Need of forward_backward in partial_fc.py for arcface_torch ?

[bokey007]

Why do we explicitly need to code the backward part in forward_backward in partial_fc.py for arcface_torch?

can't PyTorch auto-differentiation take care of this?

I would really appreciate help on this.

Thanks

[Light--]

same question here. how to simplfy the code to avoid manually compute the arcface gradient (grad and x_grad) in partial_fc.py ？@anxiangsir

[msedebrahimi]

did anyone find an answer or another solution for backward?