Serim Ryou

Hi, this code was implemented based on the previous version of PyTorch. It seems like there is a numerical instability for the log computation, so I modified the code based...

What's the number of classes? I recommend using warm-up with cross-entropy for the first few epochs (setting ``AnchorLoss(warm_up == True)``) and feed the epoch number while training. I think it's...

Hi, thanks for your interest! Could you tell me more about the target? You have to design your anchor probability based on your target distribution. Are you using the gaussian...