Works best with protogen3.4
If using automatic1111 you need an extension called Kohya additional networks
Description
FAQ
Comments (3)
Getting this error running the LoRA
LoRA weight: 1, model: bdelphineLora760x1024_bdelphine760x1024v2(c1e28729)
Error running process: C:\Users\drago\stable-diffusion-webui-master\stable-diffusion-webui-master\extensions\sd-webui-additional-networks\scripts\additional_networks.py
Traceback (most recent call last):
File "C:\Users\drago\stable-diffusion-webui-master\stable-diffusion-webui-master\modules\scripts.py", line 347, in process
script.process(p, *script_args)
File "C:\Users\drago\stable-diffusion-webui-master\stable-diffusion-webui-master\extensions\sd-webui-additional-networks\scripts\additional_networks.py", line 238, in process
network, info = lora_compvis.create_network_and_apply_compvis(du_state_dict, weight, text_encoder, unet)
File "C:\Users\drago\stable-diffusion-webui-master\stable-diffusion-webui-master\extensions\sd-webui-additional-networks\scripts\lora_compvis.py", line 73, in create_network_and_apply_compvis
network_dim = min([s for s in size if s > 1])
ValueError: min() arg is an empty sequence
I'm really struggling to make this work. I managed to get 'some' results using a very large prompt and the 'Deliberate' model, and it also required her name in the prompt. But i can't get anything close to your results with your settings. I downloaded kohya, copied your generation data to the letter and am using the protogen model, but none of my results bare any resemblance at all, with or without the same seed, it's like the lora just isn't triggering or something. Any tips?
very nice. WIll have to try this one out. I had trained my own Belle dreambooth that can be seen over at https://www.deviantart.com/erosdivinus
My recent work is my model merged with protogenx34





