Official PyTorch implementation of U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation
I am a little confused with 2 CAM loss functions (5) and (6) in your paper. I don't know why one function uses BCE for generator, one uses MSE for discriminator? Can someone explain it? Many thanks.
zihaog0724
commented
2 years ago
Dear sir,
I am a little confused with 2 CAM loss functions (5) and (6) in your paper. I don't know why one function uses BCE for generator, one uses MSE for discriminator? Can someone explain it? Many thanks.