learnopencv
learnopencv copied to clipboard
tutorial issues
Hi, rank amateur here trying to follow along with the tutorial on https://learnopencv.com/paired-image-to-image-translation-pix2pix/, am using the pytorch code from https://github.com/spmallick/learnopencv/blob/master/Image-to-Image-Translation-with-GAN/PyTorch/pix2pix_pytorch.ipynb
Basic bug report: in the import calls torchsummary has been changed to torchinfo - with that edit the Jupyter cell runs.
Bigger issue: the link to download the data on the tutorial page throws a 404 error. I found the dataset on another site and have the train and val folders, but am running into a problem in the Jupyter notebook in cells [7] and [8]; it looks like the code is calling for additional files:
FileNotFoundError Traceback (most recent call last)
~\anaconda3\envs\tf-gpu3080\lib\site-packages\torchvision\datasets\folder.py in init(self, root, transform, target_transform, loader, is_valid_file) 308 is_valid_file: Optional[Callable[[str], bool]] = None, 309 ): --> 310 super(ImageFolder, self).init(root, loader, IMG_EXTENSIONS if is_valid_file is None else None, 311 transform=transform, 312 target_transform=target_transform,
Perhaps this is happening because I downloaded the data from another site and do not have the anticipated file structure set up? Any assistance would be greatly appreciated. Thank you for this tutorial and the clear discussion you provide.