Save Training Result

[HieuQN]

Amazing work.
I want to save training results during the training, but unsuccessful. Below is the codes that you commented out.
# val_images.extend(
# [display_transform()(val_hr_restore.squeeze(0)), display_transform()(hr.data.cpu().squeeze(0)),
# display_transform()(sr.data.cpu().squeeze(0))])
# val_images = torch.stack(val_images)
# val_images = torch.chunk(val_images, val_images.size(0) // 15)
# val_save_bar = tqdm(val_images, desc='[saving training results]')
# index = 1

what is val_hr_restore and sr.data.cpu()
thanks for your time.

[HieuQN]

Would you able to do a small tutorial on how to train and eval on the network? Highly appreciate it.

[ReNginx]

Hi there. Thanks for the comment. This project was made in a hurry. So it's kind of sloppy everywhere. The commented out code actually comes from Implementation from another person's SRGAN implementation (mentioned at last of README). Those code generates and saves sample images during training. We are not using that so we just commented out. Could you give more detail about your problem? I'll try my best to help.