Nerogar
Nerogar
@supermachine77 can OneTrainer access the internet on your machine? This might not be the exact same issue, but it could be related. And do you know if the issue existed...
@mookiexl I couldn't reproduce your problem. It's always loading the local v1-inference.yaml file. Can you attach a debugger, or put some print statements in your code? Specifically, I'm interested in...
> Ok, so apparently it happens only when training a LoRA, finetune works. With LoRA training I can reproduce the issue. Let me think about a solution.
@mookiexl your issue should be solved now
I don't have multiple GPUs to test this. But it seems a bit strange that cuda:0 works, and cuda:1 doesn't. The device name "cuda:1" is just passed to pytorch without...
I don't think the pytorch bug will be fixed anytime soon. We might need to add a better workaround to the model saving code. There are two possible solutions: 1....
Yes. This is an easy to add feature. I will add it. But right now all my time is spent adding stable cascade support.
Completely forgot about this. I added the setting now to the train tab. I don't have a good dataset to test it myself, so let me know how it goes....
The conversion scripts in OneTrainer are custom built. You can find them in modules/util/convert. The Stable Diffusion convert script is called convert_sd_diffusers_to_ckpt.py. I doubt that there are any bugs that...
You mentioned before that using the diffusers CLIP improves quality. Is this not the case anymore? Did you find out something else? Any additional information that could help narrow this...