311_code
311_code
I also have a side issue sort of related. When trying to train 113,246 images, if one single image is bad it messes up the latent caching, and have to...
Edit: I just ended up reading the kohya-ss scripts cascade branch and got the answer to quesiton: ` The first time, specify --text_model_checkpoint_path and --save_text_model to save the Text Encoder...
I followed [this new video from yesterday](https://www.youtube.com/watch?v=yPOadldf6bI) and it appears he leaves "image variations" at 1 even when latent caching is on. The wiki mentions not to leave it at...
Deleted last workflow post as was too much detail for this thread. Already posted that elsewhere. I'm having problems with onetrainer results now, and putting my likeness in various situations....
The attn windowing doesn't full work as it relies on xformers and not available yet in pytorch I read, but the code still makes svd a lot better when turning...
> do you mind if I include this? Sorry for delay, thanks for asking! Yeah go right ahead. It did take me about 3 weeks though lol, so maybe you...
You can disable this artificial limitation just like the 2 people ref image limitation. Just paste code in chatgpt 4o and it will tell you the specific lines to remove.
Yes, but there is some code that was added to temporarily disable two people when using ref images, use chatgpt 4o to find it and you can simple remove it,...
New problem, everything works with the custom models but anytime you want to use offline it errors out with this: ``` To create a public link, set `share=True` in `launch()`....