input bug

[1shifeng]

Hello,when running python test_GoPro.py --rgb False --load_unet True --load_path Pretrained_Model/RE_Net_GRAY.pth,Traceback (most recent call last):
File "test_GoPro.py", line 163, in
prepare()
File "test_GoPro.py", line 61, in prepare
unet.load_state_dict(torch.load(opt.load_path)['state_dict'])
File "/home/dell/anaconda3/envs/trmd/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1051, in load_state_dict
raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
RuntimeError: Error(s) in loading state_dict for DataParallel:
size mismatch for module.blur_en_conv1.0.weight: copying a param with shape torch.Size([16, 1, 5, 5]) from checkpoint, the shape in current model is torch.Size([16, 3, 5, 5]).
size mismatch for module.pred.weight: copying a param with shape torch.Size([6, 39, 1, 1]) from checkpoint, the shape in current model is torch.Size([18, 41, 1, 1]).
size mismatch for module.pred.bias: copying a param with shape torch.Size([6]) from checkpoint, the shape in current model is torch.Size([18]) appears,what's wrong with setting it to
blurry_channels=1,
event_channels=6,
out_channels=6,
rgb=False):

[chenkang455]

Sry for the bug, I have updated the code in test_GoPro.py and you can follow the modification to run the code.

[1shifeng]

Sry for the bug, I have updated the code in test_GoPro.py and you can follow the modification to run the code.

[1shifeng]

Then how to test on the rgb data set of go pro

[chenkang455]

Then how to test on the rgb data set of go pro

Just try this command:

python test_GoPro.py --rgb True --load_unet True --load_path Pretrained_Model/RE_Net_RGB.pth

[1shifeng]

When python test_GoPro.py --rgb True --load_unet True --load_path Pretrained_Model/RE_Net_RGB.pth,Traceback (most recent call last):
File "test_GoPro.py", line 165, in
metrics = test(test_loader)
File "test_GoPro.py", line 79, in test
res_pre = unet(blur_image,voxel).clip(-1,1)
File "/home/dell/anaconda3/envs/trmd/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/dell/anaconda3/envs/trmd/lib/python3.8/site-packages/torch/nn/parallel/data_parallel.py", line 159, in forward
return self.module(*inputs[0], **kwargs[0])
File "/home/dell/anaconda3/envs/trmd/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/dell/TRMD/network.py", line 94, in forward
blur1 = self.blur_en_conv1(blurry_frame)
File "/home/dell/anaconda3/envs/trmd/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/dell/anaconda3/envs/trmd/lib/python3.8/site-packages/torch/nn/modules/container.py", line 117, in forward
input = module(input)
File "/home/dell/anaconda3/envs/trmd/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/dell/anaconda3/envs/trmd/lib/python3.8/site-packages/torch/nn/modules/conv.py", line 423, in forward
return self._conv_forward(input, self.weight)
File "/home/dell/anaconda3/envs/trmd/lib/python3.8/site-packages/torch/nn/modules/conv.py", line 419, in _conv_forward
return F.conv2d(input, weight, self.bias, self.stride,
RuntimeError: Given groups=1, weight of size [16, 3, 5, 5], expected input[1, 1, 360, 640] to have 3 channels, but got 1 channels instead appears

[chenkang455]

blur_image

Hi, @1shifeng , can you print the opt parameters by print(opt) for me?

[1shifeng]

Namespace(data_path_test='Data/GOPRO/test', load_path='Pretrained_Model/RE_Net_GRAY.pth', load_unet=False, num_bin=7, rgb=False, seed=3407, test_batch_size=1)

[chenkang455]

https://github.com/chenkang455/TRMD/blob/a5a45cfa696179c6ff2a79671980a0266a7a77f8/test_GoPro.py#L159C1-L168C15

Have you update the code?

[1shifeng]

Sorry,I didn't see it.Thank you very much.