LoRA can not be used in text encoder in "train_text_to_image_lora.py"?

[ZihaoW123]

It can be seen from [(https://github.com/cloneofsimo/lora)] that fine-tuning the lora for the text encoder works better than the lora for unet. When will this repo support the lora of the text encoder?

[haofanwang]

This PR can support it!

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.

Please note that issues that do not follow the contributing guidelines are likely to be ignored.

[patrickvonplaten]

cc @sayakpaul think this is being worked on no?

[sayakpaul]

Actively. Latest update: #2719 (comment)

The community script supports it already, though with the use of peft.

[sayakpaul]

Closing this as we support loading of LoRA into a text encoder when relevant.