About WSDAN-CAL in the train_distributed.py

[kanzmh]

I've tried to train this code, but I'm still confused about understanding the code in the train_distributed.py file

In this coding, the aug variable uses the crop method, and the aux variable uses the drop method

But why does the aug_images variable use both methods (crop&drop), not just the crop method?

Thank you

[raoyongming]

Hi, thanks for your interest in our work. It should be a typo. As the printed message here, epoch_crop_acc is the accuracy for images with extra data augmentations including both "crop" and "drop", and epoch_drop_acc is the accuracy for auxiliary predictions. It seems we use the wrong names for these two variables.

[kanzmh]

Can I ask again about the architecture of WSDAN_CAL and the visualization? Cause I'm still quite confused when I read the journal and the model results, thank you for your help.