.7.attn.proj.bias", + "backbone.layers.7.ln2.weight", + "backbone.layers.7.ln2.bias", + "backbone.layers.7.ffn.layers.0.0.bias", + "backbone.layers.7.ffn.layers.1.bias" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.0 + }, + "layer_8_decay": { + "param_names": [ + "backbone.layers.7.attn.relative_position_bias_table", + "backbone.layers.7.attn.qkv.weight", + "backbone.layers.7.attn.proj.weight", + "backbone.layers.7.ffn.layers.0.0.weight", + "backbone.layers.7.ffn.layers.1.weight" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.05 + }, + "layer_9_no_decay": { + "param_names": [ + "backbone.layers.8.gamma_1", + "backbone.layers.8.gamma_2", + "backbone.layers.8.ln1.weight", + "backbone.layers.8.ln1.bias", + "backbone.layers.8.attn.qkv.bias", + "backbone.layers.8.attn.proj.bias", + "backbone.layers.8.ln2.weight", + "backbone.layers.8.ln2.bias", + "backbone.layers.8.ffn.layers.0.0.bias", + "backbone.layers.8.ffn.layers.1.bias" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.0 + }, + "layer_9_decay": { + "param_names": [ + "backbone.layers.8.attn.relative_position_bias_table", + "backbone.layers.8.attn.qkv.weight", + "backbone.layers.8.attn.proj.weight", + "backbone.layers.8.ffn.layers.0.0.weight", + "backbone.layers.8.ffn.layers.1.weight" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.05 + }, + "layer_10_no_decay": { + "param_names": [ + "backbone.layers.9.gamma_1", + "backbone.layers.9.gamma_2", + "backbone.layers.9.ln1.weight", + "backbone.layers.9.ln1.bias", + "backbone.layers.9.attn.qkv.bias", + "backbone.layers.9.attn.proj.bias", + "backbone.layers.9.ln2.weight", + "backbone.layers.9.ln2.bias", + "backbone.layers.9.ffn.layers.0.0.bias", + "backbone.layers.9.ffn.layers.1.bias" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.0 + }, + "layer_10_decay": { + "param_names": [ + "backbone.layers.9.attn.relative_position_bias_table", + "backbone.layers.9.attn.qkv.weight", + "backbone.layers.9.attn.proj.weight", + "backbone.layers.9.ffn.layers.0.0.weight", + "backbone.layers.9.ffn.layers.1.weight" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.05 + }, + "layer_11_no_decay": { + "param_names": [ + "backbone.layers.10.gamma_1", + "backbone.layers.10.gamma_2", + "backbone.layers.10.ln1.weight", + "backbone.layers.10.ln1.bias", + "backbone.layers.10.attn.qkv.bias", + "backbone.layers.10.attn.proj.bias", + "backbone.layers.10.ln2.weight", + "backbone.layers.10.ln2.bias", + "backbone.layers.10.ffn.layers.0.0.bias", + "backbone.layers.10.ffn.layers.1.bias" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.0 + }, + "layer_11_decay": { + "param_names": [ + "backbone.layers.10.attn.relative_position_bias_table", + "backbone.layers.10.attn.qkv.weight", + "backbone.layers.10.attn.proj.weight", + "backbone.layers.10.ffn.layers.0.0.weight", + "backbone.layers.10.ffn.layers.1.weight" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.05 + }, + "layer_12_no_decay": { + "param_names": [ + "backbone.layers.11.gamma_1", + "backbone.layers.11.gamma_2", + "backbone.layers.11.ln1.weight", + "backbone.layers.11.ln1.bias", + "backbone.layers.11.attn.qkv.bias", + "backbone.layers.11.attn.proj.bias", + "backbone.layers.11.ln2.weight", + "backbone.layers.11.ln2.bias", + "backbone.layers.11.ffn.layers.0.0.bias", + "backbone.layers.11.ffn.layers.1.bias" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.0 + }, + "layer_12_decay": { + "param_names": [ + "backbone.layers.11.attn.relative_position_bias_table", + "backbone.layers.11.attn.qkv.weight", + "backbone.layers.11.attn.proj.weight", + "backbone.layers.11.ffn.layers.0.0.weight", + "backbone.layers.11.ffn.layers.1.weight" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.05 + }, + "layer_13_decay": { + "param_names": [ + "neck.upsample_4x.0.weight", + "neck.upsample_4x.3.weight", + "neck.upsample_2x.0.weight", + "decode_head.conv_seg.weight", + "decode_head.psp_modules.0.1.conv.weight", + "decode_head.psp_modules.1.1.conv.weight", + "decode_head.psp_modules.2.1.conv.weight", + "decode_head.psp_modules.3.1.conv.weight", + "decode_head.bottleneck.conv.weight", + "decode_head.lateral_convs.0.conv.weight", + "decode_head.lateral_convs.1.conv.weight", + "decode_head.lateral_convs.2.conv.weight", + "decode_head.fpn_convs.0.conv.weight", + "decode_head.fpn_convs.1.conv.weight", + "decode_head.fpn_convs.2.conv.weight", + "decode_head.fpn_bottleneck.conv.weight", + "auxiliary_head.conv_seg.weight", + "auxiliary_head.convs.0.conv.weight" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.05 + }, + "layer_13_no_decay": { + "param_names": [ + "neck.upsample_4x.0.bias", + "neck.upsample_4x.1.weight", + "neck.upsample_4x.1.bias", + "neck.upsample_4x.3.bias", + "neck.upsample_2x.0.bias", + "decode_head.conv_seg.bias", + "decode_head.psp_modules.0.1.bn.weight", + "decode_head.psp_modules.0.1.bn.bias", + "decode_head.psp_modules.1.1.bn.weight", + "decode_head.psp_modules.1.1.bn.bias", + "decode_head.psp_modules.2.1.bn.weight", + "decode_head.psp_modules.2.1.bn.bias", + "decode_head.psp_modules.3.1.bn.weight", + "decode_head.psp_modules.3.1.bn.bias", + "decode_head.bottleneck.bn.weight", + "decode_head.bottleneck.bn.bias", + "decode_head.lateral_convs.0.bn.weight", + "decode_head.lateral_convs.0.bn.bias", + "decode_head.lateral_convs.1.bn.weight", + "decode_head.lateral_convs.1.bn.bias", + "decode_head.lateral_convs.2.bn.weight", + "decode_head.lateral_convs.2.bn.bias", + "decode_head.fpn_convs.0.bn.weight", + "decode_head.fpn_convs.0.bn.bias", + "decode_head.fpn_convs.1.bn.weight", + "decode_head.fpn_convs.1.bn.bias", + "decode_head.fpn_convs.2.bn.weight", + "decode_head.fpn_convs.2.bn.bias", + "decode_head.fpn_bottleneck.bn.weight", + "decode_head.fpn_bottleneck.bn.bias", + "auxiliary_head.conv_seg.bias", + "auxiliary_head.convs.0.bn.weight", + "auxiliary_head.convs.0.bn.bias" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.0 + } +} +04/19 06:52:26 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +Loads checkpoint by local backend from path: ./pretrain/mae_pretrain_vit_base_mmcls.pth +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1748, in train + self._init_model_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 923, in _init_model_weights + model.init_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/base_module.py", line 136, in init_weights + m.init_weights() + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/mae.py", line 183, in init_weights + checkpoint = _load_checkpoint( + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError: ./pretrain/mae_pretrain_vit_base_mmcls.pth can not be found. +Loads checkpoint by local backend from path: ./pretrain/mae_pretrain_vit_base_mmcls.pth +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1748, in train + self._init_model_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 923, in _init_model_weights + model.init_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/base_module.py", line 136, in init_weights + m.init_weights() + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/mae.py", line 183, in init_weights + checkpoint = _load_checkpoint( + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError: ./pretrain/mae_pretrain_vit_base_mmcls.pth can not be found. +Loads checkpoint by local backend from path: ./pretrain/mae_pretrain_vit_base_mmcls.pth +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1748, in train + self._init_model_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 923, in _init_model_weights + model.init_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/base_module.py", line 136, in init_weights + m.init_weights() + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/mae.py", line 183, in init_weights + checkpoint = _load_checkpoint( + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError: ./pretrain/mae_pretrain_vit_base_mmcls.pth can not be found. +Loads checkpoint by local backend from path: ./pretrain/mae_pretrain_vit_base_mmcls.pth +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1748, in train + self._init_model_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 923, in _init_model_weights + model.init_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/base_module.py", line 136, in init_weights + m.init_weights() + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/mae.py", line 183, in init_weights + checkpoint = _load_checkpoint( + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError: ./pretrain/mae_pretrain_vit_base_mmcls.pth can not be found. +ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 1) local_rank: 0 (pid: 308287) of binary: /opt/conda/bin/python +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 245, in launch_agent + raise ChildFailedError( +torch.distributed.elastic.multiprocessing.errors.ChildFailedError: +============================================================ +tools/train.py FAILED +------------------------------------------------------------ +Failures: +[1]: + time : 2024-04-19_06:52:29 + host : 06540a9609a4 + rank : 1 (local_rank: 1) + exitcode : 1 (pid: 308288) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[2]: + time : 2024-04-19_06:52:29 + host : 06540a9609a4 + rank : 2 (local_rank: 2) + exitcode : 1 (pid: 308289) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[3]: + time : 2024-04-19_06:52:29 + host : 06540a9609a4 + rank : 3 (local_rank: 3) + exitcode : 1 (pid: 308290) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +------------------------------------------------------------ +Root Cause (first observed failure): +[0]: + time : 2024-04-19_06:52:29 + host : 06540a9609a4 + rank : 0 (local_rank: 0) + exitcode : 1 (pid: 308287) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +============================================================ +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/19 06:57:54 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 1351088420 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 1351088420 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/19 06:57:54 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag_10000' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +fp16 = dict(loss_scale='dynamic') +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = None +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=768, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + num_convs=1, + type='FCNHead'), + backbone=dict( + act_cfg=dict(type='GELU'), + attn_drop_rate=0.0, + drop_path_rate=0.1, + embed_dims=768, + img_size=( + 512, + 512, + ), + in_channels=3, + init_values=1.0, + mlp_ratio=4, + norm_cfg=dict(eps=1e-06, type='LN'), + norm_eval=False, + num_heads=12, + num_layers=12, + out_indices=[ + 3, + 5, + 7, + 11, + ], + patch_size=16, + type='MAE'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=768, + dropout_ratio=0.1, + in_channels=[ + 768, + 768, + 768, + 768, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + neck=dict( + embed_dim=768, rescales=[ + 4, + 2, + 1, + 0.5, + ], type='Feature2Pyramid'), + pretrained='./mae_pretrain_vit_base.pth', + test_cfg=dict(crop_size=( + 512, + 512, + ), mode='slide', stride=( + 341, + 341, + )), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + constructor='LayerDecayOptimizerConstructor', + optimizer=dict( + betas=( + 0.9, + 0.999, + ), lr=0.0001, type='AdamW', weight_decay=0.05), + paramwise_cfg=dict(layer_decay_rate=0.65, num_layers=12), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, by_epoch=False, end=1500, start_factor=1e-06, + type='LinearLR'), + dict( + begin=1500, + by_epoch=False, + end=160000, + eta_min=0.0, + power=1.0, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag_10000', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=2, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag_10000', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag_10000', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = './work_dirs/mae-base_upernet_8xb2-amp-160k_ade20k-512x512' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/19 06:57:57 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65}set param backbone.layers.0.attn.proj.weight as id 1 + +set param backbone.layers.0.attn.proj.bias as id 1 +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.cls_token as id 0 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.pos_embed as id 0set param backbone.layers.1.gamma_2 as id 2 + +set param backbone.layers.1.ln1.weight as id 2 +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65}set param backbone.layers.1.ln1.bias as id 2 + +set param backbone.patch_embed.projection.weight as id 0set param backbone.layers.1.attn.relative_position_bias_table as id 2 +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 + +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1set param backbone.layers.1.ffn.layers.0.0.weight as id 2 + +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.cls_token as id 0 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.pos_embed as id 0set param backbone.layers.0.attn.qkv.bias as id 1 + +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.0.attn.proj.weight as id 1set param backbone.layers.2.ln1.weight as id 3 + +set param backbone.layers.0.attn.proj.bias as id 1set param backbone.layers.2.ln1.bias as id 3 + +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.layers.2.attn.relative_position_bias_table as id 3set param backbone.layers.0.ln2.weight as id 1 + +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.2.attn.qkv.weight as id 3set param backbone.layers.0.ln2.bias as id 1 + +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.0.gamma_1 as id 1set param backbone.layers.2.attn.proj.bias as id 3 + +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.2.ln2.weight as id 3set param backbone.layers.0.gamma_2 as id 1set param backbone.layers.0.ffn.layers.0.0.bias as id 1 + + +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.0.ln1.weight as id 1set param backbone.layers.2.ffn.layers.0.0.weight as id 3set param backbone.layers.0.ffn.layers.1.weight as id 1 + + +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.0.ln1.bias as id 1set param backbone.layers.0.ffn.layers.1.bias as id 1 + +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.1.gamma_2 as id 2set param backbone.layers.0.attn.qkv.weight as id 1 + +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.1.ln1.weight as id 2set param backbone.layers.0.attn.qkv.bias as id 1set param backbone.layers.3.ln1.bias as id 4 + + +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.0.attn.proj.weight as id 1set param backbone.layers.1.attn.qkv.weight as id 2set param backbone.layers.3.attn.qkv.bias as id 4 + + +set param backbone.layers.0.attn.proj.bias as id 1set param backbone.layers.1.attn.qkv.bias as id 2 + +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.1.attn.proj.weight as id 2set param backbone.layers.0.ln2.weight as id 1 + +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.0.ln2.bias as id 1set param backbone.layers.3.ln2.bias as id 4 + +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2set param backbone.layers.3.ffn.layers.1.weight as id 4 + +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.1.ffn.layers.1.weight as id 2set param backbone.layers.3.ffn.layers.1.bias as id 4 + +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.0.ffn.layers.1.bias as id 1set param backbone.layers.2.gamma_1 as id 3set param backbone.layers.4.gamma_2 as id 5 + + +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.2.ln1.weight as id 3set param backbone.layers.4.ln1.bias as id 5 + +set param backbone.layers.1.gamma_1 as id 2set param backbone.layers.2.ln1.bias as id 3set param backbone.layers.4.attn.relative_position_bias_table as id 5 + + +set param backbone.layers.1.gamma_2 as id 2set param backbone.layers.2.attn.relative_position_bias_table as id 3set param backbone.layers.4.attn.qkv.weight as id 5 + + +set param backbone.layers.1.ln1.weight as id 2set param backbone.layers.4.attn.qkv.bias as id 5set param backbone.layers.2.attn.qkv.weight as id 3 + + +set param backbone.layers.1.ln1.bias as id 2set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.4.attn.proj.weight as id 5 + +set param backbone.layers.1.attn.relative_position_bias_table as id 2set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.2.attn.proj.weight as id 3 + +set param backbone.layers.1.attn.qkv.weight as id 2set param backbone.layers.2.attn.proj.bias as id 3set param backbone.layers.4.ln2.weight as id 5 + + +set param backbone.layers.1.attn.qkv.bias as id 2set param backbone.layers.4.ln2.bias as id 5set param backbone.layers.2.ln2.weight as id 3 + + +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 + +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.1.ln2.weight as id 2set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.2.ffn.layers.1.weight as id 3 + +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.2.ffn.layers.1.bias as id 3set param backbone.layers.4.ffn.layers.1.bias as id 5 + +set param backbone.layers.3.gamma_1 as id 4set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.5.gamma_1 as id 6 + +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2set param backbone.layers.5.gamma_2 as id 6 + +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.5.ln1.weight as id 6set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.1.ffn.layers.1.weight as id 2 + +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.3.attn.relative_position_bias_table as id 4set param backbone.layers.1.ffn.layers.1.bias as id 2 + +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.2.gamma_1 as id 3set param backbone.layers.5.attn.qkv.bias as id 6 + +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.3.attn.proj.bias as id 4set param backbone.layers.5.attn.proj.weight as id 6 + +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.5.attn.proj.bias as id 6set param backbone.layers.3.ln2.weight as id 4 + +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.3.ln2.bias as id 4set param backbone.layers.5.ln2.weight as id 6 + +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4set param backbone.layers.2.attn.qkv.weight as id 3 + +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.2.attn.qkv.bias as id 3set param backbone.layers.5.ffn.layers.0.0.bias as id 6 + +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.5.ffn.layers.1.weight as id 6set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.3.ffn.layers.1.bias as id 4 + +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.4.gamma_1 as id 5set param backbone.layers.6.gamma_1 as id 7 + +set param backbone.layers.2.ln2.weight as id 3set param backbone.layers.4.gamma_2 as id 5set param backbone.layers.6.gamma_2 as id 7 + + +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.4.ln1.weight as id 5set param backbone.layers.6.ln1.weight as id 7 + +set param backbone.layers.4.ln1.bias as id 5set param backbone.layers.6.ln1.bias as id 7 + +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.6.attn.qkv.bias as id 7set param backbone.layers.2.ffn.layers.1.weight as id 3 + +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.2.ffn.layers.1.bias as id 3set param backbone.layers.6.attn.proj.weight as id 7set param backbone.layers.4.attn.proj.bias as id 5 + + +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.3.gamma_1 as id 4set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.4.ln2.bias as id 5 + +set param backbone.layers.3.gamma_2 as id 4set param backbone.layers.6.ln2.bias as id 7 + +set param backbone.layers.4.ffn.layers.0.0.weight as id 5set param backbone.layers.3.ln1.weight as id 4 + +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6set param backbone.layers.3.attn.proj.weight as id 4 + +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.3.ln2.bias as id 4set param backbone.layers.5.attn.qkv.bias as id 6 + +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.5.attn.proj.bias as id 6 + +set param backbone.layers.6.ffn.layers.1.weight as id 7set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.5.ln2.weight as id 6 + +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.7.gamma_2 as id 8set param backbone.layers.5.ffn.layers.0.0.bias as id 6 + +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.5.ffn.layers.1.weight as id 6set param backbone.layers.7.ln1.bias as id 8set param backbone.layers.4.gamma_1 as id 5 + + +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.7.attn.qkv.weight as id 8set param backbone.layers.6.gamma_1 as id 7 + +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.7.attn.qkv.bias as id 8set param backbone.layers.6.gamma_2 as id 7 + +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.7.attn.proj.weight as id 8set param backbone.layers.6.ln1.weight as id 7 + +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.7.attn.proj.bias as id 8set param backbone.layers.6.ln1.bias as id 7 + +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.7.ln2.weight as id 8set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.4.attn.qkv.bias as id 5 + +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.4.attn.proj.weight as id 5set param backbone.layers.6.attn.qkv.bias as id 7 + +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8set param backbone.layers.6.attn.proj.weight as id 7 + +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.6.attn.proj.bias as id 7set param backbone.layers.7.ffn.layers.1.weight as id 8 + +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.7.ffn.layers.1.bias as id 8set param backbone.layers.6.ln2.weight as id 7 + +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5set param backbone.layers.8.gamma_2 as id 9 + +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.5.gamma_2 as id 6set param backbone.layers.8.attn.proj.weight as id 9 + +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.5.attn.qkv.weight as id 6set param backbone.layers.8.ffn.layers.0.0.bias as id 9 + +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.8.ffn.layers.1.weight as id 9set param backbone.layers.6.ffn.layers.0.0.bias as id 7 + +set param backbone.layers.8.ffn.layers.1.bias as id 9set param backbone.layers.5.attn.proj.weight as id 6 + +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.5.attn.proj.bias as id 6set param backbone.layers.6.ffn.layers.1.bias as id 7 + +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.5.ln2.weight as id 6set param backbone.layers.9.gamma_2 as id 10 + +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.5.ln2.bias as id 6set param backbone.layers.9.ln1.weight as id 10 + +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.9.attn.relative_position_bias_table as id 10set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.7.ln1.bias as id 8 + +set param backbone.layers.9.attn.qkv.weight as id 10set param backbone.layers.5.ffn.layers.0.0.bias as id 6 + +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.5.ffn.layers.1.weight as id 6set param backbone.layers.9.attn.proj.weight as id 10 + +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.5.ffn.layers.1.bias as id 6set param backbone.layers.9.attn.proj.bias as id 10 + +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8set param backbone.layers.9.ln2.weight as id 10 + +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.9.ln2.bias as id 10set param backbone.layers.7.ln2.weight as id 8 + +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10set param backbone.layers.6.ln1.weight as id 7 + +set param backbone.layers.9.ffn.layers.0.0.bias as id 10set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.6.ln1.bias as id 7 + +set param backbone.layers.9.ffn.layers.1.weight as id 10set param backbone.layers.7.ffn.layers.0.0.bias as id 8 + +set param backbone.layers.6.attn.relative_position_bias_table as id 7set param backbone.layers.9.ffn.layers.1.bias as id 10 + +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8set param backbone.layers.6.attn.qkv.weight as id 7 + +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.6.attn.qkv.bias as id 7set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.8.gamma_1 as id 9 + +set param backbone.layers.10.ln1.weight as id 11set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.6.attn.proj.weight as id 7 + +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.8.ln1.weight as id 9set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 + +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.10.attn.qkv.weight as id 11set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 + +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.6.ln2.bias as id 7set param backbone.layers.8.attn.qkv.weight as id 9 + +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9set param backbone.layers.10.ffn.layers.0.0.weight as id 11 + +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.11.gamma_1 as id 12set param backbone.layers.8.ffn.layers.1.bias as id 9 + +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.9.gamma_1 as id 10set param backbone.layers.11.ln1.weight as id 12 + +set param backbone.layers.9.gamma_2 as id 10set param backbone.layers.11.ln1.bias as id 12 + +set param backbone.layers.9.ln1.weight as id 10set param backbone.layers.11.attn.relative_position_bias_table as id 12 + +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7set param backbone.layers.11.attn.proj.bias as id 12 + +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 + +set param backbone.layers.9.ln2.weight as id 10set param backbone.layers.11.ln2.bias as id 12 + +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10set param backbone.layers.11.ffn.layers.0.0.bias as id 12 + +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.7.gamma_1 as id 8set param backbone.layers.11.ffn.layers.1.weight as id 12 + +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.11.ffn.layers.1.bias as id 12set param backbone.layers.7.gamma_2 as id 8 + +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.10.gamma_1 as id 11set param backbone.layers.7.ln1.bias as id 8 + +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param neck.upsample_4x.0.weight as id 13 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.7.attn.qkv.weight as id 8set param neck.upsample_4x.0.bias as id 13 +set param backbone.layers.10.ln1.bias as id 11 + +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.10.attn.relative_position_bias_table as id 11set param neck.upsample_4x.1.weight as id 13 + +set param backbone.layers.7.attn.proj.weight as id 8set param backbone.layers.10.attn.qkv.weight as id 11 + +set param neck.upsample_4x.1.bias as id 13 +set param backbone.layers.7.attn.proj.bias as id 8set param backbone.layers.10.attn.qkv.bias as id 11 + +set param neck.upsample_4x.3.weight as id 13 +set param backbone.layers.7.ln2.weight as id 8set param backbone.layers.10.attn.proj.weight as id 11set param neck.upsample_4x.3.bias as id 13 + + +set param backbone.layers.7.ln2.bias as id 8set param backbone.layers.10.attn.proj.bias as id 11 + +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13set param backbone.layers.10.ln2.weight as id 11 + +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8set param backbone.layers.10.ffn.layers.0.0.weight as id 11 + +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param decode_head.conv_seg.weight as id 13set param backbone.layers.10.ffn.layers.0.0.bias as id 11 + +set param decode_head.conv_seg.bias as id 13 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.11.gamma_1 as id 12 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param backbone.layers.8.attn.relative_position_bias_table as id 9set param backbone.layers.11.gamma_2 as id 12 + +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param backbone.layers.8.attn.qkv.weight as id 9set param backbone.layers.11.ln1.weight as id 12 +set param decode_head.psp_modules.0.1.bn.bias as id 13 + +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.8.attn.proj.bias as id 9set param decode_head.psp_modules.1.1.conv.weight as id 13 + +set param backbone.layers.11.attn.qkv.weight as id 12 +set param decode_head.psp_modules.1.1.bn.weight as id 13set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.11.attn.qkv.bias as id 12 + +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.11.attn.proj.weight as id 12 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param backbone.layers.11.attn.proj.bias as id 12set param backbone.layers.8.ffn.layers.0.0.weight as id 9 + +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9set param backbone.layers.11.ln2.weight as id 12 +set param decode_head.psp_modules.2.1.bn.bias as id 13 + +set param backbone.layers.8.ffn.layers.1.weight as id 9set param backbone.layers.11.ln2.bias as id 12 + +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param backbone.layers.9.gamma_1 as id 10set param backbone.layers.11.ffn.layers.0.0.bias as id 12 + +set param decode_head.bottleneck.conv.weight as id 13 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param decode_head.bottleneck.bn.weight as id 13 +set param backbone.layers.9.ln1.weight as id 10set param backbone.layers.11.ffn.layers.1.bias as id 12 + +set param decode_head.bottleneck.bn.bias as id 13 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param decode_head.lateral_convs.0.bn.weight as id 13set param neck.upsample_4x.0.weight as id 13 +set param backbone.layers.9.attn.qkv.bias as id 10 + +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10set param decode_head.lateral_convs.1.conv.weight as id 13 +set param neck.upsample_4x.1.weight as id 13 + +set param backbone.layers.9.ln2.weight as id 10set param neck.upsample_4x.1.bias as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 + +set param backbone.layers.9.ln2.bias as id 10 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10set param neck.upsample_4x.3.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 + +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param neck.upsample_2x.0.weight as id 13set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param decode_head.lateral_convs.2.bn.bias as id 13 + +set param neck.upsample_2x.0.bias as id 13 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13set param backbone.layers.10.gamma_1 as id 11 + +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param backbone.layers.10.gamma_2 as id 11 +set param decode_head.fpn_convs.1.conv.weight as id 13set param backbone.layers.10.ln1.weight as id 11 +set param decode_head.conv_seg.weight as id 13 + +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param backbone.layers.10.ln1.bias as id 11set param decode_head.conv_seg.bias as id 13 + +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param backbone.layers.10.attn.qkv.bias as id 11set param decode_head.fpn_convs.2.bn.bias as id 13 + +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param backbone.layers.10.attn.proj.weight as id 11 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13set param backbone.layers.10.attn.proj.bias as id 11 + +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13set param backbone.layers.10.ln2.weight as id 11 + +set param backbone.layers.10.ln2.bias as id 11 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11set param decode_head.psp_modules.1.1.bn.weight as id 13 + +set param backbone.layers.10.ffn.layers.0.0.bias as id 11set param decode_head.psp_modules.1.1.bn.bias as id 13set param auxiliary_head.conv_seg.weight as id 13 + + +set param auxiliary_head.conv_seg.bias as id 13set param backbone.layers.10.ffn.layers.1.weight as id 11 + +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param backbone.layers.11.gamma_1 as id 12set param auxiliary_head.convs.0.bn.weight as id 13 + +set param backbone.layers.11.gamma_2 as id 12set param auxiliary_head.convs.0.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 + +set param backbone.layers.11.ln1.weight as id 12 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param backbone.layers.11.ln1.bias as id 12 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param decode_head.bottleneck.conv.weight as id 13 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param decode_head.bottleneck.bn.weight as id 13set param backbone.layers.11.attn.qkv.bias as id 12 + +set param decode_head.bottleneck.bn.bias as id 13 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param backbone.layers.11.ln2.weight as id 12 +set param decode_head.lateral_convs.0.bn.weight as id 13set param backbone.layers.11.ln2.bias as id 12 + +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param decode_head.lateral_convs.1.conv.weight as id 13set param backbone.layers.11.ffn.layers.0.0.bias as id 12 + +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param neck.upsample_4x.0.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13set param neck.upsample_4x.1.weight as id 13 + +set param decode_head.fpn_convs.1.bn.weight as id 13set param neck.upsample_4x.1.bias as id 13 + +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13set param neck.upsample_4x.3.bias as id 13 + +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param neck.upsample_2x.0.weight as id 13set param decode_head.fpn_convs.2.bn.bias as id 13 + +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13set param auxiliary_head.convs.0.conv.weight as id 13 + +set param decode_head.psp_modules.0.1.bn.weight as id 13set param auxiliary_head.convs.0.bn.weight as id 13 + +set param decode_head.psp_modules.0.1.bn.bias as id 13set param auxiliary_head.convs.0.bn.bias as id 13 + +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +Param groups = { + "layer_0_decay": { + "param_names": [ + "backbone.cls_token", + "backbone.pos_embed", + "backbone.patch_embed.projection.weight" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.05 + }, + "layer_0_no_decay": { + "param_names": [ + "backbone.patch_embed.projection.bias" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.0 + }, + "layer_1_no_decay": { + "param_names": [ + "backbone.layers.0.gamma_1", + "backbone.layers.0.gamma_2", + "backbone.layers.0.ln1.weight", + "backbone.layers.0.ln1.bias", + "backbone.layers.0.attn.qkv.bias", + "backbone.layers.0.attn.proj.bias", + "backbone.layers.0.ln2.weight", + "backbone.layers.0.ln2.bias", + "backbone.layers.0.ffn.layers.0.0.bias", + "backbone.layers.0.ffn.layers.1.bias" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.0 + }, + "layer_1_decay": { + "param_names": [ + "backbone.layers.0.attn.relative_position_bias_table", + "backbone.layers.0.attn.qkv.weight", + "backbone.layers.0.attn.proj.weight", + "backbone.layers.0.ffn.layers.0.0.weight", + "backbone.layers.0.ffn.layers.1.weight" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.05 + }, + "layer_2_no_decay": { + "param_names": [ + "backbone.layers.1.gamma_1", + "backbone.layers.1.gamma_2", + "backbone.layers.1.ln1.weight", + "backbone.layers.1.ln1.bias", + "backbone.layers.1.attn.qkv.bias", + "backbone.layers.1.attn.proj.bias", + "backbone.layers.1.ln2.weight", + "backbone.layers.1.ln2.bias", + "backbone.layers.1.ffn.layers.0.0.bias", + "backbone.layers.1.ffn.layers.1.bias" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.0 + }, + "layer_2_decay": { + "param_names": [ + "backbone.layers.1.attn.relative_position_bias_table", + "backbone.layers.1.attn.qkv.weight", + "backbone.layers.1.attn.proj.weight", + "backbone.layers.1.ffn.layers.0.0.weight", + "backbone.layers.1.ffn.layers.1.weight" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.05 + }, + "layer_3_no_decay": { + "param_names": [ + "backbone.layers.2.gamma_1", + "backbone.layers.2.gamma_2", + "backbone.layers.2.ln1.weight", + "backbone.layers.2.ln1.bias", + "backbone.layers.2.attn.qkv.bias", + "backbone.layers.2.attn.proj.bias", + "backbone.layers.2.ln2.weight", + "backbone.layers.2.ln2.bias", + "backbone.layers.2.ffn.layers.0.0.bias", + "backbone.layers.2.ffn.layers.1.bias" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.0 + }, + "layer_3_decay": { + "param_names": [ + "backbone.layers.2.attn.relative_position_bias_table", + "backbone.layers.2.attn.qkv.weight", + "backbone.layers.2.attn.proj.weight", + "backbone.layers.2.ffn.layers.0.0.weight", + "backbone.layers.2.ffn.layers.1.weight" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.05 + }, + "layer_4_no_decay": { + "param_names": [ + "backbone.layers.3.gamma_1", + "backbone.layers.3.gamma_2", + "backbone.layers.3.ln1.weight", + "backbone.layers.3.ln1.bias", + "backbone.layers.3.attn.qkv.bias", + "backbone.layers.3.attn.proj.bias", + "backbone.layers.3.ln2.weight", + "backbone.layers.3.ln2.bias", + "backbone.layers.3.ffn.layers.0.0.bias", + "backbone.layers.3.ffn.layers.1.bias" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.0 + }, + "layer_4_decay": { + "param_names": [ + "backbone.layers.3.attn.relative_position_bias_table", + "backbone.layers.3.attn.qkv.weight", + "backbone.layers.3.attn.proj.weight", + "backbone.layers.3.ffn.layers.0.0.weight", + "backbone.layers.3.ffn.layers.1.weight" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.05 + }, + "layer_5_no_decay": { + "param_names": [ + "backbone.layers.4.gamma_1", + "backbone.layers.4.gamma_2", + "backbone.layers.4.ln1.weight", + "backbone.layers.4.ln1.bias", + "backbone.layers.4.attn.qkv.bias", + "backbone.layers.4.attn.proj.bias", + "backbone.layers.4.ln2.weight", + "backbone.layers.4.ln2.bias", + "backbone.layers.4.ffn.layers.0.0.bias", + "backbone.layers.4.ffn.layers.1.bias" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.0 + }, + "layer_5_decay": { + "param_names": [ + "backbone.layers.4.attn.relative_position_bias_table", + "backbone.layers.4.attn.qkv.weight", + "backbone.layers.4.attn.proj.weight", + "backbone.layers.4.ffn.layers.0.0.weight", + "backbone.layers.4.ffn.layers.1.weight" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.05 + }, + "layer_6_no_decay": { + "param_names": [ + "backbone.layers.5.gamma_1", + "backbone.layers.5.gamma_2", + "backbone.layers.5.ln1.weight", + "backbone.layers.5.ln1.bias", + "backbone.layers.5.attn.qkv.bias", + "backbone.layers.5.attn.proj.bias", + "backbone.layers.5.ln2.weight", + "backbone.layers.5.ln2.bias", + "backbone.layers.5.ffn.layers.0.0.bias", + "backbone.layers.5.ffn.layers.1.bias" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.0 + }, + "layer_6_decay": { + "param_names": [ + "backbone.layers.5.attn.relative_position_bias_table", + "backbone.layers.5.attn.qkv.weight", + "backbone.layers.5.attn.proj.weight", + "backbone.layers.5.ffn.layers.0.0.weight", + "backbone.layers.5.ffn.layers.1.weight" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.05 + }, + "layer_7_no_decay": { + "param_names": [ + "backbone.layers.6.gamma_1", + "backbone.layers.6.gamma_2", + "backbone.layers.6.ln1.weight", + "backbone.layers.6.ln1.bias", + "backbone.layers.6.attn.qkv.bias", + "backbone.layers.6.attn.proj.bias", + "backbone.layers.6.ln2.weight", + "backbone.layers.6.ln2.bias", + "backbone.layers.6.ffn.layers.0.0.bias", + "backbone.layers.6.ffn.layers.1.bias" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.0 + }, + "layer_7_decay": { + "param_names": [ + "backbone.layers.6.attn.relative_position_bias_table", + "backbone.layers.6.attn.qkv.weight", + "backbone.layers.6.attn.proj.weight", + "backbone.layers.6.ffn.layers.0.0.weight", + "backbone.layers.6.ffn.layers.1.weight" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.05 + }, + "layer_8_no_decay": { + "param_names": [ + "backbone.layers.7.gamma_1", + "backbone.layers.7.gamma_2", + "backbone.layers.7.ln1.weight", + "backbone.layers.7.ln1.bias", + "backbone.layers.7.attn.qkv.bias", + "backbone.layers.7.attn.proj.bias", + "backbone.layers.7.ln2.weight", + "backbone.layers.7.ln2.bias", + "backbone.layers.7.ffn.layers.0.0.bias", + "backbone.layers.7.ffn.layers.1.bias" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.0 + }, + "layer_8_decay": { + "param_names": [ + "backbone.layers.7.attn.relative_position_bias_table", + "backbone.layers.7.attn.qkv.weight", + "backbone.layers.7.attn.proj.weight", + "backbone.layers.7.ffn.layers.0.0.weight", + "backbone.layers.7.ffn.layers.1.weight" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.05 + }, + "layer_9_no_decay": { + "param_names": [ + "backbone.layers.8.gamma_1", + "backbone.layers.8.gamma_2", + "backbone.layers.8.ln1.weight", + "backbone.layers.8.ln1.bias", + "backbone.layers.8.attn.qkv.bias", + "backbone.layers.8.attn.proj.bias", + "backbone.layers.8.ln2.weight", + "backbone.layers.8.ln2.bias", + "backbone.layers.8.ffn.layers.0.0.bias", + "backbone.layers.8.ffn.layers.1.bias" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.0 + }, + "layer_9_decay": { + "param_names": [ + "backbone.layers.8.attn.relative_position_bias_table", + "backbone.layers.8.attn.qkv.weight", + "backbone.layers.8.attn.proj.weight", + "backbone.layers.8.ffn.layers.0.0.weight", + "backbone.layers.8.ffn.layers.1.weight" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.05 + }, + "layer_10_no_decay": { + "param_names": [ + "backbone.layers.9.gamma_1", + "backbone.layers.9.gamma_2", + "backbone.layers.9.ln1.weight", + "backbone.layers.9.ln1.bias", + "backbone.layers.9.attn.qkv.bias", + "backbone.layers.9.attn.proj.bias", + "backbone.layers.9.ln2.weight", + "backbone.layers.9.ln2.bias", + "backbone.layers.9.ffn.layers.0.0.bias", + "backbone.layers.9.ffn.layers.1.bias" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.0 + }, + "layer_10_decay": { + "param_names": [ + "backbone.layers.9.attn.relative_position_bias_table", + "backbone.layers.9.attn.qkv.weight", + "backbone.layers.9.attn.proj.weight", + "backbone.layers.9.ffn.layers.0.0.weight", + "backbone.layers.9.ffn.layers.1.weight" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.05 + }, + "layer_11_no_decay": { + "param_names": [ + "backbone.layers.10.gamma_1", + "backbone.layers.10.gamma_2", + "backbone.layers.10.ln1.weight", + "backbone.layers.10.ln1.bias", + "backbone.layers.10.attn.qkv.bias", + "backbone.layers.10.attn.proj.bias", + "backbone.layers.10.ln2.weight", + "backbone.layers.10.ln2.bias", + "backbone.layers.10.ffn.layers.0.0.bias", + "backbone.layers.10.ffn.layers.1.bias" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.0 + }, + "layer_11_decay": { + "param_names": [ + "backbone.layers.10.attn.relative_position_bias_table", + "backbone.layers.10.attn.qkv.weight", + "backbone.layers.10.attn.proj.weight", + "backbone.layers.10.ffn.layers.0.0.weight", + "backbone.layers.10.ffn.layers.1.weight" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.05 + }, + "layer_12_no_decay": { + "param_names": [ + "backbone.layers.11.gamma_1", + "backbone.layers.11.gamma_2", + "backbone.layers.11.ln1.weight", + "backbone.layers.11.ln1.bias", + "backbone.layers.11.attn.qkv.bias", + "backbone.layers.11.attn.proj.bias", + "backbone.layers.11.ln2.weight", + "backbone.layers.11.ln2.bias", + "backbone.layers.11.ffn.layers.0.0.bias", + "backbone.layers.11.ffn.layers.1.bias" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.0 + }, + "layer_12_decay": { + "param_names": [ + "backbone.layers.11.attn.relative_position_bias_table", + "backbone.layers.11.attn.qkv.weight", + "backbone.layers.11.attn.proj.weight", + "backbone.layers.11.ffn.layers.0.0.weight", + "backbone.layers.11.ffn.layers.1.weight" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.05 + }, + "layer_13_decay": { + "param_names": [ + "neck.upsample_4x.0.weight", + "neck.upsample_4x.3.weight", + "neck.upsample_2x.0.weight", + "decode_head.conv_seg.weight", + "decode_head.psp_modules.0.1.conv.weight", + "decode_head.psp_modules.1.1.conv.weight", + "decode_head.psp_modules.2.1.conv.weight", + "decode_head.psp_modules.3.1.conv.weight", + "decode_head.bottleneck.conv.weight", + "decode_head.lateral_convs.0.conv.weight", + "decode_head.lateral_convs.1.conv.weight", + "decode_head.lateral_convs.2.conv.weight", + "decode_head.fpn_convs.0.conv.weight", + "decode_head.fpn_convs.1.conv.weight", + "decode_head.fpn_convs.2.conv.weight", + "decode_head.fpn_bottleneck.conv.weight", + "auxiliary_head.conv_seg.weight", + "auxiliary_head.convs.0.conv.weight" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.05 + }, + "layer_13_no_decay": { + "param_names": [ + "neck.upsample_4x.0.bias", + "neck.upsample_4x.1.weight", + "neck.upsample_4x.1.bias", + "neck.upsample_4x.3.bias", + "neck.upsample_2x.0.bias", + "decode_head.conv_seg.bias", + "decode_head.psp_modules.0.1.bn.weight", + "decode_head.psp_modules.0.1.bn.bias", + "decode_head.psp_modules.1.1.bn.weight", + "decode_head.psp_modules.1.1.bn.bias", + "decode_head.psp_modules.2.1.bn.weight", + "decode_head.psp_modules.2.1.bn.bias", + "decode_head.psp_modules.3.1.bn.weight", + "decode_head.psp_modules.3.1.bn.bias", + "decode_head.bottleneck.bn.weight", + "decode_head.bottleneck.bn.bias", + "decode_head.lateral_convs.0.bn.weight", + "decode_head.lateral_convs.0.bn.bias", + "decode_head.lateral_convs.1.bn.weight", + "decode_head.lateral_convs.1.bn.bias", + "decode_head.lateral_convs.2.bn.weight", + "decode_head.lateral_convs.2.bn.bias", + "decode_head.fpn_convs.0.bn.weight", + "decode_head.fpn_convs.0.bn.bias", + "decode_head.fpn_convs.1.bn.weight", + "decode_head.fpn_convs.1.bn.bias", + "decode_head.fpn_convs.2.bn.weight", + "decode_head.fpn_convs.2.bn.bias", + "decode_head.fpn_bottleneck.bn.weight", + "decode_head.fpn_bottleneck.bn.bias", + "auxiliary_head.conv_seg.bias", + "auxiliary_head.convs.0.bn.weight", + "auxiliary_head.convs.0.bn.bias" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.0 + } +} +04/19 06:57:58 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +Loads checkpoint by local backend from path: ./mae_pretrain_vit_base.pth +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1748, in train + self._init_model_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 923, in _init_model_weights + model.init_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/base_module.py", line 136, in init_weights + m.init_weights() + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/mae.py", line 183, in init_weights + checkpoint = _load_checkpoint( + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError: ./mae_pretrain_vit_base.pth can not be found. +Loads checkpoint by local backend from path: ./mae_pretrain_vit_base.pth +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1748, in train + self._init_model_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 923, in _init_model_weights + model.init_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/base_module.py", line 136, in init_weights + m.init_weights() + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/mae.py", line 183, in init_weights + checkpoint = _load_checkpoint( + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError: ./mae_pretrain_vit_base.pth can not be found. +Loads checkpoint by local backend from path: ./mae_pretrain_vit_base.pth +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1748, in train + self._init_model_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 923, in _init_model_weights + model.init_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/base_module.py", line 136, in init_weights + m.init_weights() + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/mae.py", line 183, in init_weights + checkpoint = _load_checkpoint( + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError: ./mae_pretrain_vit_base.pth can not be found. +Loads checkpoint by local backend from path: ./mae_pretrain_vit_base.pth +Traceback (most recent call last): + File "tools/train.py", line 104, in + main() + File "tools/train.py", line 100, in main + runner.train() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1748, in train + self._init_model_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/runner.py", line 923, in _init_model_weights + model.init_weights() + File "/opt/conda/lib/python3.8/site-packages/mmengine/model/base_module.py", line 136, in init_weights + m.init_weights() + File "/workspaces/mmsegmentation-1/mmseg/models/backbones/mae.py", line 183, in init_weights + checkpoint = _load_checkpoint( + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 548, in _load_checkpoint + return CheckpointLoader.load_checkpoint(filename, map_location, logger) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 330, in load_checkpoint + return checkpoint_loader(filename, map_location) + File "/opt/conda/lib/python3.8/site-packages/mmengine/runner/checkpoint.py", line 346, in load_from_local + raise FileNotFoundError(f'{filename} can not be found.') +FileNotFoundError: ./mae_pretrain_vit_base.pth can not be found. +ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 1) local_rank: 0 (pid: 312536) of binary: /opt/conda/bin/python +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 245, in launch_agent + raise ChildFailedError( +torch.distributed.elastic.multiprocessing.errors.ChildFailedError: +============================================================ +tools/train.py FAILED +------------------------------------------------------------ +Failures: +[1]: + time : 2024-04-19_06:58:00 + host : 06540a9609a4 + rank : 1 (local_rank: 1) + exitcode : 1 (pid: 312537) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[2]: + time : 2024-04-19_06:58:00 + host : 06540a9609a4 + rank : 2 (local_rank: 2) + exitcode : 1 (pid: 312538) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +[3]: + time : 2024-04-19_06:58:00 + host : 06540a9609a4 + rank : 3 (local_rank: 3) + exitcode : 1 (pid: 312539) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +------------------------------------------------------------ +Root Cause (first observed failure): +[0]: + time : 2024-04-19_06:58:00 + host : 06540a9609a4 + rank : 0 (local_rank: 0) + exitcode : 1 (pid: 312536) + error_file: + traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html +============================================================ +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/19 06:59:25 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 330997872 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 330997872 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/19 06:59:25 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag_10000' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +fp16 = dict(loss_scale='dynamic') +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = None +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=768, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + num_convs=1, + type='FCNHead'), + backbone=dict( + act_cfg=dict(type='GELU'), + attn_drop_rate=0.0, + drop_path_rate=0.1, + embed_dims=768, + img_size=( + 512, + 512, + ), + in_channels=3, + init_values=1.0, + mlp_ratio=4, + norm_cfg=dict(eps=1e-06, type='LN'), + norm_eval=False, + num_heads=12, + num_layers=12, + out_indices=[ + 3, + 5, + 7, + 11, + ], + patch_size=16, + type='MAE'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=768, + dropout_ratio=0.1, + in_channels=[ + 768, + 768, + 768, + 768, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + neck=dict( + embed_dim=768, rescales=[ + 4, + 2, + 1, + 0.5, + ], type='Feature2Pyramid'), + pretrained= + '/workspaces/mmsegmentation-1/configs/mae/mae_pretrain_vit_base.pth', + test_cfg=dict(crop_size=( + 512, + 512, + ), mode='slide', stride=( + 341, + 341, + )), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + constructor='LayerDecayOptimizerConstructor', + optimizer=dict( + betas=( + 0.9, + 0.999, + ), lr=0.0001, type='AdamW', weight_decay=0.05), + paramwise_cfg=dict(layer_decay_rate=0.65, num_layers=12), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, by_epoch=False, end=1500, start_factor=1e-06, + type='LinearLR'), + dict( + begin=1500, + by_epoch=False, + end=160000, + eta_min=0.0, + power=1.0, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag_10000', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=2, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag_10000', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag_10000', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = './work_dirs/mae-base_upernet_8xb2-amp-160k_ade20k-512x512' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/19 06:59:28 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +Param groups = { + "layer_0_decay": { + "param_names": [ + "backbone.cls_token", + "backbone.pos_embed", + "backbone.patch_embed.projection.weight" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.05 + }, + "layer_0_no_decay": { + "param_names": [ + "backbone.patch_embed.projection.bias" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.0 + }, + "layer_1_no_decay": { + "param_names": [ + "backbone.layers.0.gamma_1", + "backbone.layers.0.gamma_2", + "backbone.layers.0.ln1.weight", + "backbone.layers.0.ln1.bias", + "backbone.layers.0.attn.qkv.bias", + "backbone.layers.0.attn.proj.bias", + "backbone.layers.0.ln2.weight", + "backbone.layers.0.ln2.bias", + "backbone.layers.0.ffn.layers.0.0.bias", + "backbone.layers.0.ffn.layers.1.bias" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.0 + }, + "layer_1_decay": { + "param_names": [ + "backbone.layers.0.attn.relative_position_bias_table", + "backbone.layers.0.attn.qkv.weight", + "backbone.layers.0.attn.proj.weight", + "backbone.layers.0.ffn.layers.0.0.weight", + "backbone.layers.0.ffn.layers.1.weight" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.05 + }, + "layer_2_no_decay": { + "param_names": [ + "backbone.layers.1.gamma_1", + "backbone.layers.1.gamma_2", + "backbone.layers.1.ln1.weight", + "backbone.layers.1.ln1.bias", + "backbone.layers.1.attn.qkv.bias", + "backbone.layers.1.attn.proj.bias", + "backbone.layers.1.ln2.weight", + "backbone.layers.1.ln2.bias", + "backbone.layers.1.ffn.layers.0.0.bias", + "backbone.layers.1.ffn.layers.1.bias" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.0 + }, + "layer_2_decay": { + "param_names": [ + "backbone.layers.1.attn.relative_position_bias_table", + "backbone.layers.1.attn.qkv.weight", + "backbone.layers.1.attn.proj.weight", + "backbone.layers.1.ffn.layers.0.0.weight", + "backbone.layers.1.ffn.layers.1.weight" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.05 + }, + "layer_3_no_decay": { + "param_names": [ + "backbone.layers.2.gamma_1", + "backbone.layers.2.gamma_2", + "backbone.layers.2.ln1.weight", + "backbone.layers.2.ln1.bias", + "backbone.layers.2.attn.qkv.bias", + "backbone.layers.2.attn.proj.bias", + "backbone.layers.2.ln2.weight", + "backbone.layers.2.ln2.bias", + "backbone.layers.2.ffn.layers.0.0.bias", + "backbone.layers.2.ffn.layers.1.bias" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.0 + }, + "layer_3_decay": { + "param_names": [ + "backbone.layers.2.attn.relative_position_bias_table", + "backbone.layers.2.attn.qkv.weight", + "backbone.layers.2.attn.proj.weight", + "backbone.layers.2.ffn.layers.0.0.weight", + "backbone.layers.2.ffn.layers.1.weight" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.05 + }, + "layer_4_no_decay": { + "param_names": [ + "backbone.layers.3.gamma_1", + "backbone.layers.3.gamma_2", + "backbone.layers.3.ln1.weight", + "backbone.layers.3.ln1.bias", + "backbone.layers.3.attn.qkv.bias", + "backbone.layers.3.attn.proj.bias", + "backbone.layers.3.ln2.weight", + "backbone.layers.3.ln2.bias", + "backbone.layers.3.ffn.layers.0.0.bias", + "backbone.layers.3.ffn.layers.1.bias" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.0 + }, + "layer_4_decay": { + "param_names": [ + "backbone.layers.3.attn.relative_position_bias_table", + "backbone.layers.3.attn.qkv.weight", + "backbone.layers.3.attn.proj.weight", + "backbone.layers.3.ffn.layers.0.0.weight", + "backbone.layers.3.ffn.layers.1.weight" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.05 + }, + "layer_5_no_decay": { + "param_names": [ + "backbone.layers.4.gamma_1", + "backbone.layers.4.gamma_2", + "backbone.layers.4.ln1.weight", + "backbone.layers.4.ln1.bias", + "backbone.layers.4.attn.qkv.bias", + "backbone.layers.4.attn.proj.bias", + "backbone.layers.4.ln2.weight", + "backbone.layers.4.ln2.bias", + "backbone.layers.4.ffn.layers.0.0.bias", + "backbone.layers.4.ffn.layers.1.bias" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.0 + }, + "layer_5_decay": { + "param_names": [ + "backbone.layers.4.attn.relative_position_bias_table", + "backbone.layers.4.attn.qkv.weight", + "backbone.layers.4.attn.proj.weight", + "backbone.layers.4.ffn.layers.0.0.weight", + "backbone.layers.4.ffn.layers.1.weight" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.05 + }, + "layer_6_no_decay": { + "param_names": [ + "backbone.layers.5.gamma_1", + "backbone.layers.5.gamma_2", + "backbone.layers.5.ln1.weight", + "backbone.layers.5.ln1.bias", + "backbone.layers.5.attn.qkv.bias", + "backbone.layers.5.attn.proj.bias", + "backbone.layers.5.ln2.weight", + "backbone.layers.5.ln2.bias", + "backbone.layers.5.ffn.layers.0.0.bias", + "backbone.layers.5.ffn.layers.1.bias" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.0 + }, + "layer_6_decay": { + "param_names": [ + "backbone.layers.5.attn.relative_position_bias_table", + "backbone.layers.5.attn.qkv.weight", + "backbone.layers.5.attn.proj.weight", + "backbone.layers.5.ffn.layers.0.0.weight", + "backbone.layers.5.ffn.layers.1.weight" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.05 + }, + "layer_7_no_decay": { + "param_names": [ + "backbone.layers.6.gamma_1", + "backbone.layers.6.gamma_2", + "backbone.layers.6.ln1.weight", + "backbone.layers.6.ln1.bias", + "backbone.layers.6.attn.qkv.bias", + "backbone.layers.6.attn.proj.bias", + "backbone.layers.6.ln2.weight", + "backbone.layers.6.ln2.bias", + "backbone.layers.6.ffn.layers.0.0.bias", + "backbone.layers.6.ffn.layers.1.bias" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.0 + }, + "layer_7_decay": { + "param_names": [ + "backbone.layers.6.attn.relative_position_bias_table", + "backbone.layers.6.attn.qkv.weight", + "backbone.layers.6.attn.proj.weight", + "backbone.layers.6.ffn.layers.0.0.weight", + "backbone.layers.6.ffn.layers.1.weight" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.05 + }, + "layer_8_no_decay": { + "param_names": [ + "backbone.layers.7.gamma_1", + "backbone.layers.7.gamma_2", + "backbone.layers.7.ln1.weight", + "backbone.layers.7.ln1.bias", + "backbone.layers.7.attn.qkv.bias", + "backbone.layers.7.attn.proj.bias", + "backbone.layers.7.ln2.weight", + "backbone.layers.7.ln2.bias", + "backbone.layers.7.ffn.layers.0.0.bias", + "backbone.layers.7.ffn.layers.1.bias" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.0 + }, + "layer_8_decay": { + "param_names": [ + "backbone.layers.7.attn.relative_position_bias_table", + "backbone.layers.7.attn.qkv.weight", + "backbone.layers.7.attn.proj.weight", + "backbone.layers.7.ffn.layers.0.0.weight", + "backbone.layers.7.ffn.layers.1.weight" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.05 + }, + "layer_9_no_decay": { + "param_names": [ + "backbone.layers.8.gamma_1", + "backbone.layers.8.gamma_2", + "backbone.layers.8.ln1.weight", + "backbone.layers.8.ln1.bias", + "backbone.layers.8.attn.qkv.bias", + "backbone.layers.8.attn.proj.bias", + "backbone.layers.8.ln2.weight", + "backbone.layers.8.ln2.bias", + "backbone.layers.8.ffn.layers.0.0.bias", + "backbone.layers.8.ffn.layers.1.bias" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.0 + }, + "layer_9_decay": { + "param_names": [ + "backbone.layers.8.attn.relative_position_bias_table", + "backbone.layers.8.attn.qkv.weight", + "backbone.layers.8.attn.proj.weight", + "backbone.layers.8.ffn.layers.0.0.weight", + "backbone.layers.8.ffn.layers.1.weight" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.05 + }, + "layer_10_no_decay": { + "param_names": [ + "backbone.layers.9.gamma_1", + "backbone.layers.9.gamma_2", + "backbone.layers.9.ln1.weight", + "backbone.layers.9.ln1.bias", + "backbone.layers.9.attn.qkv.bias", + "backbone.layers.9.attn.proj.bias", + "backbone.layers.9.ln2.weight", + "backbone.layers.9.ln2.bias", + "backbone.layers.9.ffn.layers.0.0.bias", + "backbone.layers.9.ffn.layers.1.bias" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.0 + }, + "layer_10_decay": { + "param_names": [ + "backbone.layers.9.attn.relative_position_bias_table", + "backbone.layers.9.attn.qkv.weight", + "backbone.layers.9.attn.proj.weight", + "backbone.layers.9.ffn.layers.0.0.weight", + "backbone.layers.9.ffn.layers.1.weight" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.05 + }, + "layer_11_no_decay": { + "param_names": [ + "backbone.layers.10.gamma_1", + "backbone.layers.10.gamma_2", + "backbone.layers.10.ln1.weight", + "backbone.layers.10.ln1.bias", + "backbone.layers.10.attn.qkv.bias", + "backbone.layers.10.attn.proj.bias", + "backbone.layers.10.ln2.weight", + "backbone.layers.10.ln2.bias", + "backbone.layers.10.ffn.layers.0.0.bias", + "backbone.layers.10.ffn.layers.1.bias" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.0 + }, + "layer_11_decay": { + "param_names": [ + "backbone.layers.10.attn.relative_position_bias_table", + "backbone.layers.10.attn.qkv.weight", + "backbone.layers.10.attn.proj.weight", + "backbone.layers.10.ffn.layers.0.0.weight", + "backbone.layers.10.ffn.layers.1.weight" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.05 + }, + "layer_12_no_decay": { + "param_names": [ + "backbone.layers.11.gamma_1", + "backbone.layers.11.gamma_2", + "backbone.layers.11.ln1.weight", + "backbone.layers.11.ln1.bias", + "backbone.layers.11.attn.qkv.bias", + "backbone.layers.11.attn.proj.bias", + "backbone.layers.11.ln2.weight", + "backbone.layers.11.ln2.bias", + "backbone.layers.11.ffn.layers.0.0.bias", + "backbone.layers.11.ffn.layers.1.bias" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.0 + }, + "layer_12_decay": { + "param_names": [ + "backbone.layers.11.attn.relative_position_bias_table", + "backbone.layers.11.attn.qkv.weight", + "backbone.layers.11.attn.proj.weight", + "backbone.layers.11.ffn.layers.0.0.weight", + "backbone.layers.11.ffn.layers.1.weight" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.05 + }, + "layer_13_decay": { + "param_names": [ + "neck.upsample_4x.0.weight", + "neck.upsample_4x.3.weight", + "neck.upsample_2x.0.weight", + "decode_head.conv_seg.weight", + "decode_head.psp_modules.0.1.conv.weight", + "decode_head.psp_modules.1.1.conv.weight", + "decode_head.psp_modules.2.1.conv.weight", + "decode_head.psp_modules.3.1.conv.weight", + "decode_head.bottleneck.conv.weight", + "decode_head.lateral_convs.0.conv.weight", + "decode_head.lateral_convs.1.conv.weight", + "decode_head.lateral_convs.2.conv.weight", + "decode_head.fpn_convs.0.conv.weight", + "decode_head.fpn_convs.1.conv.weight", + "decode_head.fpn_convs.2.conv.weight", + "decode_head.fpn_bottleneck.conv.weight", + "auxiliary_head.conv_seg.weight", + "auxiliary_head.convs.0.conv.weight" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.05 + }, + "layer_13_no_decay": { + "param_names": [ + "neck.upsample_4x.0.bias", + "neck.upsample_4x.1.weight", + "neck.upsample_4x.1.bias", + "neck.upsample_4x.3.bias", + "neck.upsample_2x.0.bias", + "decode_head.conv_seg.bias", + "decode_head.psp_modules.0.1.bn.weight", + "decode_head.psp_modules.0.1.bn.bias", + "decode_head.psp_modules.1.1.bn.weight", + "decode_head.psp_modules.1.1.bn.bias", + "decode_head.psp_modules.2.1.bn.weight", + "decode_head.psp_modules.2.1.bn.bias", + "decode_head.psp_modules.3.1.bn.weight", + "decode_head.psp_modules.3.1.bn.bias", + "decode_head.bottleneck.bn.weight", + "decode_head.bottleneck.bn.bias", + "decode_head.lateral_convs.0.bn.weight", + "decode_head.lateral_convs.0.bn.bias", + "decode_head.lateral_convs.1.bn.weight", + "decode_head.lateral_convs.1.bn.bias", + "decode_head.lateral_convs.2.bn.weight", + "decode_head.lateral_convs.2.bn.bias", + "decode_head.fpn_convs.0.bn.weight", + "decode_head.fpn_convs.0.bn.bias", + "decode_head.fpn_convs.1.bn.weight", + "decode_head.fpn_convs.1.bn.bias", + "decode_head.fpn_convs.2.bn.weight", + "decode_head.fpn_convs.2.bn.bias", + "decode_head.fpn_bottleneck.bn.weight", + "decode_head.fpn_bottleneck.bn.bias", + "auxiliary_head.conv_seg.bias", + "auxiliary_head.convs.0.bn.weight", + "auxiliary_head.convs.0.bn.bias" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.0 + } +} +04/19 06:59:29 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/configs/mae/mae_pretrain_vit_base.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/configs/mae/mae_pretrain_vit_base.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/configs/mae/mae_pretrain_vit_base.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/configs/mae/mae_pretrain_vit_base.pth +04/19 06:59:30 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +04/19 06:59:30 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +04/19 06:59:30 - mmengine - INFO - Checkpoints will be saved to /workspaces/mmsegmentation-1/work_dirs/mae-base_upernet_8xb2-amp-160k_ade20k-512x512. +04/19 07:00:27 - mmengine - INFO - Iter(train) [ 50/160000] base_lr: 3.2689e-06 lr: 1.2086e-08 eta: 2 days, 2:23:31 time: 1.0255 data_time: 0.0040 memory: 8935 loss: 6.8883 decode.loss_ce: 4.8707 decode.acc_seg: 29.3745 aux.loss_ce: 2.0176 aux.acc_seg: 0.3349 +04/19 07:01:18 - mmengine - INFO - Iter(train) [ 100/160000] base_lr: 6.6045e-06 lr: 2.4418e-08 eta: 1 day, 23:58:07 time: 1.0259 data_time: 0.0044 memory: 8462 loss: 6.5402 decode.loss_ce: 4.5599 decode.acc_seg: 36.9930 aux.loss_ce: 1.9803 aux.acc_seg: 2.7327 +04/19 07:02:09 - mmengine - INFO - Iter(train) [ 150/160000] base_lr: 9.9401e-06 lr: 3.6750e-08 eta: 1 day, 23:09:52 time: 1.0258 data_time: 0.0043 memory: 8462 loss: 6.1158 decode.loss_ce: 4.1781 decode.acc_seg: 36.4723 aux.loss_ce: 1.9377 aux.acc_seg: 38.2418 +04/19 07:03:00 - mmengine - INFO - Iter(train) [ 200/160000] base_lr: 1.3276e-05 lr: 4.9083e-08 eta: 1 day, 22:43:21 time: 1.0203 data_time: 0.0039 memory: 8462 loss: 5.6012 decode.loss_ce: 3.7162 decode.acc_seg: 58.7185 aux.loss_ce: 1.8850 aux.acc_seg: 35.2327 +04/19 07:03:51 - mmengine - INFO - Iter(train) [ 250/160000] base_lr: 1.6611e-05 lr: 6.1415e-08 eta: 1 day, 22:24:47 time: 1.0181 data_time: 0.0044 memory: 8462 loss: 5.1370 decode.loss_ce: 3.3229 decode.acc_seg: 60.3662 aux.loss_ce: 1.8141 aux.acc_seg: 37.0596 +04/19 07:04:42 - mmengine - INFO - Iter(train) [ 300/160000] base_lr: 1.9947e-05 lr: 7.3747e-08 eta: 1 day, 22:11:16 time: 1.0159 data_time: 0.0044 memory: 8462 loss: 4.5380 decode.loss_ce: 2.7961 decode.acc_seg: 86.9333 aux.loss_ce: 1.7419 aux.acc_seg: 43.2981 +04/19 07:05:33 - mmengine - INFO - Iter(train) [ 350/160000] base_lr: 2.3282e-05 lr: 8.6079e-08 eta: 1 day, 22:00:34 time: 1.0144 data_time: 0.0043 memory: 8462 loss: 3.8221 decode.loss_ce: 2.1781 decode.acc_seg: 91.3601 aux.loss_ce: 1.6440 aux.acc_seg: 35.6861 +04/19 07:06:24 - mmengine - INFO - Iter(train) [ 400/160000] base_lr: 2.6618e-05 lr: 9.8412e-08 eta: 1 day, 21:51:13 time: 1.0113 data_time: 0.0041 memory: 8462 loss: 3.1464 decode.loss_ce: 1.6124 decode.acc_seg: 94.2289 aux.loss_ce: 1.5340 aux.acc_seg: 58.6138 +04/19 07:07:14 - mmengine - INFO - Iter(train) [ 450/160000] base_lr: 2.9953e-05 lr: 1.1074e-07 eta: 1 day, 21:43:28 time: 1.0115 data_time: 0.0042 memory: 8462 loss: 2.4739 decode.loss_ce: 1.0814 decode.acc_seg: 97.0388 aux.loss_ce: 1.3925 aux.acc_seg: 77.4273 +04/19 07:08:05 - mmengine - INFO - Iter(train) [ 500/160000] base_lr: 3.3289e-05 lr: 1.2308e-07 eta: 1 day, 21:37:15 time: 1.0108 data_time: 0.0045 memory: 8462 loss: 1.9593 decode.loss_ce: 0.7072 decode.acc_seg: 96.0545 aux.loss_ce: 1.2521 aux.acc_seg: 82.0532 +04/19 07:08:55 - mmengine - INFO - Iter(train) [ 550/160000] base_lr: 3.6624e-05 lr: 1.3541e-07 eta: 1 day, 21:31:53 time: 1.0116 data_time: 0.0042 memory: 8462 loss: 1.5581 decode.loss_ce: 0.4585 decode.acc_seg: 97.1006 aux.loss_ce: 1.0996 aux.acc_seg: 89.4524 +04/19 07:09:46 - mmengine - INFO - Iter(train) [ 600/160000] base_lr: 3.9960e-05 lr: 1.4774e-07 eta: 1 day, 21:26:46 time: 1.0079 data_time: 0.0042 memory: 8462 loss: 1.2520 decode.loss_ce: 0.3212 decode.acc_seg: 97.2136 aux.loss_ce: 0.9308 aux.acc_seg: 95.9038 +04/19 07:10:36 - mmengine - INFO - Iter(train) [ 650/160000] base_lr: 4.3296e-05 lr: 1.6007e-07 eta: 1 day, 21:21:33 time: 1.0020 data_time: 0.0038 memory: 8462 loss: 1.0009 decode.loss_ce: 0.2412 decode.acc_seg: 96.5857 aux.loss_ce: 0.7596 aux.acc_seg: 95.6675 +04/19 07:11:26 - mmengine - INFO - Iter(train) [ 700/160000] base_lr: 4.6631e-05 lr: 1.7240e-07 eta: 1 day, 21:16:14 time: 1.0006 data_time: 0.0042 memory: 8462 loss: 0.7967 decode.loss_ce: 0.1877 decode.acc_seg: 97.4550 aux.loss_ce: 0.6090 aux.acc_seg: 96.2496 +04/19 07:12:16 - mmengine - INFO - Iter(train) [ 750/160000] base_lr: 4.9967e-05 lr: 1.8474e-07 eta: 1 day, 21:11:03 time: 0.9981 data_time: 0.0041 memory: 8462 loss: 0.5978 decode.loss_ce: 0.1384 decode.acc_seg: 97.7348 aux.loss_ce: 0.4594 aux.acc_seg: 97.1413 +04/19 07:13:06 - mmengine - INFO - Iter(train) [ 800/160000] base_lr: 5.3302e-05 lr: 1.9707e-07 eta: 1 day, 21:06:08 time: 0.9967 data_time: 0.0040 memory: 8462 loss: 0.5116 decode.loss_ce: 0.1474 decode.acc_seg: 95.7693 aux.loss_ce: 0.3642 aux.acc_seg: 95.3093 +04/19 07:13:56 - mmengine - INFO - Iter(train) [ 850/160000] base_lr: 5.6638e-05 lr: 2.0940e-07 eta: 1 day, 21:01:39 time: 0.9975 data_time: 0.0044 memory: 8462 loss: 0.3884 decode.loss_ce: 0.1213 decode.acc_seg: 98.1377 aux.loss_ce: 0.2671 aux.acc_seg: 96.9400 +04/19 07:14:45 - mmengine - INFO - Iter(train) [ 900/160000] base_lr: 5.9973e-05 lr: 2.2173e-07 eta: 1 day, 20:57:36 time: 0.9968 data_time: 0.0042 memory: 8462 loss: 0.3336 decode.loss_ce: 0.1094 decode.acc_seg: 97.8647 aux.loss_ce: 0.2242 aux.acc_seg: 96.1260 +04/19 07:15:35 - mmengine - INFO - Iter(train) [ 950/160000] base_lr: 6.3309e-05 lr: 2.3407e-07 eta: 1 day, 20:53:56 time: 0.9966 data_time: 0.0043 memory: 8462 loss: 0.2584 decode.loss_ce: 0.0931 decode.acc_seg: 97.6976 aux.loss_ce: 0.1653 aux.acc_seg: 97.0474 +04/19 07:16:25 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_ade20k-512x512_20240419_065922 +04/19 07:16:25 - mmengine - INFO - Iter(train) [ 1000/160000] base_lr: 6.6644e-05 lr: 2.4640e-07 eta: 1 day, 20:50:32 time: 0.9976 data_time: 0.0051 memory: 8462 loss: 0.2484 decode.loss_ce: 0.1043 decode.acc_seg: 96.5818 aux.loss_ce: 0.1441 aux.acc_seg: 96.0867 +04/19 07:17:15 - mmengine - INFO - Iter(train) [ 1050/160000] base_lr: 6.9980e-05 lr: 2.5873e-07 eta: 1 day, 20:47:24 time: 0.9968 data_time: 0.0044 memory: 8462 loss: 0.2135 decode.loss_ce: 0.0918 decode.acc_seg: 97.1264 aux.loss_ce: 0.1217 aux.acc_seg: 96.5210 +04/19 07:18:05 - mmengine - INFO - Iter(train) [ 1100/160000] base_lr: 7.3316e-05 lr: 2.7106e-07 eta: 1 day, 20:44:25 time: 0.9965 data_time: 0.0043 memory: 8462 loss: 0.2053 decode.loss_ce: 0.0867 decode.acc_seg: 96.3804 aux.loss_ce: 0.1187 aux.acc_seg: 95.7396 +04/19 07:18:55 - mmengine - INFO - Iter(train) [ 1150/160000] base_lr: 7.6651e-05 lr: 2.8339e-07 eta: 1 day, 20:41:40 time: 0.9979 data_time: 0.0045 memory: 8462 loss: 0.1674 decode.loss_ce: 0.0742 decode.acc_seg: 97.7476 aux.loss_ce: 0.0932 aux.acc_seg: 96.7743 +04/19 07:19:45 - mmengine - INFO - Iter(train) [ 1200/160000] base_lr: 7.9987e-05 lr: 2.9573e-07 eta: 1 day, 20:39:08 time: 0.9967 data_time: 0.0042 memory: 8462 loss: 0.1592 decode.loss_ce: 0.0774 decode.acc_seg: 98.5477 aux.loss_ce: 0.0818 aux.acc_seg: 97.7438 +04/19 07:20:34 - mmengine - INFO - Iter(train) [ 1250/160000] base_lr: 8.3322e-05 lr: 3.0806e-07 eta: 1 day, 20:36:46 time: 0.9985 data_time: 0.0044 memory: 8462 loss: 0.1593 decode.loss_ce: 0.0811 decode.acc_seg: 97.6845 aux.loss_ce: 0.0782 aux.acc_seg: 96.8628 +04/19 07:21:24 - mmengine - INFO - Iter(train) [ 1300/160000] base_lr: 8.6658e-05 lr: 3.2039e-07 eta: 1 day, 20:34:27 time: 0.9981 data_time: 0.0043 memory: 8462 loss: 0.1575 decode.loss_ce: 0.0764 decode.acc_seg: 98.2693 aux.loss_ce: 0.0811 aux.acc_seg: 97.4092 +04/19 07:22:14 - mmengine - INFO - Iter(train) [ 1350/160000] base_lr: 8.9993e-05 lr: 3.3272e-07 eta: 1 day, 20:32:15 time: 0.9968 data_time: 0.0042 memory: 8462 loss: 0.1320 decode.loss_ce: 0.0655 decode.acc_seg: 98.1331 aux.loss_ce: 0.0666 aux.acc_seg: 97.2891 +04/19 07:23:04 - mmengine - INFO - Iter(train) [ 1400/160000] base_lr: 9.3329e-05 lr: 3.4506e-07 eta: 1 day, 20:30:10 time: 0.9974 data_time: 0.0048 memory: 8462 loss: 0.1308 decode.loss_ce: 0.0717 decode.acc_seg: 98.7215 aux.loss_ce: 0.0591 aux.acc_seg: 97.6765 +04/19 07:23:54 - mmengine - INFO - Iter(train) [ 1450/160000] base_lr: 9.6664e-05 lr: 3.5739e-07 eta: 1 day, 20:28:10 time: 0.9978 data_time: 0.0047 memory: 8462 loss: 0.1409 decode.loss_ce: 0.0761 decode.acc_seg: 97.6967 aux.loss_ce: 0.0648 aux.acc_seg: 96.7949 +04/19 07:24:44 - mmengine - INFO - Iter(train) [ 1500/160000] base_lr: 1.0000e-04 lr: 3.6972e-07 eta: 1 day, 20:26:20 time: 0.9988 data_time: 0.0042 memory: 8462 loss: 0.1213 decode.loss_ce: 0.0638 decode.acc_seg: 97.8680 aux.loss_ce: 0.0575 aux.acc_seg: 97.0081 +04/19 07:25:34 - mmengine - INFO - Iter(train) [ 1550/160000] base_lr: 9.9969e-05 lr: 3.6961e-07 eta: 1 day, 20:24:31 time: 0.9977 data_time: 0.0038 memory: 8462 loss: 0.1194 decode.loss_ce: 0.0642 decode.acc_seg: 98.4535 aux.loss_ce: 0.0553 aux.acc_seg: 96.6642 +04/19 07:26:24 - mmengine - INFO - Iter(train) [ 1600/160000] base_lr: 9.9938e-05 lr: 3.6949e-07 eta: 1 day, 20:22:50 time: 0.9984 data_time: 0.0045 memory: 8462 loss: 0.1192 decode.loss_ce: 0.0650 decode.acc_seg: 98.8091 aux.loss_ce: 0.0542 aux.acc_seg: 97.6425 +04/19 07:27:14 - mmengine - INFO - Iter(train) [ 1650/160000] base_lr: 9.9906e-05 lr: 3.6937e-07 eta: 1 day, 20:21:09 time: 0.9986 data_time: 0.0043 memory: 8462 loss: 0.1103 decode.loss_ce: 0.0602 decode.acc_seg: 97.7407 aux.loss_ce: 0.0500 aux.acc_seg: 96.1254 +04/19 07:28:04 - mmengine - INFO - Iter(train) [ 1700/160000] base_lr: 9.9874e-05 lr: 3.6926e-07 eta: 1 day, 20:19:34 time: 0.9983 data_time: 0.0041 memory: 8462 loss: 0.0946 decode.loss_ce: 0.0512 decode.acc_seg: 98.4478 aux.loss_ce: 0.0435 aux.acc_seg: 98.2590 +WARNING:torch.distributed.elastic.agent.server.api:Received 2 death signal, shutting down workers +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 314189 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 314190 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 314191 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 314192 closing signal SIGINT +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 314189 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 314190 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 314191 closing signal SIGTERM +WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 314192 closing signal SIGTERM +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 709, in run + result = self._invoke_run(role) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 850, in _invoke_run + time.sleep(monitor_interval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 314155 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 716, in run + self._shutdown(e.sigval) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 314155 got signal: 2 + +During handling of the above exception, another exception occurred: + +Traceback (most recent call last): + File "/opt/conda/lib/python3.8/runpy.py", line 194, in _run_module_as_main + return _run_code(code, main_globals, None, + File "/opt/conda/lib/python3.8/runpy.py", line 87, in _run_code + exec(code, run_globals) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 193, in + main() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 189, in main + launch(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py", line 174, in launch + run(args) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/run.py", line 715, in run + elastic_launch( + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in __call__ + return launch_agent(self._config, self._entrypoint, list(args)) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 236, in launch_agent + result = agent.run() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/metrics/api.py", line 125, in wrapper + result = f(*args, **kwargs) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/api.py", line 721, in run + self._shutdown() + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/agent/server/local_elastic_agent.py", line 190, in _shutdown + self._pcontext.close(death_sig) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 330, in close + self._close(death_sig=death_sig, timeout=timeout) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 720, in _close + handler.proc.wait(time_to_wait) + File "/opt/conda/lib/python3.8/subprocess.py", line 1083, in wait + return self._wait(timeout=timeout) + File "/opt/conda/lib/python3.8/subprocess.py", line 1800, in _wait + time.sleep(delay) + File "/opt/conda/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/api.py", line 60, in _terminate_process_handler + raise SignalException(f"Process {os.getpid()} got signal: {sigval}", sigval=sigval) +torch.distributed.elastic.multiprocessing.api.SignalException: Process 314155 got signal: 2 +/opt/conda/lib/python3.8/site-packages/torch/distributed/launch.py:178: FutureWarning: The module torch.distributed.launch is deprecated +and will be removed in future. Use torchrun. +Note that --use_env is set by default in torchrun. +If your script expects `--local_rank` argument to be set, please +change it to read from `os.environ['LOCAL_RANK']` instead. See +https://pytorch.org/docs/stable/distributed.html#launch-utility for +further instructions + + warnings.warn( +WARNING:torch.distributed.run: +***************************************** +Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +***************************************** +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +/opt/conda/lib/python3.8/site-packages/mmengine/utils/dl_utils/setup_env.py:56: UserWarning: Setting MKL_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. + warnings.warn( +04/19 07:32:08 - mmengine - INFO - +------------------------------------------------------------ +System environment: + sys.platform: linux + Python: 3.8.12 (default, Oct 12 2021, 13:49:34) [GCC 7.5.0] + CUDA available: True + MUSA available: False + numpy_random_seed: 930945155 + GPU 0,1,2,3: NVIDIA GeForce RTX 2080 Ti + CUDA_HOME: /usr/local/cuda + NVCC: Cuda compilation tools, release 11.3, V11.3.109 + GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 + PyTorch: 1.11.0 + PyTorch compiling details: PyTorch built with: + - GCC 7.3 + - C++ Version: 201402 + - Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications + - Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e) + - OpenMP 201511 (a.k.a. OpenMP 4.5) + - LAPACK is enabled (usually provided by MKL) + - NNPACK is enabled + - CPU capability usage: AVX2 + - CUDA Runtime 11.3 + - NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37 + - CuDNN 8.2 + - Magma 2.5.2 + - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, + + TorchVision: 0.12.0 + OpenCV: 4.9.0 + MMEngine: 0.10.3 + +Runtime environment: + cudnn_benchmark: True + mp_cfg: {'mp_start_method': 'fork', 'opencv_num_threads': 0} + dist_cfg: {'backend': 'nccl'} + seed: 930945155 + Distributed launcher: pytorch + Distributed training: True + GPU number: 4 +------------------------------------------------------------ + +04/19 07:32:09 - mmengine - INFO - Config: +crop_size = ( + 512, + 512, +) +data_preprocessor = dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor') +data_root = 'data/cag' +dataset_type = 'CoronaryAngiographyDataset' +default_hooks = dict( + checkpoint=dict(by_epoch=False, interval=10000, type='CheckpointHook'), + logger=dict(interval=50, log_metric_by_epoch=False, type='LoggerHook'), + param_scheduler=dict(type='ParamSchedulerHook'), + sampler_seed=dict(type='DistSamplerSeedHook'), + timer=dict(type='IterTimerHook'), + visualization=dict(type='SegVisualizationHook')) +default_scope = 'mmseg' +env_cfg = dict( + cudnn_benchmark=True, + dist_cfg=dict(backend='nccl'), + mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0)) +fp16 = dict(loss_scale='dynamic') +img_ratios = [ + 1.0, +] +launcher = 'pytorch' +load_from = None +log_level = 'INFO' +log_processor = dict(by_epoch=False) +model = dict( + auxiliary_head=dict( + align_corners=False, + channels=256, + concat_input=False, + dropout_ratio=0.1, + in_channels=768, + in_index=2, + loss_decode=dict( + loss_weight=0.4, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + num_convs=1, + type='FCNHead'), + backbone=dict( + act_cfg=dict(type='GELU'), + attn_drop_rate=0.0, + drop_path_rate=0.1, + embed_dims=768, + img_size=( + 512, + 512, + ), + in_channels=3, + init_values=1.0, + mlp_ratio=4, + norm_cfg=dict(eps=1e-06, type='LN'), + norm_eval=False, + num_heads=12, + num_layers=12, + out_indices=[ + 3, + 5, + 7, + 11, + ], + patch_size=16, + type='MAE'), + data_preprocessor=dict( + bgr_to_rgb=True, + mean=[ + 123.675, + 116.28, + 103.53, + ], + pad_val=0, + seg_pad_val=255, + size=( + 512, + 512, + ), + std=[ + 58.395, + 57.12, + 57.375, + ], + type='SegDataPreProcessor'), + decode_head=dict( + align_corners=False, + channels=768, + dropout_ratio=0.1, + in_channels=[ + 768, + 768, + 768, + 768, + ], + in_index=[ + 0, + 1, + 2, + 3, + ], + loss_decode=dict( + loss_weight=1.0, type='CrossEntropyLoss', use_sigmoid=False), + norm_cfg=dict(requires_grad=True, type='SyncBN'), + num_classes=150, + pool_scales=( + 1, + 2, + 3, + 6, + ), + type='UPerHead'), + neck=dict( + embed_dim=768, rescales=[ + 4, + 2, + 1, + 0.5, + ], type='Feature2Pyramid'), + pretrained= + '/workspaces/mmsegmentation-1/configs/mae/mae_pretrain_vit_base.pth', + test_cfg=dict(crop_size=( + 512, + 512, + ), mode='slide', stride=( + 341, + 341, + )), + train_cfg=dict(), + type='EncoderDecoder') +norm_cfg = dict(requires_grad=True, type='SyncBN') +optim_wrapper = dict( + constructor='LayerDecayOptimizerConstructor', + optimizer=dict( + betas=( + 0.9, + 0.999, + ), lr=0.0001, type='AdamW', weight_decay=0.05), + paramwise_cfg=dict(layer_decay_rate=0.65, num_layers=12), + type='OptimWrapper') +optimizer = dict(lr=0.01, momentum=0.9, type='SGD', weight_decay=0.0005) +param_scheduler = [ + dict( + begin=0, by_epoch=False, end=1500, start_factor=1e-06, + type='LinearLR'), + dict( + begin=1500, + by_epoch=False, + end=160000, + eta_min=0.0, + power=1.0, + type='PolyLR'), +] +resume = False +test_cfg = dict(type='TestLoop') +test_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/test', seg_map_path='annotations/test'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +test_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +test_pipeline = [ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), +] +train_cfg = dict( + max_iters=160000, type='IterBasedTrainLoop', val_interval=10000) +train_dataloader = dict( + batch_size=2, + dataset=dict( + data_prefix=dict( + img_path='images/training', seg_map_path='annotations/training'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=True, type='InfiniteSampler')) +train_pipeline = [ + dict(type='LoadImageFromFile'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict( + border_mode=0, + p=1, + rotate_limit=20, + scale_limit=( + -0.2, + 0, + ), + shift_limit=0.1, + type='AlbuShiftScaleRotateTransform', + value=[ + 0.3, + 0.4, + 0.5, + ]), + dict( + brightness_limit=( + -0.2, + 0.2, + ), + contrast_limit=( + -0.2, + 0.2, + ), + p=0.5, + type='AlbuRandomContrastTransform'), + dict(p=0.5, type='AlbuGaussNoiseTransform', var_limit=( + 0, + 0.01, + )), + dict(prob=0.5, type='RandomFlip'), + dict(type='PackSegInputs'), +] +tta_model = dict(type='SegTTAModel') +tta_pipeline = [ + dict(backend_args=None, type='LoadImageFromFile'), + dict( + transforms=[ + [ + dict(keep_ratio=True, scale_factor=1.0, type='Resize'), + ], + [ + dict(direction='horizontal', prob=0.0, type='RandomFlip'), + dict(direction='horizontal', prob=1.0, type='RandomFlip'), + ], + [ + dict(type='LoadAnnotations'), + ], + [ + dict(type='PackSegInputs'), + ], + ], + type='TestTimeAug'), +] +val_cfg = dict(type='ValLoop') +val_dataloader = dict( + batch_size=1, + dataset=dict( + data_prefix=dict( + img_path='images/validation', + seg_map_path='annotations/validation'), + data_root='data/cag', + pipeline=[ + dict(type='LoadImageFromFile'), + dict(keep_ratio=True, scale=( + 512, + 512, + ), type='Resize'), + dict(reduce_zero_label=False, type='LoadAnnotations'), + dict(type='PackSegInputs'), + ], + reduce_zero_label=False, + type='CoronaryAngiographyDataset'), + num_workers=4, + persistent_workers=True, + sampler=dict(shuffle=False, type='DefaultSampler')) +val_evaluator = dict( + iou_metrics=[ + 'mIoU', + 'mFscore', + ], type='IoUMetric') +vis_backends = [ + dict(type='LocalVisBackend'), +] +visualizer = dict( + name='visualizer', + type='SegLocalVisualizer', + vis_backends=[ + dict(type='LocalVisBackend'), + ]) +work_dir = '/workspaces/mmsegmentation-1/work_dirs/MAE_Pre_4000' + +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/backbones/beit.py:299: UserWarning: DeprecationWarning: pretrained is deprecated, please use "init_cfg" instead + warnings.warn('DeprecationWarning: pretrained is deprecated, ' +/opt/conda/lib/python3.8/site-packages/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at /opt/conda/conda-bld/pytorch_1646755903507/work/aten/src/ATen/native/TensorShape.cpp:2228.) + return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/models/builder.py:36: UserWarning: ``build_loss`` would be deprecated soon, please use ``mmseg.registry.MODELS.build()`` + warnings.warn('``build_loss`` would be deprecated soon, please use ' +/workspaces/mmsegmentation-1/mmseg/models/losses/cross_entropy_loss.py:235: UserWarning: Default ``avg_non_ignore`` is False, if you would like to ignore the certain label and average loss over non-ignore labels, which is the same with PyTorch official cross_entropy, set ``avg_non_ignore=True``. + warnings.warn( +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +/workspaces/mmsegmentation-1/mmseg/engine/hooks/visualization_hook.py:61: UserWarning: The draw is False, it means that the hook for visualization will not take effect. The results will NOT be visualized or stored. + warnings.warn('The draw is False, it means that the ' +04/19 07:32:12 - mmengine - INFO - Hooks will be executed in the following order: +before_run: +(VERY_HIGH ) RuntimeInfoHook +(BELOW_NORMAL) LoggerHook + -------------------- +before_train: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_train_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) DistSamplerSeedHook + -------------------- +before_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook + -------------------- +after_train_iter: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_train_epoch: +(NORMAL ) IterTimerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +before_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_val_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_val_iter: +(NORMAL ) IterTimerHook + -------------------- +after_val_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_val_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook +(LOW ) ParamSchedulerHook +(VERY_LOW ) CheckpointHook + -------------------- +after_val: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_train: +(VERY_HIGH ) RuntimeInfoHook +(VERY_LOW ) CheckpointHook + -------------------- +before_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +before_test_epoch: +(NORMAL ) IterTimerHook + -------------------- +before_test_iter: +(NORMAL ) IterTimerHook + -------------------- +after_test_iter: +(NORMAL ) IterTimerHook +(NORMAL ) SegVisualizationHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test_epoch: +(VERY_HIGH ) RuntimeInfoHook +(NORMAL ) IterTimerHook +(BELOW_NORMAL) LoggerHook + -------------------- +after_test: +(VERY_HIGH ) RuntimeInfoHook + -------------------- +after_run: +(BELOW_NORMAL) LoggerHook + -------------------- +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65}set param backbone.layers.0.ln1.bias as id 1 + +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.cls_token as id 0set param backbone.layers.0.ln2.weight as id 1 + +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.pos_embed as id 0 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.0.ln1.weight as id 1set param backbone.layers.1.gamma_2 as id 2 + +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2set param backbone.layers.0.attn.relative_position_bias_table as id 1 + +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.1.ffn.layers.1.weight as id 2set param backbone.layers.0.ffn.layers.0.0.bias as id 1 + +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.1.ln1.weight as id 2set param backbone.layers.2.attn.qkv.bias as id 3 + +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2set param backbone.layers.2.ffn.layers.0.0.weight as id 3 + +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2set param backbone.layers.2.ffn.layers.1.weight as id 3 + +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2set param backbone.layers.3.gamma_1 as id 4 + +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.3.ln1.weight as id 4set param backbone.layers.1.ffn.layers.1.bias as id 2 + +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4set param backbone.layers.2.gamma_1 as id 3 + +set param backbone.layers.3.attn.qkv.weight as id 4set param backbone.layers.2.gamma_2 as id 3 + +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3set param backbone.layers.3.attn.proj.weight as id 4 + +set param backbone.layers.3.attn.proj.bias as id 4set param backbone.layers.2.attn.relative_position_bias_table as id 3 + +set param backbone.layers.3.ln2.weight as id 4set param backbone.layers.2.attn.qkv.weight as id 3 + +set param backbone.layers.3.ln2.bias as id 4set param backbone.layers.2.attn.qkv.bias as id 3 + +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.4.gamma_1 as id 5set param backbone.layers.2.ffn.layers.0.0.bias as id 3 + +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5set param backbone.layers.3.gamma_1 as id 4 + +set param backbone.layers.4.attn.qkv.weight as id 5set param backbone.layers.3.gamma_2 as id 4 + +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.3.attn.relative_position_bias_table as id 4set param backbone.layers.4.attn.proj.bias as id 5 + +set param backbone.layers.3.attn.qkv.weight as id 4set param backbone.layers.4.ln2.weight as id 5 + +set param backbone.layers.3.attn.qkv.bias as id 4set param backbone.layers.4.ln2.bias as id 5 + +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5set param backbone.layers.3.attn.proj.bias as id 4 + +set param backbone.layers.4.ffn.layers.0.0.bias as id 5set param backbone.layers.3.ln2.weight as id 4 + +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4set param backbone.layers.5.gamma_1 as id 6 + +set param backbone.layers.5.gamma_2 as id 6set param backbone.layers.3.ffn.layers.1.weight as id 4 + +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.5.attn.proj.weight as id 6set param backbone.layers.4.attn.relative_position_bias_table as id 5 + +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.5.ln2.weight as id 6set param backbone.layers.4.attn.qkv.bias as id 5 + +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.4.ln2.weight as id 5set param backbone.layers.5.ffn.layers.0.0.bias as id 6 + +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5set param backbone.layers.5.ffn.layers.1.bias as id 6 + +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7set param backbone.layers.4.ffn.layers.1.weight as id 5 + +set param backbone.layers.4.ffn.layers.1.bias as id 5set param backbone.layers.6.ln1.weight as id 7 + +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7set param backbone.layers.6.ffn.layers.1.weight as id 7 + +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8set param backbone.layers.6.ln2.weight as id 7 + +set param backbone.layers.6.ln2.bias as id 7set param backbone.layers.7.gamma_2 as id 8 + +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8set param backbone.layers.6.ffn.layers.0.0.weight as id 7 + +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9set param backbone.layers.7.gamma_1 as id 8 + +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.7.attn.qkv.weight as id 8set param backbone.layers.8.attn.proj.weight as id 9 + +set param backbone.layers.7.attn.qkv.bias as id 8set param backbone.layers.8.attn.proj.bias as id 9 + +set param backbone.layers.7.attn.proj.weight as id 8set param backbone.layers.8.ln2.weight as id 9 + +set param backbone.layers.7.attn.proj.bias as id 8set param backbone.layers.8.ln2.bias as id 9 + +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9set param backbone.layers.7.ln2.bias as id 8 + +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.8.attn.qkv.weight as id 9set param backbone.layers.9.attn.qkv.bias as id 10 + +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.8.attn.proj.weight as id 9set param backbone.layers.9.attn.proj.bias as id 10 + +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.8.ln2.weight as id 9set param backbone.layers.9.ln2.bias as id 10 + +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9set param backbone.layers.9.ffn.layers.0.0.bias as id 10 + +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.8.ffn.layers.1.weight as id 9set param backbone.layers.9.ffn.layers.1.bias as id 10 + +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11set param backbone.layers.9.gamma_1 as id 10 + +set param backbone.layers.10.ln1.weight as id 11set param backbone.layers.9.gamma_2 as id 10 + +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.10.attn.qkv.weight as id 11set param backbone.layers.9.attn.relative_position_bias_table as id 10 + +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10set param backbone.layers.10.ffn.layers.0.0.weight as id 11 + +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.10.ffn.layers.1.weight as id 11set param backbone.layers.9.ffn.layers.0.0.bias as id 10 + +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12set param backbone.layers.10.attn.qkv.weight as id 11 + +set param backbone.layers.11.attn.proj.weight as id 12set param backbone.layers.10.attn.qkv.bias as id 11 + +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.11.ln2.weight as id 12set param backbone.layers.10.attn.proj.bias as id 11 + +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param backbone.layers.11.gamma_2 as id 12 +set param neck.upsample_4x.0.bias as id 13 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param neck.upsample_4x.1.weight as id 13 +set param backbone.layers.11.attn.relative_position_bias_table as id 12set param neck.upsample_4x.1.bias as id 13 + +set param neck.upsample_4x.3.weight as id 13 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param neck.upsample_4x.3.bias as id 13 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param neck.upsample_2x.0.weight as id 13set param backbone.layers.11.attn.proj.weight as id 12 + +set param neck.upsample_2x.0.bias as id 13set param backbone.layers.11.attn.proj.bias as id 12 + +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param decode_head.conv_seg.weight as id 13 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param decode_head.conv_seg.bias as id 13 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param neck.upsample_4x.0.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13set param neck.upsample_4x.1.weight as id 13 + +set param neck.upsample_4x.1.bias as id 13set param decode_head.psp_modules.1.1.bn.weight as id 13 + +set param decode_head.psp_modules.1.1.bn.bias as id 13set param neck.upsample_4x.3.weight as id 13 + +set param neck.upsample_4x.3.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13set param neck.upsample_2x.0.weight as id 13 + +set param decode_head.psp_modules.2.1.bn.weight as id 13set param neck.upsample_2x.0.bias as id 13 + +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13set param decode_head.psp_modules.0.1.bn.weight as id 13 + +set param decode_head.lateral_convs.0.bn.weight as id 13set param decode_head.psp_modules.0.1.bn.bias as id 13 + +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13set param decode_head.lateral_convs.1.conv.weight as id 13 + +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13set param decode_head.psp_modules.2.1.conv.weight as id 13 + +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13set param decode_head.fpn_convs.0.bn.bias as id 13 + +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.bottleneck.conv.weight as id 13set param decode_head.fpn_convs.1.bn.weight as id 13 + +set param decode_head.fpn_convs.1.bn.bias as id 13set param decode_head.bottleneck.bn.weight as id 13 + +set param decode_head.bottleneck.bn.bias as id 13set param decode_head.fpn_convs.2.conv.weight as id 13 + +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13set param decode_head.fpn_bottleneck.conv.weight as id 13 + +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +/workspaces/mmsegmentation-1/mmseg/datasets/transforms/loading.py:83: UserWarning: `reduce_zero_label` will be deprecated, if you would like to ignore the zero label, please set `reduce_zero_label=True` when dataset initialized + warnings.warn('`reduce_zero_label` will be deprecated, ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:198: UserWarning: DeprecationWarning: Original LayerDecayOptimizerConstructor of BEiT will be deprecated. Please use LearningRateDecayOptimizerConstructor instead, and set decay_type = layer_wise_vit in paramwise_cfg. + warnings.warn('DeprecationWarning: Original ' +/workspaces/mmsegmentation-1/mmseg/engine/optimizers/layer_decay_optimizer_constructor.py:204: UserWarning: DeprecationWarning: Layer_decay_rate will be deleted, please use decay_rate instead. + warnings.warn('DeprecationWarning: Layer_decay_rate will ' +self.paramwise_cfg is {'num_layers': 12, 'decay_type': 'layer_wise_vit', 'decay_rate': 0.65} +Build LearningRateDecayOptimizerConstructor layer_wise_vit 0.65 - 14 +set param backbone.cls_token as id 0 +set param backbone.pos_embed as id 0 +set param backbone.patch_embed.projection.weight as id 0 +set param backbone.patch_embed.projection.bias as id 0 +set param backbone.layers.0.gamma_1 as id 1 +set param backbone.layers.0.gamma_2 as id 1 +set param backbone.layers.0.ln1.weight as id 1 +set param backbone.layers.0.ln1.bias as id 1 +set param backbone.layers.0.attn.relative_position_bias_table as id 1 +set param backbone.layers.0.attn.qkv.weight as id 1 +set param backbone.layers.0.attn.qkv.bias as id 1 +set param backbone.layers.0.attn.proj.weight as id 1 +set param backbone.layers.0.attn.proj.bias as id 1 +set param backbone.layers.0.ln2.weight as id 1 +set param backbone.layers.0.ln2.bias as id 1 +set param backbone.layers.0.ffn.layers.0.0.weight as id 1 +set param backbone.layers.0.ffn.layers.0.0.bias as id 1 +set param backbone.layers.0.ffn.layers.1.weight as id 1 +set param backbone.layers.0.ffn.layers.1.bias as id 1 +set param backbone.layers.1.gamma_1 as id 2 +set param backbone.layers.1.gamma_2 as id 2 +set param backbone.layers.1.ln1.weight as id 2 +set param backbone.layers.1.ln1.bias as id 2 +set param backbone.layers.1.attn.relative_position_bias_table as id 2 +set param backbone.layers.1.attn.qkv.weight as id 2 +set param backbone.layers.1.attn.qkv.bias as id 2 +set param backbone.layers.1.attn.proj.weight as id 2 +set param backbone.layers.1.attn.proj.bias as id 2 +set param backbone.layers.1.ln2.weight as id 2 +set param backbone.layers.1.ln2.bias as id 2 +set param backbone.layers.1.ffn.layers.0.0.weight as id 2 +set param backbone.layers.1.ffn.layers.0.0.bias as id 2 +set param backbone.layers.1.ffn.layers.1.weight as id 2 +set param backbone.layers.1.ffn.layers.1.bias as id 2 +set param backbone.layers.2.gamma_1 as id 3 +set param backbone.layers.2.gamma_2 as id 3 +set param backbone.layers.2.ln1.weight as id 3 +set param backbone.layers.2.ln1.bias as id 3 +set param backbone.layers.2.attn.relative_position_bias_table as id 3 +set param backbone.layers.2.attn.qkv.weight as id 3 +set param backbone.layers.2.attn.qkv.bias as id 3 +set param backbone.layers.2.attn.proj.weight as id 3 +set param backbone.layers.2.attn.proj.bias as id 3 +set param backbone.layers.2.ln2.weight as id 3 +set param backbone.layers.2.ln2.bias as id 3 +set param backbone.layers.2.ffn.layers.0.0.weight as id 3 +set param backbone.layers.2.ffn.layers.0.0.bias as id 3 +set param backbone.layers.2.ffn.layers.1.weight as id 3 +set param backbone.layers.2.ffn.layers.1.bias as id 3 +set param backbone.layers.3.gamma_1 as id 4 +set param backbone.layers.3.gamma_2 as id 4 +set param backbone.layers.3.ln1.weight as id 4 +set param backbone.layers.3.ln1.bias as id 4 +set param backbone.layers.3.attn.relative_position_bias_table as id 4 +set param backbone.layers.3.attn.qkv.weight as id 4 +set param backbone.layers.3.attn.qkv.bias as id 4 +set param backbone.layers.3.attn.proj.weight as id 4 +set param backbone.layers.3.attn.proj.bias as id 4 +set param backbone.layers.3.ln2.weight as id 4 +set param backbone.layers.3.ln2.bias as id 4 +set param backbone.layers.3.ffn.layers.0.0.weight as id 4 +set param backbone.layers.3.ffn.layers.0.0.bias as id 4 +set param backbone.layers.3.ffn.layers.1.weight as id 4 +set param backbone.layers.3.ffn.layers.1.bias as id 4 +set param backbone.layers.4.gamma_1 as id 5 +set param backbone.layers.4.gamma_2 as id 5 +set param backbone.layers.4.ln1.weight as id 5 +set param backbone.layers.4.ln1.bias as id 5 +set param backbone.layers.4.attn.relative_position_bias_table as id 5 +set param backbone.layers.4.attn.qkv.weight as id 5 +set param backbone.layers.4.attn.qkv.bias as id 5 +set param backbone.layers.4.attn.proj.weight as id 5 +set param backbone.layers.4.attn.proj.bias as id 5 +set param backbone.layers.4.ln2.weight as id 5 +set param backbone.layers.4.ln2.bias as id 5 +set param backbone.layers.4.ffn.layers.0.0.weight as id 5 +set param backbone.layers.4.ffn.layers.0.0.bias as id 5 +set param backbone.layers.4.ffn.layers.1.weight as id 5 +set param backbone.layers.4.ffn.layers.1.bias as id 5 +set param backbone.layers.5.gamma_1 as id 6 +set param backbone.layers.5.gamma_2 as id 6 +set param backbone.layers.5.ln1.weight as id 6 +set param backbone.layers.5.ln1.bias as id 6 +set param backbone.layers.5.attn.relative_position_bias_table as id 6 +set param backbone.layers.5.attn.qkv.weight as id 6 +set param backbone.layers.5.attn.qkv.bias as id 6 +set param backbone.layers.5.attn.proj.weight as id 6 +set param backbone.layers.5.attn.proj.bias as id 6 +set param backbone.layers.5.ln2.weight as id 6 +set param backbone.layers.5.ln2.bias as id 6 +set param backbone.layers.5.ffn.layers.0.0.weight as id 6 +set param backbone.layers.5.ffn.layers.0.0.bias as id 6 +set param backbone.layers.5.ffn.layers.1.weight as id 6 +set param backbone.layers.5.ffn.layers.1.bias as id 6 +set param backbone.layers.6.gamma_1 as id 7 +set param backbone.layers.6.gamma_2 as id 7 +set param backbone.layers.6.ln1.weight as id 7 +set param backbone.layers.6.ln1.bias as id 7 +set param backbone.layers.6.attn.relative_position_bias_table as id 7 +set param backbone.layers.6.attn.qkv.weight as id 7 +set param backbone.layers.6.attn.qkv.bias as id 7 +set param backbone.layers.6.attn.proj.weight as id 7 +set param backbone.layers.6.attn.proj.bias as id 7 +set param backbone.layers.6.ln2.weight as id 7 +set param backbone.layers.6.ln2.bias as id 7 +set param backbone.layers.6.ffn.layers.0.0.weight as id 7 +set param backbone.layers.6.ffn.layers.0.0.bias as id 7 +set param backbone.layers.6.ffn.layers.1.weight as id 7 +set param backbone.layers.6.ffn.layers.1.bias as id 7 +set param backbone.layers.7.gamma_1 as id 8 +set param backbone.layers.7.gamma_2 as id 8 +set param backbone.layers.7.ln1.weight as id 8 +set param backbone.layers.7.ln1.bias as id 8 +set param backbone.layers.7.attn.relative_position_bias_table as id 8 +set param backbone.layers.7.attn.qkv.weight as id 8 +set param backbone.layers.7.attn.qkv.bias as id 8 +set param backbone.layers.7.attn.proj.weight as id 8 +set param backbone.layers.7.attn.proj.bias as id 8 +set param backbone.layers.7.ln2.weight as id 8 +set param backbone.layers.7.ln2.bias as id 8 +set param backbone.layers.7.ffn.layers.0.0.weight as id 8 +set param backbone.layers.7.ffn.layers.0.0.bias as id 8 +set param backbone.layers.7.ffn.layers.1.weight as id 8 +set param backbone.layers.7.ffn.layers.1.bias as id 8 +set param backbone.layers.8.gamma_1 as id 9 +set param backbone.layers.8.gamma_2 as id 9 +set param backbone.layers.8.ln1.weight as id 9 +set param backbone.layers.8.ln1.bias as id 9 +set param backbone.layers.8.attn.relative_position_bias_table as id 9 +set param backbone.layers.8.attn.qkv.weight as id 9 +set param backbone.layers.8.attn.qkv.bias as id 9 +set param backbone.layers.8.attn.proj.weight as id 9 +set param backbone.layers.8.attn.proj.bias as id 9 +set param backbone.layers.8.ln2.weight as id 9 +set param backbone.layers.8.ln2.bias as id 9 +set param backbone.layers.8.ffn.layers.0.0.weight as id 9 +set param backbone.layers.8.ffn.layers.0.0.bias as id 9 +set param backbone.layers.8.ffn.layers.1.weight as id 9 +set param backbone.layers.8.ffn.layers.1.bias as id 9 +set param backbone.layers.9.gamma_1 as id 10 +set param backbone.layers.9.gamma_2 as id 10 +set param backbone.layers.9.ln1.weight as id 10 +set param backbone.layers.9.ln1.bias as id 10 +set param backbone.layers.9.attn.relative_position_bias_table as id 10 +set param backbone.layers.9.attn.qkv.weight as id 10 +set param backbone.layers.9.attn.qkv.bias as id 10 +set param backbone.layers.9.attn.proj.weight as id 10 +set param backbone.layers.9.attn.proj.bias as id 10 +set param backbone.layers.9.ln2.weight as id 10 +set param backbone.layers.9.ln2.bias as id 10 +set param backbone.layers.9.ffn.layers.0.0.weight as id 10 +set param backbone.layers.9.ffn.layers.0.0.bias as id 10 +set param backbone.layers.9.ffn.layers.1.weight as id 10 +set param backbone.layers.9.ffn.layers.1.bias as id 10 +set param backbone.layers.10.gamma_1 as id 11 +set param backbone.layers.10.gamma_2 as id 11 +set param backbone.layers.10.ln1.weight as id 11 +set param backbone.layers.10.ln1.bias as id 11 +set param backbone.layers.10.attn.relative_position_bias_table as id 11 +set param backbone.layers.10.attn.qkv.weight as id 11 +set param backbone.layers.10.attn.qkv.bias as id 11 +set param backbone.layers.10.attn.proj.weight as id 11 +set param backbone.layers.10.attn.proj.bias as id 11 +set param backbone.layers.10.ln2.weight as id 11 +set param backbone.layers.10.ln2.bias as id 11 +set param backbone.layers.10.ffn.layers.0.0.weight as id 11 +set param backbone.layers.10.ffn.layers.0.0.bias as id 11 +set param backbone.layers.10.ffn.layers.1.weight as id 11 +set param backbone.layers.10.ffn.layers.1.bias as id 11 +set param backbone.layers.11.gamma_1 as id 12 +set param backbone.layers.11.gamma_2 as id 12 +set param backbone.layers.11.ln1.weight as id 12 +set param backbone.layers.11.ln1.bias as id 12 +set param backbone.layers.11.attn.relative_position_bias_table as id 12 +set param backbone.layers.11.attn.qkv.weight as id 12 +set param backbone.layers.11.attn.qkv.bias as id 12 +set param backbone.layers.11.attn.proj.weight as id 12 +set param backbone.layers.11.attn.proj.bias as id 12 +set param backbone.layers.11.ln2.weight as id 12 +set param backbone.layers.11.ln2.bias as id 12 +set param backbone.layers.11.ffn.layers.0.0.weight as id 12 +set param backbone.layers.11.ffn.layers.0.0.bias as id 12 +set param backbone.layers.11.ffn.layers.1.weight as id 12 +set param backbone.layers.11.ffn.layers.1.bias as id 12 +set param neck.upsample_4x.0.weight as id 13 +set param neck.upsample_4x.0.bias as id 13 +set param neck.upsample_4x.1.weight as id 13 +set param neck.upsample_4x.1.bias as id 13 +set param neck.upsample_4x.3.weight as id 13 +set param neck.upsample_4x.3.bias as id 13 +set param neck.upsample_2x.0.weight as id 13 +set param neck.upsample_2x.0.bias as id 13 +set param decode_head.conv_seg.weight as id 13 +set param decode_head.conv_seg.bias as id 13 +set param decode_head.psp_modules.0.1.conv.weight as id 13 +set param decode_head.psp_modules.0.1.bn.weight as id 13 +set param decode_head.psp_modules.0.1.bn.bias as id 13 +set param decode_head.psp_modules.1.1.conv.weight as id 13 +set param decode_head.psp_modules.1.1.bn.weight as id 13 +set param decode_head.psp_modules.1.1.bn.bias as id 13 +set param decode_head.psp_modules.2.1.conv.weight as id 13 +set param decode_head.psp_modules.2.1.bn.weight as id 13 +set param decode_head.psp_modules.2.1.bn.bias as id 13 +set param decode_head.psp_modules.3.1.conv.weight as id 13 +set param decode_head.psp_modules.3.1.bn.weight as id 13 +set param decode_head.psp_modules.3.1.bn.bias as id 13 +set param decode_head.bottleneck.conv.weight as id 13 +set param decode_head.bottleneck.bn.weight as id 13 +set param decode_head.bottleneck.bn.bias as id 13 +set param decode_head.lateral_convs.0.conv.weight as id 13 +set param decode_head.lateral_convs.0.bn.weight as id 13 +set param decode_head.lateral_convs.0.bn.bias as id 13 +set param decode_head.lateral_convs.1.conv.weight as id 13 +set param decode_head.lateral_convs.1.bn.weight as id 13 +set param decode_head.lateral_convs.1.bn.bias as id 13 +set param decode_head.lateral_convs.2.conv.weight as id 13 +set param decode_head.lateral_convs.2.bn.weight as id 13 +set param decode_head.lateral_convs.2.bn.bias as id 13 +set param decode_head.fpn_convs.0.conv.weight as id 13 +set param decode_head.fpn_convs.0.bn.weight as id 13 +set param decode_head.fpn_convs.0.bn.bias as id 13 +set param decode_head.fpn_convs.1.conv.weight as id 13 +set param decode_head.fpn_convs.1.bn.weight as id 13 +set param decode_head.fpn_convs.1.bn.bias as id 13 +set param decode_head.fpn_convs.2.conv.weight as id 13 +set param decode_head.fpn_convs.2.bn.weight as id 13 +set param decode_head.fpn_convs.2.bn.bias as id 13 +set param decode_head.fpn_bottleneck.conv.weight as id 13 +set param decode_head.fpn_bottleneck.bn.weight as id 13 +set param decode_head.fpn_bottleneck.bn.bias as id 13 +set param auxiliary_head.conv_seg.weight as id 13 +set param auxiliary_head.conv_seg.bias as id 13 +set param auxiliary_head.convs.0.conv.weight as id 13 +set param auxiliary_head.convs.0.bn.weight as id 13 +set param auxiliary_head.convs.0.bn.bias as id 13 +Param groups = { + "layer_0_decay": { + "param_names": [ + "backbone.cls_token", + "backbone.pos_embed", + "backbone.patch_embed.projection.weight" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.05 + }, + "layer_0_no_decay": { + "param_names": [ + "backbone.patch_embed.projection.bias" + ], + "lr_scale": 0.003697205891018715, + "lr": 3.697205891018715e-07, + "weight_decay": 0.0 + }, + "layer_1_no_decay": { + "param_names": [ + "backbone.layers.0.gamma_1", + "backbone.layers.0.gamma_2", + "backbone.layers.0.ln1.weight", + "backbone.layers.0.ln1.bias", + "backbone.layers.0.attn.qkv.bias", + "backbone.layers.0.attn.proj.bias", + "backbone.layers.0.ln2.weight", + "backbone.layers.0.ln2.bias", + "backbone.layers.0.ffn.layers.0.0.bias", + "backbone.layers.0.ffn.layers.1.bias" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.0 + }, + "layer_1_decay": { + "param_names": [ + "backbone.layers.0.attn.relative_position_bias_table", + "backbone.layers.0.attn.qkv.weight", + "backbone.layers.0.attn.proj.weight", + "backbone.layers.0.ffn.layers.0.0.weight", + "backbone.layers.0.ffn.layers.1.weight" + ], + "lr_scale": 0.005688009063105715, + "lr": 5.688009063105716e-07, + "weight_decay": 0.05 + }, + "layer_2_no_decay": { + "param_names": [ + "backbone.layers.1.gamma_1", + "backbone.layers.1.gamma_2", + "backbone.layers.1.ln1.weight", + "backbone.layers.1.ln1.bias", + "backbone.layers.1.attn.qkv.bias", + "backbone.layers.1.attn.proj.bias", + "backbone.layers.1.ln2.weight", + "backbone.layers.1.ln2.bias", + "backbone.layers.1.ffn.layers.0.0.bias", + "backbone.layers.1.ffn.layers.1.bias" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.0 + }, + "layer_2_decay": { + "param_names": [ + "backbone.layers.1.attn.relative_position_bias_table", + "backbone.layers.1.attn.qkv.weight", + "backbone.layers.1.attn.proj.weight", + "backbone.layers.1.ffn.layers.0.0.weight", + "backbone.layers.1.ffn.layers.1.weight" + ], + "lr_scale": 0.008750783174008792, + "lr": 8.750783174008792e-07, + "weight_decay": 0.05 + }, + "layer_3_no_decay": { + "param_names": [ + "backbone.layers.2.gamma_1", + "backbone.layers.2.gamma_2", + "backbone.layers.2.ln1.weight", + "backbone.layers.2.ln1.bias", + "backbone.layers.2.attn.qkv.bias", + "backbone.layers.2.attn.proj.bias", + "backbone.layers.2.ln2.weight", + "backbone.layers.2.ln2.bias", + "backbone.layers.2.ffn.layers.0.0.bias", + "backbone.layers.2.ffn.layers.1.bias" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.0 + }, + "layer_3_decay": { + "param_names": [ + "backbone.layers.2.attn.relative_position_bias_table", + "backbone.layers.2.attn.qkv.weight", + "backbone.layers.2.attn.proj.weight", + "backbone.layers.2.ffn.layers.0.0.weight", + "backbone.layers.2.ffn.layers.1.weight" + ], + "lr_scale": 0.013462743344628911, + "lr": 1.3462743344628912e-06, + "weight_decay": 0.05 + }, + "layer_4_no_decay": { + "param_names": [ + "backbone.layers.3.gamma_1", + "backbone.layers.3.gamma_2", + "backbone.layers.3.ln1.weight", + "backbone.layers.3.ln1.bias", + "backbone.layers.3.attn.qkv.bias", + "backbone.layers.3.attn.proj.bias", + "backbone.layers.3.ln2.weight", + "backbone.layers.3.ln2.bias", + "backbone.layers.3.ffn.layers.0.0.bias", + "backbone.layers.3.ffn.layers.1.bias" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.0 + }, + "layer_4_decay": { + "param_names": [ + "backbone.layers.3.attn.relative_position_bias_table", + "backbone.layers.3.attn.qkv.weight", + "backbone.layers.3.attn.proj.weight", + "backbone.layers.3.ffn.layers.0.0.weight", + "backbone.layers.3.ffn.layers.1.weight" + ], + "lr_scale": 0.02071191283789063, + "lr": 2.0711912837890633e-06, + "weight_decay": 0.05 + }, + "layer_5_no_decay": { + "param_names": [ + "backbone.layers.4.gamma_1", + "backbone.layers.4.gamma_2", + "backbone.layers.4.ln1.weight", + "backbone.layers.4.ln1.bias", + "backbone.layers.4.attn.qkv.bias", + "backbone.layers.4.attn.proj.bias", + "backbone.layers.4.ln2.weight", + "backbone.layers.4.ln2.bias", + "backbone.layers.4.ffn.layers.0.0.bias", + "backbone.layers.4.ffn.layers.1.bias" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.0 + }, + "layer_5_decay": { + "param_names": [ + "backbone.layers.4.attn.relative_position_bias_table", + "backbone.layers.4.attn.qkv.weight", + "backbone.layers.4.attn.proj.weight", + "backbone.layers.4.ffn.layers.0.0.weight", + "backbone.layers.4.ffn.layers.1.weight" + ], + "lr_scale": 0.03186448128906251, + "lr": 3.1864481289062513e-06, + "weight_decay": 0.05 + }, + "layer_6_no_decay": { + "param_names": [ + "backbone.layers.5.gamma_1", + "backbone.layers.5.gamma_2", + "backbone.layers.5.ln1.weight", + "backbone.layers.5.ln1.bias", + "backbone.layers.5.attn.qkv.bias", + "backbone.layers.5.attn.proj.bias", + "backbone.layers.5.ln2.weight", + "backbone.layers.5.ln2.bias", + "backbone.layers.5.ffn.layers.0.0.bias", + "backbone.layers.5.ffn.layers.1.bias" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.0 + }, + "layer_6_decay": { + "param_names": [ + "backbone.layers.5.attn.relative_position_bias_table", + "backbone.layers.5.attn.qkv.weight", + "backbone.layers.5.attn.proj.weight", + "backbone.layers.5.ffn.layers.0.0.weight", + "backbone.layers.5.ffn.layers.1.weight" + ], + "lr_scale": 0.049022278906250015, + "lr": 4.9022278906250015e-06, + "weight_decay": 0.05 + }, + "layer_7_no_decay": { + "param_names": [ + "backbone.layers.6.gamma_1", + "backbone.layers.6.gamma_2", + "backbone.layers.6.ln1.weight", + "backbone.layers.6.ln1.bias", + "backbone.layers.6.attn.qkv.bias", + "backbone.layers.6.attn.proj.bias", + "backbone.layers.6.ln2.weight", + "backbone.layers.6.ln2.bias", + "backbone.layers.6.ffn.layers.0.0.bias", + "backbone.layers.6.ffn.layers.1.bias" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.0 + }, + "layer_7_decay": { + "param_names": [ + "backbone.layers.6.attn.relative_position_bias_table", + "backbone.layers.6.attn.qkv.weight", + "backbone.layers.6.attn.proj.weight", + "backbone.layers.6.ffn.layers.0.0.weight", + "backbone.layers.6.ffn.layers.1.weight" + ], + "lr_scale": 0.07541889062500001, + "lr": 7.541889062500002e-06, + "weight_decay": 0.05 + }, + "layer_8_no_decay": { + "param_names": [ + "backbone.layers.7.gamma_1", + "backbone.layers.7.gamma_2", + "backbone.layers.7.ln1.weight", + "backbone.layers.7.ln1.bias", + "backbone.layers.7.attn.qkv.bias", + "backbone.layers.7.attn.proj.bias", + "backbone.layers.7.ln2.weight", + "backbone.layers.7.ln2.bias", + "backbone.layers.7.ffn.layers.0.0.bias", + "backbone.layers.7.ffn.layers.1.bias" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.0 + }, + "layer_8_decay": { + "param_names": [ + "backbone.layers.7.attn.relative_position_bias_table", + "backbone.layers.7.attn.qkv.weight", + "backbone.layers.7.attn.proj.weight", + "backbone.layers.7.ffn.layers.0.0.weight", + "backbone.layers.7.ffn.layers.1.weight" + ], + "lr_scale": 0.11602906250000002, + "lr": 1.1602906250000002e-05, + "weight_decay": 0.05 + }, + "layer_9_no_decay": { + "param_names": [ + "backbone.layers.8.gamma_1", + "backbone.layers.8.gamma_2", + "backbone.layers.8.ln1.weight", + "backbone.layers.8.ln1.bias", + "backbone.layers.8.attn.qkv.bias", + "backbone.layers.8.attn.proj.bias", + "backbone.layers.8.ln2.weight", + "backbone.layers.8.ln2.bias", + "backbone.layers.8.ffn.layers.0.0.bias", + "backbone.layers.8.ffn.layers.1.bias" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.0 + }, + "layer_9_decay": { + "param_names": [ + "backbone.layers.8.attn.relative_position_bias_table", + "backbone.layers.8.attn.qkv.weight", + "backbone.layers.8.attn.proj.weight", + "backbone.layers.8.ffn.layers.0.0.weight", + "backbone.layers.8.ffn.layers.1.weight" + ], + "lr_scale": 0.17850625000000003, + "lr": 1.7850625000000003e-05, + "weight_decay": 0.05 + }, + "layer_10_no_decay": { + "param_names": [ + "backbone.layers.9.gamma_1", + "backbone.layers.9.gamma_2", + "backbone.layers.9.ln1.weight", + "backbone.layers.9.ln1.bias", + "backbone.layers.9.attn.qkv.bias", + "backbone.layers.9.attn.proj.bias", + "backbone.layers.9.ln2.weight", + "backbone.layers.9.ln2.bias", + "backbone.layers.9.ffn.layers.0.0.bias", + "backbone.layers.9.ffn.layers.1.bias" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.0 + }, + "layer_10_decay": { + "param_names": [ + "backbone.layers.9.attn.relative_position_bias_table", + "backbone.layers.9.attn.qkv.weight", + "backbone.layers.9.attn.proj.weight", + "backbone.layers.9.ffn.layers.0.0.weight", + "backbone.layers.9.ffn.layers.1.weight" + ], + "lr_scale": 0.274625, + "lr": 2.7462500000000003e-05, + "weight_decay": 0.05 + }, + "layer_11_no_decay": { + "param_names": [ + "backbone.layers.10.gamma_1", + "backbone.layers.10.gamma_2", + "backbone.layers.10.ln1.weight", + "backbone.layers.10.ln1.bias", + "backbone.layers.10.attn.qkv.bias", + "backbone.layers.10.attn.proj.bias", + "backbone.layers.10.ln2.weight", + "backbone.layers.10.ln2.bias", + "backbone.layers.10.ffn.layers.0.0.bias", + "backbone.layers.10.ffn.layers.1.bias" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.0 + }, + "layer_11_decay": { + "param_names": [ + "backbone.layers.10.attn.relative_position_bias_table", + "backbone.layers.10.attn.qkv.weight", + "backbone.layers.10.attn.proj.weight", + "backbone.layers.10.ffn.layers.0.0.weight", + "backbone.layers.10.ffn.layers.1.weight" + ], + "lr_scale": 0.42250000000000004, + "lr": 4.2250000000000004e-05, + "weight_decay": 0.05 + }, + "layer_12_no_decay": { + "param_names": [ + "backbone.layers.11.gamma_1", + "backbone.layers.11.gamma_2", + "backbone.layers.11.ln1.weight", + "backbone.layers.11.ln1.bias", + "backbone.layers.11.attn.qkv.bias", + "backbone.layers.11.attn.proj.bias", + "backbone.layers.11.ln2.weight", + "backbone.layers.11.ln2.bias", + "backbone.layers.11.ffn.layers.0.0.bias", + "backbone.layers.11.ffn.layers.1.bias" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.0 + }, + "layer_12_decay": { + "param_names": [ + "backbone.layers.11.attn.relative_position_bias_table", + "backbone.layers.11.attn.qkv.weight", + "backbone.layers.11.attn.proj.weight", + "backbone.layers.11.ffn.layers.0.0.weight", + "backbone.layers.11.ffn.layers.1.weight" + ], + "lr_scale": 0.65, + "lr": 6.500000000000001e-05, + "weight_decay": 0.05 + }, + "layer_13_decay": { + "param_names": [ + "neck.upsample_4x.0.weight", + "neck.upsample_4x.3.weight", + "neck.upsample_2x.0.weight", + "decode_head.conv_seg.weight", + "decode_head.psp_modules.0.1.conv.weight", + "decode_head.psp_modules.1.1.conv.weight", + "decode_head.psp_modules.2.1.conv.weight", + "decode_head.psp_modules.3.1.conv.weight", + "decode_head.bottleneck.conv.weight", + "decode_head.lateral_convs.0.conv.weight", + "decode_head.lateral_convs.1.conv.weight", + "decode_head.lateral_convs.2.conv.weight", + "decode_head.fpn_convs.0.conv.weight", + "decode_head.fpn_convs.1.conv.weight", + "decode_head.fpn_convs.2.conv.weight", + "decode_head.fpn_bottleneck.conv.weight", + "auxiliary_head.conv_seg.weight", + "auxiliary_head.convs.0.conv.weight" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.05 + }, + "layer_13_no_decay": { + "param_names": [ + "neck.upsample_4x.0.bias", + "neck.upsample_4x.1.weight", + "neck.upsample_4x.1.bias", + "neck.upsample_4x.3.bias", + "neck.upsample_2x.0.bias", + "decode_head.conv_seg.bias", + "decode_head.psp_modules.0.1.bn.weight", + "decode_head.psp_modules.0.1.bn.bias", + "decode_head.psp_modules.1.1.bn.weight", + "decode_head.psp_modules.1.1.bn.bias", + "decode_head.psp_modules.2.1.bn.weight", + "decode_head.psp_modules.2.1.bn.bias", + "decode_head.psp_modules.3.1.bn.weight", + "decode_head.psp_modules.3.1.bn.bias", + "decode_head.bottleneck.bn.weight", + "decode_head.bottleneck.bn.bias", + "decode_head.lateral_convs.0.bn.weight", + "decode_head.lateral_convs.0.bn.bias", + "decode_head.lateral_convs.1.bn.weight", + "decode_head.lateral_convs.1.bn.bias", + "decode_head.lateral_convs.2.bn.weight", + "decode_head.lateral_convs.2.bn.bias", + "decode_head.fpn_convs.0.bn.weight", + "decode_head.fpn_convs.0.bn.bias", + "decode_head.fpn_convs.1.bn.weight", + "decode_head.fpn_convs.1.bn.bias", + "decode_head.fpn_convs.2.bn.weight", + "decode_head.fpn_convs.2.bn.bias", + "decode_head.fpn_bottleneck.bn.weight", + "decode_head.fpn_bottleneck.bn.bias", + "auxiliary_head.conv_seg.bias", + "auxiliary_head.convs.0.bn.weight", + "auxiliary_head.convs.0.bn.bias" + ], + "lr_scale": 1.0, + "lr": 0.0001, + "weight_decay": 0.0 + } +} +04/19 07:32:13 - mmengine - WARNING - The prefix is not set in metric class IoUMetric. +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/configs/mae/mae_pretrain_vit_base.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/configs/mae/mae_pretrain_vit_base.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/configs/mae/mae_pretrain_vit_base.pth +Loads checkpoint by local backend from path: /workspaces/mmsegmentation-1/configs/mae/mae_pretrain_vit_base.pth +04/19 07:32:14 - mmengine - WARNING - "FileClient" will be deprecated in future. Please use io functions in https://mmengine.readthedocs.io/en/latest/api/fileio.html#file-io +04/19 07:32:14 - mmengine - WARNING - "HardDiskBackend" is the alias of "LocalBackend" and the former will be deprecated in future. +04/19 07:32:14 - mmengine - INFO - Checkpoints will be saved to /workspaces/mmsegmentation-1/work_dirs/MAE_Pre_4000. +04/19 07:33:10 - mmengine - INFO - Iter(train) [ 50/160000] base_lr: 3.2689e-06 lr: 1.2086e-08 eta: 2 days, 2:26:12 time: 1.0282 data_time: 0.0044 memory: 8935 loss: 6.8929 decode.loss_ce: 4.8994 decode.acc_seg: 13.5880 aux.loss_ce: 1.9936 aux.acc_seg: 0.0000 +04/19 07:34:02 - mmengine - INFO - Iter(train) [ 100/160000] base_lr: 6.6045e-06 lr: 2.4418e-08 eta: 2 days, 0:05:00 time: 1.0281 data_time: 0.0044 memory: 8462 loss: 6.5884 decode.loss_ce: 4.6293 decode.acc_seg: 34.7708 aux.loss_ce: 1.9591 aux.acc_seg: 11.9486 +04/19 07:34:53 - mmengine - INFO - Iter(train) [ 150/160000] base_lr: 9.9401e-06 lr: 3.6750e-08 eta: 1 day, 23:14:54 time: 1.0264 data_time: 0.0038 memory: 8462 loss: 6.0648 decode.loss_ce: 4.1524 decode.acc_seg: 36.6734 aux.loss_ce: 1.9124 aux.acc_seg: 23.0017 +04/19 07:35:45 - mmengine - INFO - Iter(train) [ 200/160000] base_lr: 1.3276e-05 lr: 4.9083e-08 eta: 1 day, 22:47:50 time: 1.0232 data_time: 0.0040 memory: 8462 loss: 5.5490 decode.loss_ce: 3.6809 decode.acc_seg: 73.1098 aux.loss_ce: 1.8681 aux.acc_seg: 61.9604 +04/19 07:36:35 - mmengine - INFO - Iter(train) [ 250/160000] base_lr: 1.6611e-05 lr: 6.1415e-08 eta: 1 day, 22:28:20 time: 1.0183 data_time: 0.0051 memory: 8462 loss: 5.0071 decode.loss_ce: 3.2241 decode.acc_seg: 74.2155 aux.loss_ce: 1.7830 aux.acc_seg: 79.6803 +04/19 07:37:26 - mmengine - INFO - Iter(train) [ 300/160000] base_lr: 1.9947e-05 lr: 7.3747e-08 eta: 1 day, 22:14:09 time: 1.0157 data_time: 0.0040 memory: 8462 loss: 4.4336 decode.loss_ce: 2.7250 decode.acc_seg: 81.5327 aux.loss_ce: 1.7086 aux.acc_seg: 69.2970 +04/19 07:38:17 - mmengine - INFO - Iter(train) [ 350/160000] base_lr: 2.3282e-05 lr: 8.6079e-08 eta: 1 day, 22:02:23 time: 1.0125 data_time: 0.0042 memory: 8462 loss: 3.8821 decode.loss_ce: 2.2462 decode.acc_seg: 86.4904 aux.loss_ce: 1.6358 aux.acc_seg: 69.9341 +04/19 07:39:08 - mmengine - INFO - Iter(train) [ 400/160000] base_lr: 2.6618e-05 lr: 9.8412e-08 eta: 1 day, 21:52:57 time: 1.0134 data_time: 0.0043 memory: 8462 loss: 3.1226 decode.loss_ce: 1.6190 decode.acc_seg: 94.7613 aux.loss_ce: 1.5036 aux.acc_seg: 83.1755 +04/19 07:39:58 - mmengine - INFO - Iter(train) [ 450/160000] base_lr: 2.9953e-05 lr: 1.1074e-07 eta: 1 day, 21:45:08 time: 1.0109 data_time: 0.0041 memory: 8462 loss: 2.5017 decode.loss_ce: 1.1098 decode.acc_seg: 96.2904 aux.loss_ce: 1.3919 aux.acc_seg: 81.9899 +04/19 07:40:49 - mmengine - INFO - Iter(train) [ 500/160000] base_lr: 3.3289e-05 lr: 1.2308e-07 eta: 1 day, 21:38:21 time: 1.0099 data_time: 0.0041 memory: 8462 loss: 1.9654 decode.loss_ce: 0.7101 decode.acc_seg: 98.1623 aux.loss_ce: 1.2554 aux.acc_seg: 88.5754 +04/19 07:41:39 - mmengine - INFO - Iter(train) [ 550/160000] base_lr: 3.6624e-05 lr: 1.3541e-07 eta: 1 day, 21:32:18 time: 1.0070 data_time: 0.0040 memory: 8462 loss: 1.5712 decode.loss_ce: 0.4877 decode.acc_seg: 98.3290 aux.loss_ce: 1.0835 aux.acc_seg: 93.5476 +04/19 07:42:30 - mmengine - INFO - Iter(train) [ 600/160000] base_lr: 3.9960e-05 lr: 1.4774e-07 eta: 1 day, 21:26:52 time: 1.0063 data_time: 0.0044 memory: 8462 loss: 1.2105 decode.loss_ce: 0.3179 decode.acc_seg: 96.7295 aux.loss_ce: 0.8926 aux.acc_seg: 95.1063 +04/19 07:43:20 - mmengine - INFO - Iter(train) [ 650/160000] base_lr: 4.3296e-05 lr: 1.6007e-07 eta: 1 day, 21:21:34 time: 1.0028 data_time: 0.0043 memory: 8462 loss: 0.9846 decode.loss_ce: 0.2559 decode.acc_seg: 96.0199 aux.loss_ce: 0.7287 aux.acc_seg: 95.8773 +04/19 07:44:10 - mmengine - INFO - Iter(train) [ 700/160000] base_lr: 4.6631e-05 lr: 1.7240e-07 eta: 1 day, 21:16:11 time: 0.9998 data_time: 0.0043 memory: 8462 loss: 0.7521 decode.loss_ce: 0.1831 decode.acc_seg: 97.2086 aux.loss_ce: 0.5690 aux.acc_seg: 96.8592 +04/19 07:45:00 - mmengine - INFO - Iter(train) [ 750/160000] base_lr: 4.9967e-05 lr: 1.8474e-07 eta: 1 day, 21:11:06 time: 1.0000 data_time: 0.0042 memory: 8462 loss: 0.6118 decode.loss_ce: 0.1588 decode.acc_seg: 97.6543 aux.loss_ce: 0.4529 aux.acc_seg: 97.6311 +04/19 07:45:50 - mmengine - INFO - Iter(train) [ 800/160000] base_lr: 5.3302e-05 lr: 1.9707e-07 eta: 1 day, 21:06:10 time: 0.9954 data_time: 0.0041 memory: 8462 loss: 0.4727 decode.loss_ce: 0.1355 decode.acc_seg: 96.8500 aux.loss_ce: 0.3372 aux.acc_seg: 96.2122 +04/19 07:46:39 - mmengine - INFO - Iter(train) [ 850/160000] base_lr: 5.6638e-05 lr: 2.0940e-07 eta: 1 day, 21:01:45 time: 0.9967 data_time: 0.0042 memory: 8462 loss: 0.4170 decode.loss_ce: 0.1385 decode.acc_seg: 98.6187 aux.loss_ce: 0.2784 aux.acc_seg: 97.8769 +04/19 07:47:29 - mmengine - INFO - Iter(train) [ 900/160000] base_lr: 5.9973e-05 lr: 2.2173e-07 eta: 1 day, 20:57:43 time: 0.9982 data_time: 0.0044 memory: 8462 loss: 0.3119 decode.loss_ce: 0.1100 decode.acc_seg: 96.6267 aux.loss_ce: 0.2019 aux.acc_seg: 96.7194 +04/19 07:48:19 - mmengine - INFO - Iter(train) [ 950/160000] base_lr: 6.3309e-05 lr: 2.3407e-07 eta: 1 day, 20:54:01 time: 0.9974 data_time: 0.0042 memory: 8462 loss: 0.2706 decode.loss_ce: 0.1000 decode.acc_seg: 97.2006 aux.loss_ce: 0.1706 aux.acc_seg: 96.2858 +04/19 07:49:09 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_ade20k-512x512_20240419_073206 +04/19 07:49:09 - mmengine - INFO - Iter(train) [ 1000/160000] base_lr: 6.6644e-05 lr: 2.4640e-07 eta: 1 day, 20:50:37 time: 0.9962 data_time: 0.0046 memory: 8462 loss: 0.2227 decode.loss_ce: 0.0922 decode.acc_seg: 98.8609 aux.loss_ce: 0.1306 aux.acc_seg: 97.1130 +04/19 07:49:59 - mmengine - INFO - Iter(train) [ 1050/160000] base_lr: 6.9980e-05 lr: 2.5873e-07 eta: 1 day, 20:47:26 time: 0.9968 data_time: 0.0046 memory: 8462 loss: 0.2316 decode.loss_ce: 0.0922 decode.acc_seg: 97.4508 aux.loss_ce: 0.1394 aux.acc_seg: 96.8807 +04/19 07:50:49 - mmengine - INFO - Iter(train) [ 1100/160000] base_lr: 7.3316e-05 lr: 2.7106e-07 eta: 1 day, 20:44:28 time: 0.9962 data_time: 0.0041 memory: 8462 loss: 0.1786 decode.loss_ce: 0.0795 decode.acc_seg: 97.5788 aux.loss_ce: 0.0991 aux.acc_seg: 97.0448 +04/19 07:51:39 - mmengine - INFO - Iter(train) [ 1150/160000] base_lr: 7.6651e-05 lr: 2.8339e-07 eta: 1 day, 20:41:37 time: 0.9960 data_time: 0.0041 memory: 8462 loss: 0.1799 decode.loss_ce: 0.0788 decode.acc_seg: 96.7302 aux.loss_ce: 0.1011 aux.acc_seg: 95.9280 +04/19 07:52:28 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_ade20k-512x512_20240419_073206 +04/19 07:52:28 - mmengine - INFO - Iter(train) [ 1200/160000] base_lr: 7.9987e-05 lr: 2.9573e-07 eta: 1 day, 20:39:01 time: 0.9964 data_time: 0.0042 memory: 8462 loss: 0.1665 decode.loss_ce: 0.0772 decode.acc_seg: 98.2353 aux.loss_ce: 0.0893 aux.acc_seg: 97.0778 +04/19 07:53:18 - mmengine - INFO - Iter(train) [ 1250/160000] base_lr: 8.3322e-05 lr: 3.0806e-07 eta: 1 day, 20:36:34 time: 0.9961 data_time: 0.0044 memory: 8462 loss: 0.1596 decode.loss_ce: 0.0793 decode.acc_seg: 98.0207 aux.loss_ce: 0.0804 aux.acc_seg: 96.4180 +04/19 07:54:08 - mmengine - INFO - Iter(train) [ 1300/160000] base_lr: 8.6658e-05 lr: 3.2039e-07 eta: 1 day, 20:34:14 time: 0.9980 data_time: 0.0041 memory: 8462 loss: 0.1510 decode.loss_ce: 0.0773 decode.acc_seg: 96.0957 aux.loss_ce: 0.0737 aux.acc_seg: 95.7628 +04/19 07:54:58 - mmengine - INFO - Iter(train) [ 1350/160000] base_lr: 8.9993e-05 lr: 3.3272e-07 eta: 1 day, 20:32:00 time: 0.9967 data_time: 0.0043 memory: 8462 loss: 0.1365 decode.loss_ce: 0.0683 decode.acc_seg: 97.9652 aux.loss_ce: 0.0682 aux.acc_seg: 97.5304 +04/19 07:55:48 - mmengine - INFO - Iter(train) [ 1400/160000] base_lr: 9.3329e-05 lr: 3.4506e-07 eta: 1 day, 20:29:56 time: 0.9980 data_time: 0.0042 memory: 8462 loss: 0.1231 decode.loss_ce: 0.0644 decode.acc_seg: 97.4878 aux.loss_ce: 0.0587 aux.acc_seg: 95.0882 +04/19 07:56:38 - mmengine - INFO - Iter(train) [ 1450/160000] base_lr: 9.6664e-05 lr: 3.5739e-07 eta: 1 day, 20:27:59 time: 0.9969 data_time: 0.0040 memory: 8462 loss: 0.1321 decode.loss_ce: 0.0677 decode.acc_seg: 96.7474 aux.loss_ce: 0.0644 aux.acc_seg: 95.5687 +04/19 07:57:28 - mmengine - INFO - Iter(train) [ 1500/160000] base_lr: 1.0000e-04 lr: 3.6972e-07 eta: 1 day, 20:26:09 time: 0.9976 data_time: 0.0041 memory: 8462 loss: 0.1305 decode.loss_ce: 0.0714 decode.acc_seg: 97.9761 aux.loss_ce: 0.0591 aux.acc_seg: 96.3438 +04/19 07:58:18 - mmengine - INFO - Iter(train) [ 1550/160000] base_lr: 9.9969e-05 lr: 3.6961e-07 eta: 1 day, 20:24:27 time: 0.9994 data_time: 0.0040 memory: 8462 loss: 0.1054 decode.loss_ce: 0.0567 decode.acc_seg: 97.9790 aux.loss_ce: 0.0487 aux.acc_seg: 96.3348 +04/19 07:59:08 - mmengine - INFO - Iter(train) [ 1600/160000] base_lr: 9.9938e-05 lr: 3.6949e-07 eta: 1 day, 20:22:48 time: 0.9983 data_time: 0.0041 memory: 8462 loss: 0.1074 decode.loss_ce: 0.0558 decode.acc_seg: 97.5245 aux.loss_ce: 0.0516 aux.acc_seg: 97.0989 +04/19 07:59:57 - mmengine - INFO - Iter(train) [ 1650/160000] base_lr: 9.9906e-05 lr: 3.6937e-07 eta: 1 day, 20:21:15 time: 0.9995 data_time: 0.0044 memory: 8462 loss: 0.1062 decode.loss_ce: 0.0588 decode.acc_seg: 98.6660 aux.loss_ce: 0.0473 aux.acc_seg: 97.1249 +04/19 08:00:47 - mmengine - INFO - Iter(train) [ 1700/160000] base_lr: 9.9874e-05 lr: 3.6926e-07 eta: 1 day, 20:19:43 time: 0.9997 data_time: 0.0045 memory: 8462 loss: 0.1060 decode.loss_ce: 0.0581 decode.acc_seg: 98.2771 aux.loss_ce: 0.0478 aux.acc_seg: 97.3316 +04/19 08:01:37 - mmengine - INFO - Iter(train) [ 1750/160000] base_lr: 9.9843e-05 lr: 3.6914e-07 eta: 1 day, 20:18:12 time: 0.9998 data_time: 0.0044 memory: 8462 loss: 0.1101 decode.loss_ce: 0.0612 decode.acc_seg: 98.4524 aux.loss_ce: 0.0489 aux.acc_seg: 97.2885 +04/19 08:02:27 - mmengine - INFO - Iter(train) [ 1800/160000] base_lr: 9.9811e-05 lr: 3.6902e-07 eta: 1 day, 20:16:44 time: 0.9986 data_time: 0.0043 memory: 8462 loss: 0.1074 decode.loss_ce: 0.0601 decode.acc_seg: 98.3191 aux.loss_ce: 0.0473 aux.acc_seg: 97.3679 +04/19 08:03:17 - mmengine - INFO - Iter(train) [ 1850/160000] base_lr: 9.9780e-05 lr: 3.6891e-07 eta: 1 day, 20:15:20 time: 1.0000 data_time: 0.0046 memory: 8462 loss: 0.1145 decode.loss_ce: 0.0646 decode.acc_seg: 96.8958 aux.loss_ce: 0.0499 aux.acc_seg: 95.6266 +04/19 08:04:07 - mmengine - INFO - Iter(train) [ 1900/160000] base_lr: 9.9748e-05 lr: 3.6879e-07 eta: 1 day, 20:13:58 time: 0.9999 data_time: 0.0045 memory: 8462 loss: 0.0987 decode.loss_ce: 0.0551 decode.acc_seg: 98.0410 aux.loss_ce: 0.0437 aux.acc_seg: 97.2126 +04/19 08:04:57 - mmengine - INFO - Iter(train) [ 1950/160000] base_lr: 9.9717e-05 lr: 3.6867e-07 eta: 1 day, 20:12:40 time: 1.0001 data_time: 0.0042 memory: 8462 loss: 0.1058 decode.loss_ce: 0.0595 decode.acc_seg: 98.5418 aux.loss_ce: 0.0463 aux.acc_seg: 96.3310 +04/19 08:05:47 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_ade20k-512x512_20240419_073206 +04/19 08:05:47 - mmengine - INFO - Iter(train) [ 2000/160000] base_lr: 9.9685e-05 lr: 3.6856e-07 eta: 1 day, 20:11:20 time: 1.0007 data_time: 0.0041 memory: 8462 loss: 0.1130 decode.loss_ce: 0.0679 decode.acc_seg: 96.9311 aux.loss_ce: 0.0452 aux.acc_seg: 95.4798 +04/19 08:06:37 - mmengine - INFO - Iter(train) [ 2050/160000] base_lr: 9.9654e-05 lr: 3.6844e-07 eta: 1 day, 20:10:02 time: 0.9987 data_time: 0.0041 memory: 8462 loss: 0.0946 decode.loss_ce: 0.0536 decode.acc_seg: 98.1184 aux.loss_ce: 0.0410 aux.acc_seg: 97.6820 +04/19 08:07:27 - mmengine - INFO - Iter(train) [ 2100/160000] base_lr: 9.9622e-05 lr: 3.6832e-07 eta: 1 day, 20:08:43 time: 0.9996 data_time: 0.0043 memory: 8462 loss: 0.0973 decode.loss_ce: 0.0559 decode.acc_seg: 98.9077 aux.loss_ce: 0.0414 aux.acc_seg: 97.5845 +04/19 08:08:17 - mmengine - INFO - Iter(train) [ 2150/160000] base_lr: 9.9591e-05 lr: 3.6821e-07 eta: 1 day, 20:07:30 time: 1.0001 data_time: 0.0046 memory: 8462 loss: 0.0883 decode.loss_ce: 0.0493 decode.acc_seg: 98.4428 aux.loss_ce: 0.0390 aux.acc_seg: 97.5756 +04/19 08:09:07 - mmengine - INFO - Iter(train) [ 2200/160000] base_lr: 9.9559e-05 lr: 3.6809e-07 eta: 1 day, 20:06:16 time: 1.0000 data_time: 0.0041 memory: 8462 loss: 0.0858 decode.loss_ce: 0.0488 decode.acc_seg: 98.9340 aux.loss_ce: 0.0370 aux.acc_seg: 98.4314 +04/19 08:09:57 - mmengine - INFO - Iter(train) [ 2250/160000] base_lr: 9.9527e-05 lr: 3.6797e-07 eta: 1 day, 20:05:02 time: 0.9992 data_time: 0.0041 memory: 8462 loss: 0.0857 decode.loss_ce: 0.0492 decode.acc_seg: 97.9370 aux.loss_ce: 0.0365 aux.acc_seg: 96.0619 +04/19 08:10:47 - mmengine - INFO - Iter(train) [ 2300/160000] base_lr: 9.9496e-05 lr: 3.6786e-07 eta: 1 day, 20:03:48 time: 0.9986 data_time: 0.0042 memory: 8462 loss: 0.0937 decode.loss_ce: 0.0543 decode.acc_seg: 99.1913 aux.loss_ce: 0.0394 aux.acc_seg: 97.6168 +04/19 08:11:37 - mmengine - INFO - Iter(train) [ 2350/160000] base_lr: 9.9464e-05 lr: 3.6774e-07 eta: 1 day, 20:02:34 time: 0.9974 data_time: 0.0042 memory: 8462 loss: 0.0839 decode.loss_ce: 0.0480 decode.acc_seg: 96.4069 aux.loss_ce: 0.0359 aux.acc_seg: 95.8021 +04/19 08:12:27 - mmengine - INFO - Iter(train) [ 2400/160000] base_lr: 9.9433e-05 lr: 3.6762e-07 eta: 1 day, 20:01:23 time: 0.9996 data_time: 0.0042 memory: 8462 loss: 0.0740 decode.loss_ce: 0.0406 decode.acc_seg: 98.9935 aux.loss_ce: 0.0333 aux.acc_seg: 97.8920 +04/19 08:13:17 - mmengine - INFO - Iter(train) [ 2450/160000] base_lr: 9.9401e-05 lr: 3.6751e-07 eta: 1 day, 20:00:12 time: 0.9989 data_time: 0.0042 memory: 8462 loss: 0.0819 decode.loss_ce: 0.0465 decode.acc_seg: 97.6427 aux.loss_ce: 0.0354 aux.acc_seg: 95.5507 +04/19 08:14:07 - mmengine - INFO - Iter(train) [ 2500/160000] base_lr: 9.9370e-05 lr: 3.6739e-07 eta: 1 day, 19:59:02 time: 0.9996 data_time: 0.0041 memory: 8462 loss: 0.0806 decode.loss_ce: 0.0457 decode.acc_seg: 98.3541 aux.loss_ce: 0.0350 aux.acc_seg: 97.8451 +04/19 08:14:57 - mmengine - INFO - Iter(train) [ 2550/160000] base_lr: 9.9338e-05 lr: 3.6727e-07 eta: 1 day, 19:57:55 time: 1.0000 data_time: 0.0046 memory: 8462 loss: 0.0901 decode.loss_ce: 0.0521 decode.acc_seg: 96.8052 aux.loss_ce: 0.0380 aux.acc_seg: 95.4477 +04/19 08:15:47 - mmengine - INFO - Iter(train) [ 2600/160000] base_lr: 9.9307e-05 lr: 3.6716e-07 eta: 1 day, 19:56:46 time: 0.9992 data_time: 0.0044 memory: 8462 loss: 0.0891 decode.loss_ce: 0.0503 decode.acc_seg: 98.3551 aux.loss_ce: 0.0388 aux.acc_seg: 96.2929 +04/19 08:16:37 - mmengine - INFO - Iter(train) [ 2650/160000] base_lr: 9.9275e-05 lr: 3.6704e-07 eta: 1 day, 19:55:38 time: 0.9993 data_time: 0.0042 memory: 8462 loss: 0.0795 decode.loss_ce: 0.0454 decode.acc_seg: 97.8434 aux.loss_ce: 0.0341 aux.acc_seg: 97.0446 +04/19 08:17:27 - mmengine - INFO - Iter(train) [ 2700/160000] base_lr: 9.9244e-05 lr: 3.6692e-07 eta: 1 day, 19:54:33 time: 0.9995 data_time: 0.0044 memory: 8462 loss: 0.0840 decode.loss_ce: 0.0501 decode.acc_seg: 98.5741 aux.loss_ce: 0.0339 aux.acc_seg: 96.9194 +04/19 08:18:17 - mmengine - INFO - Iter(train) [ 2750/160000] base_lr: 9.9212e-05 lr: 3.6681e-07 eta: 1 day, 19:53:27 time: 0.9998 data_time: 0.0043 memory: 8462 loss: 0.0900 decode.loss_ce: 0.0518 decode.acc_seg: 98.2058 aux.loss_ce: 0.0381 aux.acc_seg: 96.4069 +04/19 08:19:07 - mmengine - INFO - Iter(train) [ 2800/160000] base_lr: 9.9180e-05 lr: 3.6669e-07 eta: 1 day, 19:52:24 time: 1.0003 data_time: 0.0045 memory: 8462 loss: 0.0752 decode.loss_ce: 0.0434 decode.acc_seg: 98.1392 aux.loss_ce: 0.0318 aux.acc_seg: 97.6873 +04/19 08:19:57 - mmengine - INFO - Iter(train) [ 2850/160000] base_lr: 9.9149e-05 lr: 3.6657e-07 eta: 1 day, 19:51:18 time: 0.9991 data_time: 0.0041 memory: 8462 loss: 0.0800 decode.loss_ce: 0.0476 decode.acc_seg: 98.0701 aux.loss_ce: 0.0324 aux.acc_seg: 95.6417 +04/19 08:20:47 - mmengine - INFO - Iter(train) [ 2900/160000] base_lr: 9.9117e-05 lr: 3.6646e-07 eta: 1 day, 19:50:15 time: 1.0000 data_time: 0.0045 memory: 8462 loss: 0.0739 decode.loss_ce: 0.0421 decode.acc_seg: 98.0650 aux.loss_ce: 0.0317 aux.acc_seg: 97.2357 +04/19 08:21:37 - mmengine - INFO - Iter(train) [ 2950/160000] base_lr: 9.9086e-05 lr: 3.6634e-07 eta: 1 day, 19:49:14 time: 1.0000 data_time: 0.0042 memory: 8462 loss: 0.0739 decode.loss_ce: 0.0434 decode.acc_seg: 98.9803 aux.loss_ce: 0.0306 aux.acc_seg: 97.1418 +04/19 08:22:27 - mmengine - INFO - Exp name: mae-base_upernet_8xb2-amp-160k_ade20k-512x512_20240419_073206 +04/19 08:22:27 - mmengine - INFO - Iter(train) [ 3000/160000] base_lr: 9.9054e-05 lr: 3.6622e-07 eta: 1 day, 19:48:15 time: 1.0009 data_time: 0.0045 memory: 8462 loss: 0.0741 decode.loss_ce: 0.0436 decode.acc_seg: 99.0244 aux.loss_ce: 0.0305 aux.acc_seg: 98.1958