PyTorch-GAN icon indicating copy to clipboard operation
PyTorch-GAN copied to clipboard

Published 20 hours ago • verified publisher icon eriklindernoren

Pixelwise loss in AAE

Open bhaswara opened this issue 4 years ago • 2 comments

@eriklindernoren Is there any reasons why L1 loss is used here instead of using MSE loss for the pixelwise? And why it has to be multiplied with 0.999?

bhaswara avatar Jul 28 '20 00:07 bhaswara

The same question about Context Encoder

russab0 avatar Dec 21 '20 11:12 russab0

I have the exact same question. Can you provide any answer?

LucaLaFisca avatar Jan 12 '23 17:01 LucaLaFisca