Adversarial Loss without any change at 0.69

[sand2sand]

Hello, bro. I want to finetune my own dataset on the 256 model, but why the adversarial loss always remains at 0.69 without any change. I would very much like to get your advice.

[lmvgjp]

hi @sand2sand,

how did you solve the problem?
I am training a model but gives exact images output as original model

thanks!