ValueError: (InvalidArgument) The type of data we are trying to retrieve (float64) does not match the type of data (float32) currently contained in the container. [Hint: Expected dtype() == phi::CppTypeToDataType<T>::Type(), but received dtype():10 != phi::CppTypeToDataType<T>::Type():11.] (at ..\paddle\phi\core\dense_tensor.cc:171)

[qijix]

(C:\Users\qijix\D\segger) PS D:\PaddleSeg_2.5> python train.py --config configs/base/F_shutong.yml --do_eval --use_vdl --save_interval 500 --save_dir output
2024-07-23 00:57:08 [INFO]
------------Environment Information-------------
platform: Windows-10-10.0.22631-SP0
Python: 3.8.19 | packaged by conda-forge | (default, Mar 20 2024, 12:38:07) [MSC v.1929 64 bit (AMD64)]
Paddle compiled with cuda: True
NVCC: Build cuda_11.8.r11.8/compiler.31833905_0
cudnn: 8.8
GPUs used: 1
CUDA_VISIBLE_DEVICES: None
GPU: ['GPU 0: NVIDIA GeForce']
PaddleSeg: 2.5.0
PaddlePaddle: 2.6.1
OpenCV: 4.10.0

2024-07-23 00:57:09 [INFO]
---------------Config Information---------------
batch_size: 1
iters: 10000
loss:
coef:

• 1
  types:
• ignore_index: 255
  type: CrossEntropyLoss
  lr_scheduler:
  end_lr: 0
  learning_rate: 6.0e-05
  power: 1
  type: PolynomialDecay
  model:
  backbone:
  pretrained: https://bj.bcebos.com/paddleseg/dygraph/backbone/mix_vision_transformer_b3.tar.gz
  type: MixVisionTransformer_B3
  embedding_dim: 768
  num_classes: 2
  type: SegFormer
  optimizer:
  beta1: 0.9
  beta2: 0.999
  type: AdamW
  weight_decay: 0.01
  train_dataset:
  dataset_root: D:\new_data
  mode: train
  num_classes: 2
  train_path: D:\new_data\train.txt
  transforms:
• type: RandomHorizontalFlip
• type: Normalize
  type: Dataset
  val_dataset:
  dataset_root: D:\new_data
  mode: val
  num_classes: 2
  transforms:
• type: Normalize
  type: Dataset
  val_path: D:\new_data\val.txt

---

W0723 00:57:09.035965 13308 gpu_resources.cc:119] Please NOTE: device: 0, GPU Compute Capability: 8.9, Driver API Version: 12.4, Runtime API Version: 11.8
W0723 00:57:09.036965 13308 gpu_resources.cc:164] device: 0, cuDNN Version: 8.8.
2024-07-23 00:57:09 [INFO] Loading pretrained model from https://bj.bcebos.com/paddleseg/dygraph/backbone/mix_vision_transformer_b3.tar.gz
2024-07-23 00:57:09 [INFO] There are 572/572 variables loaded into MixVisionTransformer.
W0723 00:57:10.168778 13308 dygraph_functions.cc:52647] got different data type, run type protmotion automatically, this may cause data type been changed.
Traceback (most recent call last):
File "train.py", line 230, in
main(args)
File "train.py", line 206, in main
train(
File "D:\PaddleSeg_2.5\paddleseg\core\train.py", line 204, in train
logits_list = ddp_model(images) if nranks > 1 else model(images)
File "C:\Users\qijix\D\segger\lib\site-packages\paddle\nn\layer\layers.py", line 1429, in call
return self.forward(*inputs, **kwargs)
File "D:\PaddleSeg_2.5\paddleseg\models\segformer.py", line 83, in forward
feats = self.backbone(x)
File "C:\Users\qijix\D\segger\lib\site-packages\paddle\nn\layer\layers.py", line 1429, in call
return self.forward(*inputs, **kwargs)
File "D:\PaddleSeg_2.5\paddleseg\models\backbones\mix_transformer.py", line 472, in forward
x = self.forward_features(x)
File "D:\PaddleSeg_2.5\paddleseg\models\backbones\mix_transformer.py", line 439, in forward_features
x = blk(x, H, W)
File "C:\Users\qijix\D\segger\lib\site-packages\paddle\nn\layer\layers.py", line 1429, in call
return self.forward(*inputs, **kwargs)
File "D:\PaddleSeg_2.5\paddleseg\models\backbones\mix_transformer.py", line 199, in forward
x = x + self.drop_path(self.attn(self.norm1(x), H, W))
File "C:\Users\qijix\D\segger\lib\site-packages\paddle\nn\layer\layers.py", line 1429, in call
return self.forward(*inputs, **kwargs)
File "D:\PaddleSeg_2.5\paddleseg\models\backbones\transformer_utils.py", line 53, in forward
return drop_path(x, self.drop_prob, self.training)
File "D:\PaddleSeg_2.5\paddleseg\models\backbones\transformer_utils.py", line 40, in drop_path
output = x.divide(keep_prob) * random_tensor
File "C:\Users\qijix\D\segger\lib\site-packages\paddle\tensor\math.py", line 895, in divide
return _C_ops.divide(x, y)
ValueError: (InvalidArgument) The type of data we are trying to retrieve (float64) does not match the type of data (float32) currently contained in the container.
[Hint: Expected dtype() == phi::CppTypeToDataType::Type(), but received dtype():10 != phi::CppTypeToDataType::Type():11.] (at ..\paddle\phi\core\dense_tensor.cc:171)
一直出现数据类型不对。我使用的是paddlepaddle-gpu==2.6.1，cuda11.8，paddleseg2.5，数据类型为8uint的tif文件，但转换为float32的tif格式图像又失真了，请问有什么解决办法吗？