Confused about the reason for refactoring Dataloader code

sanghyun-son / EDSR-PyTorch

PyTorch version of the paper 'Enhanced Deep Residual Networks for Single Image Super-Resolution' (CVPRW 2017)

MIT License

2.43k stars 669 forks source link

Confused about the reason for refactoring Dataloader code #236

Closed wyf0912 closed 4 years ago

wyf0912 commented 4 years ago

Thanks for your contribution.

According to my understand, the refactored dataloader can add the scale label. But the scale label can be returned in getitem. In addation, I think it is thread-unsafe to set scale in _ms_loop

Looking forward to your repy.