Training LoRa necessitate a lot of coffee. If you think that what I do is useful and want to help me keep doing this you can buy me a coffe here: https://ko-fi.com/clumsy_trainer
This style is a mixture of modern and classic illustration, it works very well for caricature and cartoonish style
!!!!!!! THE MODEL IS OVERTRAINED ON PURPOSE !!!!!!!
use it at 0.4 weight or sh!t will happen
Description
first version, lets see how it goes
FAQ
Comments (5)
I get "RuntimeError: The size of tensor a (64) must match the size of tensor b (62) at non-singleton dimension 3" only when I run this lora on latest Auto1111
if you have installed Locon/lycoris extensions try to put them in a backup extension directory (effectively "uninstalling" the extensions) and then restart A1111. I, and many others, noticed that the latest A1111 update broke the Locon/lycoris extensions
I got this error:
loading Lora F:\sd\stable-diffusion-webui\models\Lora\GOTCHA_V0.01.safetensors: RuntimeError
Traceback (most recent call last):
File "F:\sd\stable-diffusion-webui\extensions-builtin\Lora\lora.py", line 215, in load_loras
lora = load_lora(name, lora_on_disk.filename)
File "F:\sd\stable-diffusion-webui\extensions-builtin\Lora\lora.py", line 176, in load_lora
module.weight.copy_(weight)
RuntimeError: output with shape [1, 320, 1, 1] doesn't match the broadcast shape [1, 320, 3, 3]
another issue is the link of your version of RealisticVision referenced in your demo is broken, could you kindly verify if it's still working? thanks.
Everything is working for me. If you are using a1111 and have Lycoris or LoCon extensions installed remove them or in alternative you can try to use additional network extension instead of the native LoRa plugin of a1111. Also make sure to use Gotcha with 1.5 based model. Hope this will help. Let me know
hello.. is it possible to create an SDXl version?