RepDistiller Why using log_softmax instead of softmax?

Why using log_softmax instead of softmax?

Open nguyenvulong opened this issue 2 years ago • 1 comments

I think it should be softmax instead. Otherwise p_t and p_s are not comparable.

Could you please explain why? https://github.com/HobbitLong/RepDistiller/blob/dcc043277f2820efafd679ffb82b8e8195b7e222/distiller_zoo/KD.py#L13-L17

Jul 04 '22 08:07 nguyenvulong

for those who have similar question like i did: https://github.com/yoshitomo-matsubara/torchdistill/issues/233

Jul 04 '22 18:07 nguyenvulong