search

va1shn9v / PromptIR

PromptIR: Prompting for All-in-One Blind Image Restoration [NeurIPS 2023]
https://arxiv.org/abs/2306.13090
Other
297 stars 22 forks source link

Demo Code Execution Error #33

Closed munib94 closed 2 months ago

munib94 commented 2 months ago

Hi,

When I try to run the following command:

python demo.py --test_path './test/artifacts_dataset/rain/IMG_1665.JPG' --output_path './output/demo/'

I get the following error:

Total Images : ['../artifacts_dataset/rain/IMG_1665.JPG']
Start testing...
  0%|                                                                                                                                                                                                                                                                                | 0/1 [00:02<?, ?it/s]
Traceback (most recent call last):
  File "demo.py", line 120, in <module>
    restored = net(degrad_patch)
  File "/home/user/.pyenv/versions/miniconda3-3.8-23.9.0-0/envs/promptir/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
  File "demo.py", line 56, in forward
    return self.net(x)
  File "/home/user/.pyenv/versions/miniconda3-3.8-23.9.0-0/envs/promptir/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
  File "/home/user/MSU/Research/PromptIR/net/model.py", line 326, in forward
    out_enc_level1 = self.encoder_level1(inp_enc_level1)
  File "/home/user/.pyenv/versions/miniconda3-3.8-23.9.0-0/envs/promptir/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
  File "/home/user/.pyenv/versions/miniconda3-3.8-23.9.0-0/envs/promptir/lib/python3.8/site-packages/torch/nn/modules/container.py", line 217, in forward
    input = module(input)
  File "/home/user/.pyenv/versions/miniconda3-3.8-23.9.0-0/envs/promptir/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
  File "/home/user/MSU/Research/PromptIR/net/model.py", line 194, in forward
    x = x + self.ffn(self.norm2(x))
  File "/home/user/.pyenv/versions/miniconda3-3.8-23.9.0-0/envs/promptir/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
  File "/home/user/MSU/Research/PromptIR/net/model.py", line 96, in forward
    x1, x2 = self.dwconv(x).chunk(2, dim=1)
  File "/home/user/.pyenv/versions/miniconda3-3.8-23.9.0-0/envs/promptir/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
  File "/home/user/.pyenv/versions/miniconda3-3.8-23.9.0-0/envs/promptir/lib/python3.8/site-packages/torch/nn/modules/conv.py", line 463, in forward
    return self._conv_forward(input, self.weight, self.bias)
  File "/home/user/.pyenv/versions/miniconda3-3.8-23.9.0-0/envs/promptir/lib/python3.8/site-packages/torch/nn/modules/conv.py", line 459, in _conv_forward
    return F.conv2d(input, weight, bias, self.stride,
RuntimeError: Expected canUse32BitIndexMath(input) && canUse32BitIndexMath(output) to be true, but got false.  (Could this error message be improved?  If so, please report an enhancement request to PyTorch.)

Could this be because my tensor sizes are too big? Would resizing my input images help? Is there a specific size you used for images when training and testing? I did not make any modifications to the code.

va1shn9v commented 2 months ago

Thanks a lot for your interest in our work, I think this is happening due to large sizes, you can resize your images to 256x256 or 128x128 or alternatively using the --tile option in the demo.py file