SAFMN icon indicating copy to clipboard operation
SAFMN copied to clipboard

Published 20 hours ago • verified publisher icon sunny2109

RealGANModel模式各个损失的loss_weight设置

Open erxianzhe opened this issue 1 year ago • 3 comments

在RealGANModel模式下，各个损失的loss_weight应该如何设置比较合理

erxianzhe avatar Dec 26 '23 12:12 erxianzhe

在RealGANModel模式下，各个损失的loss_weight应该如何设置比较合理

不好意思，回复晚了。 训练的时候，我用到了这些loss: L1, FFT Loss, Perceptual Loss, LDL Loss, GAN Loss Stage 1: L1 + 0.1*FFT Loss; Stage2: L1 + 0.1*GAN Loss + Perceptual Loss; Stage 3: L1 + 0.1*GAN Loss + Perceptual Loss + 0.5*LDL Loss

sunny2109 avatar Jan 05 '24 03:01 sunny2109

In RealGANModel mode, how should the loss_weight of each loss be set more reasonably?

Hi, did you manage to train the model in RealGANModel mode?

harusaju avatar Jan 11 '24 12:01 harusaju

In RealGANModel mode, how should the loss_weight of each loss be set more reasonably?

Hi, did you manage to train the model in RealGANModel mode?

Yep, the degradation settings are almost identical to RealESRGAN.

sunny2109 avatar Jan 12 '24 05:01 sunny2109