Problematic Implementation of EMD Loss

[VoVAllen]

Should be the L2 distance between CDF of two distributions but not between the PDF of two distributions

And there's some typo in the naming such as emb and emd

[yunxiaoshi]

As you say, the implementation is strictly mirrored from the paper, if you take a closer look you will see that the implementation is correct.

[VoVAllen]

Sorry for the mistake. I realized it's right. Thanks!
Another minor problem I found is Dropout. Due to pytorch features, you need to explicit call model.eval() in valuation and test period to turn the Dropout into right mode to get right forward result.

[yunxiaoshi]

Fixed the dropout issue
I didn't see any typo that you mentioned

[VoVAllen]

可能是我看错了……应该没有typo……

[VoVAllen]

Also I see you might also need to add model.eval() at valuation period

[luqiang360]

https://github.com/kentsyx/Neural-IMage-Assessment/blob/9c50b3e384a88a8afdc00333d01656be5526bfed/model.py#L36

The EMD loss still seems to be wrong, my opinion is the sum operation should be inside of torch.abs