Jianyi Wang
Jianyi Wang
We follow the same settings as RealESRGAN. We did not process the data in advance.
> > We follow the same settings as RealESRGAN. We did not process the data in advance. > > Hi, this is still a bit confusing. You mentioned following RealESRGAN,...
Hi. The details of the command has been updated. Feel free to have a try :)
> FYI, while it was said the SD-Turbo model does -not- work with the 768 resolution SD2.1 model, I can confirm that this new "[Trajectory Consistency Distillation](https://arxiv.org/abs/2402.19159)" LoRA model does....
The loss logged out is the loss at the random step. Generally, the loss at a large step, e.g., step 800 is certainly bigger than the loss at later steps,...
Pls show the input and the result if possible.
Hi. Which script are you running?
I did not test on the CPU. I am afraid that the current code may not support CPU.
Hi. Please refer to #44 .
there should be something wrong. most likely to be the same issue as above.