LSEP as a loss function

I was reading up on some of the latest Kaggle competition winners and it seems like LSEP loss is a good contender for multi-label classification.

Has anyone ever used it with any success? If so, could you share your implementation? I’ve been stuck trying to create one.


i spend some time implementing LSEP loss function for keras with classweights. It seems to be working great for me, i be glad to have some feedback on some improvments.

1 Like

Thanks for sharing! I actually got a working version too but it was terribly slow (I had ~300 possible labels and I modified it so that it would work with MixUp and Spectral Masking, so there was a lot going on). Have you tested it against BCE or with a high number of labels? Curious if you had the same issues as me…