search

mcahny / Deep-Video-Inpainting

Official pytorch implementation for "Deep Video Inpainting" (CVPR 2019)
508 stars 95 forks source link

I can not run the code #4

Closed 1145520074 closed 5 years ago

1145520074 commented 5 years ago

I miss the problem as follows, what is wrong ? Thank you very much File "demovi.py", line 166, in outputs, , , , _ = model(maskedinputs, masks_, lstm_state, prev_feed, t) File "/home/vis/zhaoxiang05/video_inpainting/Deep-Video-Inpainting/python3.6/lib/python3.6/site-packages/torch/nn/modules/module.py", line 477, in call result = self.forward(*input, kwargs) File "/home/vis/zhaoxiang05/video_inpainting/Deep-Video-Inpainting/python3.6/lib/python3.6/site-packages/torch/nn/parallel/data_parallel.py", line 121, in forward return self.module(*inputs[0], *kwargs[0]) File "/home/vis/zhaoxiang05/video_inpainting/Deep-Video-Inpainting/python3.6/lib/python3.6/site-packages/torch/nn/modules/module.py", line 477, in call result = self.forward(input, kwargs) File "/home/vis/zhaoxiang05/video_inpainting/Deep-Video-Inpainting-master/models/vinet.py", line 174, in forward flow2 = self.flownet(f1, f2) File "/home/vis/zhaoxiang05/video_inpainting/Deep-Video-Inpainting/python3.6/lib/python3.6/site-packages/torch/nn/modules/module.py", line 477, in call result = self.forward(*input, kwargs) File "/home/vis/zhaoxiang05/video_inpainting/Deep-Video-Inpainting-master/models/flow_modules.py", line 132, in forward corr = self.corr(x1.contiguous(), x2.contiguous()) File "/home/vis/zhaoxiang05/video_inpainting/Deep-Video-Inpainting/python3.6/lib/python3.6/site-packages/torch/nn/modules/module.py", line 477, in call result = self.forward(*input, *kwargs) File "/home/vis/zhaoxiang05/video_inpainting/Deep-Video-Inpainting-master/models/correlation_package/modules/correlation.py", line 17, in forward result = CorrelationFunction(self.pad_size, self.kernel_size, self.max_displacement,self.stride1, self.stride2, self.corr_multiply)(input1, input2) File "/home/vis/zhaoxiang05/video_inpainting/Deep-Video-Inpainting-master/models/correlation_package/functions/correlation.py", line 30, in forward self.pad_size, self.kernel_size, self.max_displacement,self.stride1, self.stride2, self.corr_multiply) File "/home/vis/zhaoxiang05/video_inpainting/Deep-Video-Inpainting/python3.6/lib/python3.6/site-packages/torch/utils/ffi/init.py", line 202, in safe_call result = torch._C._safe_call(args, kwargs) torch.FatalError: sizes must be non-negative (THCTensor_resizeNd at /pytorch/aten/src/THC/THCTensor.cpp:108)

mcahny commented 5 years ago

It seems to be related to the flownet2. Please compile the dependencies by running ./install.sh.

AziziShekoofeh commented 5 years ago

@1145520074 I have installed a separate python 3.6 and downgraded my pytorch to 0.4.0 and have been able to use the built-in files directly. My guess is that if you try to re-built the modules from install.sh and make.sh inside lib directory you will bumped to the cuda versioning issue again. Also, Ubuntu 16.04, Python 3.5, Pytorch 0.4.0, CUDA 8.0 is another tested and working setting.

to downgrade to pytorch 0.4.0 you can use:

pip install torch==0.4.0 torchvision

you may need to replace pip with pip3 install or python3.6/3m/3 -m pip install based on your python and pip version.