Changing supervised regularization term (alpha) from scalar to balanced weight

[stompsjo]

It may be advantageous to change the current implementation of alpha from a scalar directly applied to the labeled contrastive loss term to a balanced weight that works on both versions. i.e. changing this line:

RadClass/models/PyTorch/lightModel.py

Line 242 in 093040a

curr_loss['loss']['losses'][indices] *= self.alpha

To this:

curr_loss['loss']['losses'][indices] *= self.alpha 
curr_loss['loss']['losses'][~indices] *= (1-self.alpha)