huggingface / huggingface/diffusers

Full SDXL DreamBooth finetuning, LoRA extraction

Open
#7,997 7 comments 1 reaction 0 assignees View on GitHub
stale
Dominant language
Python
Stars
34.5k
Forks
7.3k
Avg merge
3d 3h
Merged PRs (30d)
91

Description

In the dreambooth community, it has been empirically shown that extracting a LoRA results in better performance than directly training a LoRA. Enabling a full dreambooth finetune of SDXL would not only enable this functionality, but also further enable users to extract and test out LoRAs of multiple different network ranks without having to rerun the training script multiple times.

Contributor guide

Open the contributing guide

Research direction

Start by locating the repository's SDXL DreamBooth training entry point and the existing LoRA training or extraction support. Read the related implementation and discussion before deciding the scope. Done means users can perform full SDXL DreamBooth finetuning, extract LoRAs, and test multiple network ranks without rerunning training.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.