question about new label

kaiwang960112 / AFM

This is novel noisy-robust Attentive Feature MixUp method.

21 stars 5 forks source link

question about new label #1

Open carry-xz opened 3 years ago

carry-xz commented 3 years ago

hi，nice work! I found that in your code， you didnot calcuate the new soft label as in parper, why? loss = args.weight_naive criterion(scores, labels) + args.weight_afm criterion(scores_naive, labels)

kaiwang960112 commented 3 years ago

hello, thank you for your advice. we will check our code again and update it recently.

hwei-cs commented 2 years ago

any update?