The size of tensor a (192) must match the size of tensor b (3) at non-singleton dimension 2

[ayush-angelium]

Hii @sergeywong , when I tested this model for custom images it gives an error, I have also implemented all the possible solution that mentioned in Issues like :

1. I used im_h.permute in cp_dataset.py
2. Installed different torchvision version(0.2.1).
3. converted rgb to grey image for tensor resize.
4. used different normalize in cp_dataset.py self.transform = transform = transforms.Compose([transforms.ToTensor(),transforms.Normalize((0.5,), (0.5,))])
5. manually resize in (1,256,192) but still facing tensor size issue.

Here are all possible outcomes that I applied when implementing this model. Please help me to solve this problem.

[greentree20131221]

Because your images in image-parse are 24 bit RGB images, and the images in image-parse in dataset are indexed images. You should save the parsed images as indexed color image format or grayscale.

[ayush-angelium]

"Down to the latest detail" is better than cpvton model in terms of accuracy. That's I'm switching this model.

[xuCW1]

Hii @sergeywong , when I tested this model for custom images it gives an error, I have also implemented all the possible solution that mentioned in Issues like :

1. I used im_h.permute in cp_dataset.py
2. Installed different torchvision version(0.2.1).
3. converted rgb to grey image for tensor resize.
4. used different normalize in cp_dataset.py self.transform = transform = transforms.Compose([transforms.ToTensor(),transforms.Normalize((0.5,), (0.5,))])
5. manually resize in (1,256,192) but still facing tensor size issue.

Here are all possible outcomes that I applied when implementing this model. Please help me to solve this problem.

Hello, I also encountered this problem when testing the model for custom images. How did you solve it at last?