Seeking explanation on attribute classification task loss

fdjingyuan / Deep-Fashion-Analysis-ECCV2018

Codes of ECCV 2018 workshop paper "Deep Fashion Analysis with Feature Map Upsampling and Landmark-driven Attention"

83 stars 27 forks source link

Seeking explanation on attribute classification task loss #2

Open pbamotra opened 5 years ago

pbamotra commented 5 years ago

Hi authors,

Can you please explain your implementation of loss for the attribute classification branch. I'm no able to comprehend the 2*1000 output layer and the use of cross entropy on that. That data is multi-label classification if I'm not wrong, how do you modify the cross entropy loss to use on multi-label data, it should've been Sigmoid+BCELoss if I'm following the online PyTorch tutorials correctly.

Thanks

Gaumera commented 5 years ago

hi, I also have confusion about the 2*1000 output size while data have 1000 attributes itself. can you please explain about this issue?