PaddlePaddle / PaddleGAN

PaddlePaddle GAN library, including lots of interesting applications like First-Order motion transfer, Wav2Lip, picture repair, image editing, photo2cartoon, image style transfer, GPEN, and so on.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

CUDNN error(9), CUDNN_STATUS_NOT_SUPPORTED.

monkeycc opened this issue · comments

bug描述 Describe the Bug

官网写着
CUDA 工具包 11.7 配合 cuDNN v8.4.1,

https://github.com/PaddlePaddle/PaddleGAN/blob/develop/docs/zh_CN/tutorials/nafnet.md

测试报错

paddle2.51 和 2.4
cuDNN v8.4.0
cuDNN v8.6
都一样报错

[09/26 14:26:03] ppgan INFO: Found C:\Users\monke\.cache\ppgan\NAFNet_Denoising.pdparams
W0926 14:26:06.026473 14536 gpu_resources.cc:61] Please NOTE: device: 0, GPU Compute Capability: 8.6, Driver API Version: 12.2, Runtime API Version: 11.7
W0926 14:26:06.052229 14536 gpu_resources.cc:91] device: 0, cuDNN Version: 8.4.
  0%|                                                                                                                                                                           | 0/1 [00:03<?, ?it/s]
Traceback (most recent call last):
  File "applications/tools/nafnet_denoising.py", line 59, in <module>
    predictor.run(images_path=args.images_path)
  File "E:\2023_ShuJu_WEB\PaddleGAN-develop\ppgan\apps\nafnet_predictor.py", line 142, in run
    output = self.generator(img_L)
  File "E:\anaconda3\envs\PaddleGan\lib\site-packages\paddle\fluid\dygraph\layers.py", line 1012, in __call__
    return self.forward(*inputs, **kwargs)
  File "E:\2023_ShuJu_WEB\PaddleGAN-develop\ppgan\models\generators\nafnet.py", line 360, in forward
    x = self.intro(inp)
  File "E:\anaconda3\envs\PaddleGan\lib\site-packages\paddle\fluid\dygraph\layers.py", line 1012, in __call__
    return self.forward(*inputs, **kwargs)
  File "E:\anaconda3\envs\PaddleGan\lib\site-packages\paddle\nn\layer\conv.py", line 712, in forward
    out = F.conv._conv_nd(
  File "E:\anaconda3\envs\PaddleGan\lib\site-packages\paddle\nn\functional\conv.py", line 140, in _conv_nd
    pre_bias = _C_ops.conv2d(
OSError: (External) CUDNN error(9), CUDNN_STATUS_NOT_SUPPORTED.
  [Hint: Please search for the error code(9) on website (https://docs.nvidia.com/deeplearning/cudnn/api/index.html#cudnnStatus_t) to get Nvidia's official solution and advice about CUDNN Error.] (at ../paddle/fluid/platform/device/gpu/cuda/cudnn_desc.h:177)

其他补充信息 Additional Supplementary Information

No response