Question about softmax loss

[hobbitlong]

Question about softmax loss

[hyBlue]

I'm an idiot

[vinsis]

@hyBlue seems you found out the answer yourself. Could you explain what NCESoftmaxLoss is trying to do here? I am having the same doubt as you initially had (and feeling like an idiot).

[vinsis]

Explanation: label = torch.zeros([bsz]).cuda().long() means the labels are [1, 0, 0, 0, ...] for each sample. label specifies the label of the positive sample, which is 0 in each case.

It's funny how easy it is to get confused even with something you think you are fairly familiar with.

[hyBlue]

@vinsis Great! You're right

[zhyever]

Thanks a lot! (A new idiot)

[jessico]

triple idiot.