TorchDistiller icon indicating copy to clipboard operation
TorchDistiller copied to clipboard

Published 20 hours ago • verified publisher icon irfanICMLL

Is it better to combine CWD loss with other losses than just CWD loss?

Open jialeli1 opened this issue 3 years ago • 0 comments

Hi!

As mentioned in README: To train a model with channel-wise distillation, GAN loss and Pixel-wise distillation.

Is it better to combine CWD loss with other losses than just CWD loss?

jialeli1 avatar Dec 03 '21 06:12 jialeli1