sd-scripts
sd-scripts copied to clipboard
kohya-ss
Hi There my issue is that if i train an sdxl Model on 1 GPU the training takes like 5 hours for 50 epocs which sounds about right  as...
bmaltais says this is a problem in sd-scripts Changing the prompt.txt during training produces the following message when the next sample point is reached, at which point it hangs. I...
Using sd3 Branch code training Flux LoRA the following error occured. ```python /opt/conda/lib/python3.10/site-packages/torch/utils/checkpoint.py:61: UserWarning: None of the inputs have requires_grad=True. Gradients will be None warnings.warn( Traceback (most recent call last):...
I'll attach two screenshots of what I see when attempting to start up my trainings. No idea what's going on with it.  
hello and thanks i found the tokenizer in comfyui inside comfy > text_encoders , and i was edit it i need to do same edit for traning but where i...
FLUX's layer structure is huge and confusing. I'd like to be able to isolate which layers I'm training (which on SDXL allows me to train style-only LoRAs more quickly that...
2 Issues 1. sometimes it works, but it looks nothing like the fine tuned model. 2. Doesnt work and i get the error below During handling of the above exception,...
I am able to successful db training and samples look good, but after 600 steps it gives me this error 2024-08-19 07:57:06 INFO train_util.py:5123 INFO saving checkpoint: train_util.py:5124 /home/Ubuntu/Downloads/model 4/Rick-step00000600.safetensors...
