How to properly use focal_loss from Pytorch with the Hugging face trainer?
A multi-class audio classification task with the hubert model
https://pytorch.org/vision/main/_modules/torchvision/ops/focal_loss.html
1 Like