stable-diffusion-webui icon indicating copy to clipboard operation
stable-diffusion-webui copied to clipboard

Published 20 hours ago • verified publisher icon AUTOMATIC1111

[Feature Request]: PLEASE AUTOMATIC1111 add Lora

Open alexfredo opened this issue 2 years ago • 5 comments

Is there an existing issue for this?

  • [X] I have searched the existing issues and checked the recent builds/commits

What would your feature do ?

Hi I want to know if is possible to add Lora it's a dreambooth that require only 6 go of vram https://github.com/cloneofsimo/lora it's could be awesome, thanks !

Proposed workflow

  1. Go to ....
  2. Press ....
  3. ...

Additional information

No response

alexfredo avatar Dec 08 '22 15:12 alexfredo

Wow, the output is very small. 3.24mb

ClashSAN avatar Dec 08 '22 15:12 ClashSAN

@d8ahazard from Dreambooth extension is already working on it

leppie avatar Dec 09 '22 11:12 leppie

Fwiw, there's already a working implementation in the v21 branch my dream booth extension. It should get merged into main today.

d8ahazard avatar Dec 09 '22 13:12 d8ahazard

Fwiw, there's already a working implementation in the v21 branch my dream booth extension. It should get merged into main today.

I just tried it, but I can't seem to be able to select the Lora Model (below the Model dropdown) for training. Also, if you click "Train", you get an error: TypeError: main() got an unexpected keyword argument 'lora_model' leppie has already made a comment regarding this on your last commit. By the way, thanks for all this work!

athu16 avatar Dec 09 '22 13:12 athu16

Just pushed a fix for that. Sorry. That's why it's not in main last night. I was working on it last evening until I couldn't keep my eyes open, pushed to git just for a work sync. ;)

On Fri, Dec 9, 2022 at 7:44 AM Atharva Vaidya @.***> wrote:

Fwiw, there's already a working implementation in the v21 branch my dream booth extension. It should get merged into main today.

I just tried it, but I can't seem to be able to select the Lora Model (below the Model dropdown) for training. Also, if you click "Train", you get an error: TypeError: main() got an unexpected keyword argument 'lora_model' leppie has already made a comment regarding this on your last commit. By the way, thanks for all this work!

— Reply to this email directly, view it on GitHub https://github.com/AUTOMATIC1111/stable-diffusion-webui/issues/5524#issuecomment-1344326739, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAMO4NEB53N76QVOUNEMOFDWMMZUZANCNFSM6AAAAAASYH6N44 . You are receiving this because you were mentioned.Message ID: @.***>

d8ahazard avatar Dec 09 '22 13:12 d8ahazard

When can/will it be merged into main?

Centurion-Rome avatar Dec 18 '22 17:12 Centurion-Rome

I created the two pt files from lora's colab, can it be used in automatic 1111 ? i tried but without success.

svumme avatar Dec 27 '22 16:12 svumme

As an end user, actually using the lora models is not currently very friendly. I would suggest that the checkpoint merger tab gain the ability to merge the lora pt files into a base model.

Zarxrax avatar Jan 01 '23 18:01 Zarxrax

extension https://github.com/kohya-ss/sd-webui-additional-networks and DB repo have support for it now

ClashSAN avatar Jan 08 '23 06:01 ClashSAN