Generative_Adversarial_Networks_PyTorch icon indicating copy to clipboard operation
Generative_Adversarial_Networks_PyTorch copied to clipboard
verified publisher icon AaronYALai

Some question about D_loss

Open ran337287 opened this issue 8 years ago • 0 comments

hello @AaronYALai , in run_InfoGAN.py, 'D_loss=D_criterion(outputs[:,0],labels)' is the mean of 2batchSize loss between outputs and labels, but I think D_loss=truth_input_loss+fake_input_loss, that should be the mean of 1batchSize loss. I'm looking forward to your reply.

ran337287 avatar Aug 14 '17 15:08 ran337287