Closed cankucuksozen closed 3 years ago
Hi! Have you found any possible explanation for the my_softmax? I am quite confused about this one. Maybe I missed some details in the paper.
I think that was done because at that time the pytorch softmax did not have a dim input variable as it does now.
On Fri, May 28, 2021 at 4:38 PM Aoran Wang @.***> wrote:
Hi! Have you found any possible explanation for the my_softmax? I am quite confused about this one. Maybe I missed some details in the paper.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/ethanfetaya/NRI/issues/25#issuecomment-850425344, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABG6XQQDXEZB424FXEKZMSTTP6ME5ANCNFSM4PNRMHAQ .
Many thanks!
I think that was done because at that time the pytorch softmax did not have a dim input variable as it does now. … On Fri, May 28, 2021 at 4:38 PM Aoran Wang @.***> wrote: Hi! Have you found any possible explanation for the my_softmax? I am quite confused about this one. Maybe I missed some details in the paper. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#25 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABG6XQQDXEZB424FXEKZMSTTP6ME5ANCNFSM4PNRMHAQ .
Why does the my_softmax function seems to be normalizing alongside the batch dimension instead of classes dimension?