first, sigmoid -> BCE loss
second, softmax -> multilabel classification cross-entropy
but, parameter sharing situation
sorry, I am not good at writing English.
multi label learning very well, but binary learning not good.
loss = loss1 + loss2
Should I give weight to loss?
I would appreciate your reply.
I assume you have two different outputs in your model, i.e. one using the
nn.BCELoss and the other for the
Now one part of your model learn quite good, while the other gets stuck?
A weighting of these losses might be a good idea.
Could you compare the ranges of both losses and try to rescale them to a similar range?
Also, as a small side note, if you are using
nn.CrossEntropyLoss for classification, you should pass the logits to this criterion, not the probabilities using
Thank you. Your answer was a great help