focal_loss_alt VS focal_loss

kuangliu / pytorch-retinanet

RetinaNet in PyTorch

992 stars 250 forks source link

focal_loss_alt VS focal_loss #37

Open LifeBeyondExpectations opened 6 years ago

LifeBeyondExpectations commented 6 years ago

Why there are two version of focal loss methods in _"class FocalLoss(nn.Module): URL: https://github.com/kuangliu/pytorch-retinanet/blob/2199fd9711fd787ae409800a499db73e6d466fd7/loss.py" ????

bitcurian commented 5 years ago

focal_loss_alt() is a better implementation, it uses a simpler implementation of F.binary_cross_entropy(). Besides in torch 0.4.1 binary_cross_entropy() does not perform backprop. for the weight variable which is crucial for the focal loss since the weight variable includes the modulating factor.

sahu-ji commented 5 years ago

It seems the focal_loss_alt() function doesn't use the parameter gamma. Is that an oversight or has it been implicitly taken care of and I am missing something?

gunshi commented 5 years ago

Was wondering if there is a reference for this alternate loss?