4_cifar10_tutorial.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 124
    },
    "colab_type": "code",
    "id": "cpjthnrHynVb",
    "outputId": "33652392-2017-4475-ba9f-908896297851"
   },
   "outputs": [],
   "source": [
    "from google.colab import drive\n",
    "drive.mount('/content/gdrive')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "p6-jjb94ynVn"
   },
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import torch\n",
    "import torchvision\n",
    "import torchvision.transforms as transforms\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "import torch.optim as optim\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "PWq_Lnm7ynVu"
   },
   "outputs": [],
   "source": [
    "# Hyperparameters\n",
    "batch_size = 64\n",
    "# Learning rate\n",
    "lr = 0.001\n",
    "# Number of training epochs\n",
    "num_epochs = 20"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "k3IGFjb4ynV1"
   },
   "source": [
    "Training an image classifier\n",
    "----------------------------\n",
    "\n",
    "We will do the following steps in order:\n",
    "\n",
    "1. Load and normalizing the CIFAR10 training and test datasets using\n",
    "   ``torchvision``\n",
    "2. Define a Convolutional Neural Network\n",
    "3. Define a loss function\n",
    "4. Train the network on the training data\n",
    "5. Test the network on the test data\n",
    "\n",
    "1. Loading and normalizing CIFAR10\n",
    "^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n",
    "\n",
    "Using ``torchvision``, it’s extremely easy to load CIFAR10."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "gHcq5xZMynV3"
   },
   "outputs": [],
   "source": [
    "# Device configuration\n",
    "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n",
    "\n",
    "# Path to dataset\n",
    "# dataset_root_path = '/scratch2/datasets/cifar10/'\n",
    "dataset_root_path = '/content/gdrive/My Drive/cifar10/'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Zw2KgMPhynV7"
   },
   "source": [
    "The output of torchvision datasets are PILImage images of range [0, 1].\n",
    "We transform them to Tensors of normalized range [-1, 1].\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 52
    },
    "colab_type": "code",
    "id": "MxHt55XiynV9",
    "outputId": "cf8b8c47-25dc-4d76-de6e-17092f075a21"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz to /content/gdrive/My Drive/FashionMNIST/cifar-10-python.tar.gz\n",
      "Files already downloaded and verified\n"
     ]
    }
   ],
   "source": [
    "transform = transforms.Compose(\n",
    "    [transforms.ToTensor(),\n",
    "     transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])\n",
    "\n",
    "trainset = torchvision.datasets.CIFAR10(root=dataset_root_path, train=True, download=True, transform=transform)\n",
    "trainloader = torch.utils.data.DataLoader(trainset, batch_size=batch_size, shuffle=True, num_workers=2)\n",
    "\n",
    "testset = torchvision.datasets.CIFAR10(root=dataset_root_path, train=False, download=True, transform=transform)\n",
    "testloader = torch.utils.data.DataLoader(testset, batch_size=4, shuffle=False, num_workers=2)\n",
    "\n",
    "classes = ('plane', 'car', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "goWkZoqfynWE"
   },
   "source": [
    "Let us show some of the training images\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 155
    },
    "colab_type": "code",
    "id": "lMrdfL4lynWG",
    "outputId": "495fff1b-8771-4d9b-b5eb-ac912a3e4699"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAW8AAAB5CAYAAAD2xKAmAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nM4j/A6L8H5sI8HyMAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " ship  bird  ship   cat\n"
     ]
    }
   ],
   "source": [
    "# functions to show an image\n",
    "def imshow(img):\n",
    "    img = img / 2 + 0.5     # unnormalize\n",
    "    npimg = img.numpy()\n",
    "    plt.imshow(np.transpose(npimg, (1, 2, 0)))\n",
    "    plt.axis('off')\n",
    "    plt.show()\n",
    "\n",
    "\n",
    "# get some random training images\n",
    "dataiter = iter(trainloader)\n",
    "images, labels = dataiter.next()\n",
    "\n",
    "# show images\n",
    "imshow(torchvision.utils.make_grid(images[:4]))\n",
    "# print labels\n",
    "print(' '.join('%5s' % classes[labels[j]] for j in range(4)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "uT2kipKlynWP"
   },
   "source": [
    "2. Define a Convolutional Neural Network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "99nevj9dynWS"
   },
   "outputs": [],
   "source": [
    "class Net(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(Net, self).__init__()\n",
    "        self.conv1 = nn.Conv2d(3, 6, 5)\n",
    "        self.pool = nn.MaxPool2d(2, 2)\n",
    "        self.conv2 = nn.Conv2d(6, 16, 5)\n",
    "        self.fc1 = nn.Linear(16 * 5 * 5, 120)\n",
    "        self.fc2 = nn.Linear(120, 84)\n",
    "        self.fc3 = nn.Linear(84, 10)\n",
    "\n",
    "    def forward(self, x):\n",
    "        x = self.pool(F.relu(self.conv1(x)))\n",
    "        x = self.pool(F.relu(self.conv2(x)))\n",
    "        x = x.view(-1, 16 * 5 * 5)\n",
    "        x = F.relu(self.fc1(x))\n",
    "        x = F.relu(self.fc2(x))\n",
    "        x = self.fc3(x)\n",
    "        return x\n",
    "\n",
    "net = Net()\n",
    "net = net.to(device)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "yZVw_l2hynWb"
   },
   "source": [
    "3. Define a Loss function and optimizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "zpjIa0LIynWh"
   },
   "outputs": [],
   "source": [
    "criterion = nn.CrossEntropyLoss()\n",
    "# optimizer = optim.SGD(net.parameters(), lr=lr, momentum=0.9)\n",
    "optimizer = optim.Adam(net.parameters(), lr=lr)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "181Eu5XOynWl"
   },
   "source": [
    "4. Train the network"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "8NyXek_GynWo"
   },
   "outputs": [],
   "source": [
    "# Helper function to test the model\n",
    "# In test phase, we don't need to compute gradients (for memory efficiency)\n",
    "def eval(testloader, epoch):\n",
    "    with torch.no_grad():\n",
    "        running_loss_test = 0.0\n",
    "        test_correct = 0\n",
    "        test_total = 0\n",
    "        test_i = 0\n",
    "        for test_images, test_labels in testloader:\n",
    "            # get the inputs\n",
    "            # test_images, test_labels = test_data\n",
    "            test_images = test_images.to(device)\n",
    "            test_labels = test_labels.to(device)\n",
    "            test_outputs = net(test_images)\n",
    "            test_loss = criterion(test_outputs, test_labels)\n",
    "            _, test_predicted = torch.max(test_outputs.data, 1)\n",
    "            \n",
    "            running_loss_test += test_loss.item()\n",
    "            test_total += test_labels.size(0)\n",
    "            test_correct += (test_predicted == test_labels).sum().item()\n",
    "            test_i += 1\n",
    "\n",
    "            if test_total >= testloader.dataset.test_data.shape[0]-batch_size:\n",
    "                print('[{:d}], test accuracy: {:4.2f} %, test_loss {:.5f}'.format(epoch+1, \\\n",
    "                                                100*test_correct/test_total, running_loss_test/test_i))\n",
    "                break\n",
    "        return 100*test_correct/test_total, running_loss_test/test_i"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "8paZ2VlKynWs"
   },
   "outputs": [],
   "source": [
    "# Helper function to compute (true) training loss\n",
    "# Eval on training data. This is to be run with initial model only\n",
    "# In eval phase, we don't need to compute gradients (for memory efficiency)\n",
    "def eval_initial_model(trainloader):\n",
    "    with torch.no_grad():\n",
    "        running_loss_train = 0.0\n",
    "        train_correct = 0\n",
    "        train_total = 0\n",
    "        train_i = 0\n",
    "        for train_images, train_labels in trainloader:\n",
    "            # get the inputs\n",
    "            # train_images, train_labels = train_data\n",
    "            train_images = train_images.to(device)\n",
    "            train_labels = train_labels.to(device)\n",
    "            train_outputs = net(train_images)\n",
    "            train_loss = criterion(train_outputs, train_labels)\n",
    "            _, train_predicted = torch.max(train_outputs.data, 1)\n",
    "            \n",
    "            running_loss_train += train_loss.item()\n",
    "            train_total += train_labels.size(0)\n",
    "            train_correct += (train_predicted == train_labels).sum().item()\n",
    "            train_i += 1\n",
    "\n",
    "        print('[0], train accuracy: {:4.2f} %, train_loss {:.5f}'.format(100*train_correct/train_total, \\\n",
    "                                                                        running_loss_train/train_i))\n",
    "        return 100*train_correct/train_total, running_loss_train/train_i"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 764
    },
    "colab_type": "code",
    "id": "MY7L8LMPynWx",
    "outputId": "b9b0c845-3095-4c72-bace-a3d0e09b103a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0], train accuracy: 10.00 %, train_loss 2.30092\n",
      "[1], test accuracy: 9.96 %, test_loss 2.30288\n",
      "[1] train accuracy: 40.74 % train loss: 1.61012\n",
      "[1], test accuracy: 49.21 %, test_loss 1.39504\n",
      "[2] train accuracy: 52.25 % train loss: 1.32285\n",
      "[2], test accuracy: 54.20 %, test_loss 1.25957\n",
      "[3] train accuracy: 56.56 % train loss: 1.21228\n",
      "[3], test accuracy: 56.88 %, test_loss 1.20294\n",
      "[4] train accuracy: 59.36 % train loss: 1.14003\n",
      "[4], test accuracy: 60.03 %, test_loss 1.12609\n",
      "[5] train accuracy: 61.70 % train loss: 1.07300\n",
      "[5], test accuracy: 59.70 %, test_loss 1.12261\n",
      "[6] train accuracy: 63.64 % train loss: 1.01982\n",
      "[6], test accuracy: 61.65 %, test_loss 1.08210\n",
      "[7] train accuracy: 65.85 % train loss: 0.96916\n",
      "[7], test accuracy: 62.39 %, test_loss 1.07559\n",
      "[8] train accuracy: 67.10 % train loss: 0.92929\n",
      "[8], test accuracy: 62.80 %, test_loss 1.05416\n",
      "[9] train accuracy: 68.67 % train loss: 0.88925\n",
      "[9], test accuracy: 63.77 %, test_loss 1.03570\n",
      "[10] train accuracy: 69.95 % train loss: 0.85322\n",
      "[10], test accuracy: 62.64 %, test_loss 1.08020\n",
      "[11] train accuracy: 70.93 % train loss: 0.82108\n",
      "[11], test accuracy: 64.30 %, test_loss 1.03824\n",
      "[12] train accuracy: 72.19 % train loss: 0.79056\n",
      "[12], test accuracy: 63.41 %, test_loss 1.05233\n",
      "[13] train accuracy: 73.06 % train loss: 0.76417\n",
      "[13], test accuracy: 63.84 %, test_loss 1.05638\n",
      "[14] train accuracy: 74.07 % train loss: 0.73698\n",
      "[14], test accuracy: 64.39 %, test_loss 1.05694\n",
      "[15] train accuracy: 74.81 % train loss: 0.71573\n",
      "[15], test accuracy: 64.38 %, test_loss 1.06781\n",
      "[16] train accuracy: 75.61 % train loss: 0.68754\n",
      "[16], test accuracy: 64.14 %, test_loss 1.06743\n",
      "[17] train accuracy: 76.24 % train loss: 0.66983\n",
      "[17], test accuracy: 64.74 %, test_loss 1.08165\n",
      "[18] train accuracy: 76.97 % train loss: 0.64773\n",
      "[18], test accuracy: 63.86 %, test_loss 1.13096\n",
      "[19] train accuracy: 77.73 % train loss: 0.62535\n",
      "[19], test accuracy: 64.21 %, test_loss 1.13570\n",
      "[20] train accuracy: 78.28 % train loss: 0.61100\n",
      "[20], test accuracy: 63.90 %, test_loss 1.14781\n",
      "Finished Training\n"
     ]
    }
   ],
   "source": [
    "test_acc = [0.0]*(num_epochs+1)\n",
    "test_loss = [0.0]*(num_epochs+1)\n",
    "train_acc = [0.0]*(num_epochs+1)\n",
    "train_loss = [0.0]*(num_epochs+1)\n",
    "lst_counter = 0\n",
    "\n",
    "train_acc[lst_counter], train_loss[lst_counter] = eval_initial_model(trainloader)\n",
    "test_acc[lst_counter], test_loss[lst_counter] = eval(testloader, 0)\n",
    "lst_counter += 1\n",
    "\n",
    "for epoch in range(num_epochs):  # loop over the dataset multiple times\n",
    "\n",
    "    running_loss_training = 0.0\n",
    "    correct = 0\n",
    "    total = 0\n",
    "    i = 0\n",
    "    # for i, data in enumerate(trainloader, 0):\n",
    "    for inputs, labels in trainloader:\n",
    "        # get the inputs\n",
    "        inputs = inputs.to(device); labels = labels.to(device)\n",
    "\n",
    "        # zero the parameter gradients\n",
    "        optimizer.zero_grad()\n",
    "\n",
    "        # forward + backward + optimize\n",
    "        outputs = net(inputs)\n",
    "        training_loss = criterion(outputs, labels)\n",
    "        _, predicted = torch.max(outputs.data, 1)\n",
    "        \n",
    "        training_loss.backward()\n",
    "        optimizer.step()\n",
    "\n",
    "        # print statistics\n",
    "        running_loss_training += training_loss.item()\n",
    "        total += labels.size(0)\n",
    "        correct += (predicted == labels).sum().item()\n",
    "        i += 1\n",
    "        if total >= trainloader.dataset.train_data.shape[0]:\n",
    "            print('[{:d}] train accuracy: {:4.2f} % train loss: {:.5f}'.format(epoch+1, \\\n",
    "                                                            100*correct/total, running_loss_training/i))\n",
    "            train_acc[lst_counter] = 100*correct/total\n",
    "            train_loss[lst_counter] = running_loss_training/i\n",
    "            \n",
    "            running_loss_training = 0.0\n",
    "            \n",
    "            test_acc[lst_counter], test_loss[lst_counter] = eval(testloader, epoch)\n",
    "            lst_counter += 1\n",
    "\n",
    "print('Finished Training')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 279
    },
    "colab_type": "code",
    "id": "dvSLaE_gynW5",
    "outputId": "86df9706-199f-483b-9a6e-fa04ebf3c0a7"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEGCAYAAACKB4k+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQggRIElBCCFEgCQFIYQQAf8PlRbk2+8D8u8AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting train and test loss vs Epochs\n",
    "fig1 = plt.figure(1)        \n",
    "plt.plot(range(num_epochs+1),train_loss,'r-',label='train_loss')        \n",
    "plt.plot(range(num_epochs+1),test_loss,'g-',label='test_loss')\n",
    "plt.legend(loc='upper right')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Losses')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 279
    },
    "colab_type": "code",
    "id": "MYxv-o7tynW_",
    "outputId": "5fd77623-29a3-47ed-dc41-1eabd92ede82"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAEGCAYAAABsLkJ6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig2 = plt.figure(1)        \n",
    "plt.plot(range(1, num_epochs+1),train_loss[1:],'r-',label='train_loss')        \n",
    "plt.plot(range(1, num_epochs+1),test_loss[1:],'g-',label='test_loss')\n",
    "plt.legend(loc='upper right')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Losses')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 279
    },
    "colab_type": "code",
    "id": "7WyfqqzUynXF",
    "outputId": "e1d09845-11fb-49cf-c062-4bc5d94a674a"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAEGCAYAAACJnEVTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nECKCSNIXQogIIklfCCEiiCR9IYSIIP8fc8RrYjBl4JkAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plotting train and test accuracy vs Epochs\n",
    "fig3 = plt.figure(1)        \n",
    "plt.plot(range(num_epochs+1),train_acc,'r-',label='train_acc')        \n",
    "plt.plot(range(num_epochs+1),test_acc,'g-',label='test_acc')\n",
    "plt.legend(loc='lower right')\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Accuracies')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "7arCcH20ynXP"
   },
   "source": [
    "5. Test the network on the test data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 173
    },
    "colab_type": "code",
    "id": "Bg_BiZa5ynXQ",
    "outputId": "0d577472-b7cd-4c5c-a759-c9f3fdbe3ae4"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAW8AAAB5CAYAAAD2xKAmAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAU4QUWAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GroundTruth:    cat  ship  ship plane\n",
      "Predicted:    dog (0.50) plane (0.85)  ship (0.67) plane (0.65)\n"
     ]
    }
   ],
   "source": [
    "dataiter = iter(testloader)\n",
    "images, labels = dataiter.next()\n",
    "\n",
    "# print images\n",
    "imshow(torchvision.utils.make_grid(images[:4], nrow=4))\n",
    "print('GroundTruth: ', ' '.join('%5s' % classes[labels[j]] for j in range(4)))\n",
    "\n",
    "# Get the predicted outputs\n",
    "images = images.to(device)\n",
    "labels = labels.to(device)\n",
    "outputs = net(images)\n",
    "outputs_softmax = F.softmax(outputs, dim=1)\n",
    "probabilities, predicted = torch.max(outputs_softmax, 1)\n",
    "probabilities_np = probabilities.cpu().data.numpy()\n",
    "print('Predicted: ', ' '.join('%5s (%.2f)' % (classes[predicted[j]], probabilities[j]) for j in range(4)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "kisIJYWpynXc"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "4_cifar10_tutorial.ipynb",
   "provenance": [],
   "version": "0.3.2"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}