本地训练LORA

飞书用户4246

2024年10月24日修改

成果展示

训练步骤

下载地址：https://github.com/cocktailpeanut/fluxgym

common.docs_name - LarkCCM_Docs_Menu_Image

Supported Models

1.
Flux1-dev​

Flux1-dev2pro (as explained here: https://medium.com/@zhiwangshi28/why-flux-lora-so-hard-to-train-and-how-to-overcome-it-a0c70bc59eaf)

3.
Flux1-schnell (Couldn't get high quality results, so not really recommended, but feel free to experiment with it)​

4.
More?​

The models are automatically downloaded when you start training with the model selected.​

You can easily add more to the supported models list by editing the models.yaml file. If you want to share some interesting base models, please send a PR.

Install Manually

First clone Fluxgym and kohya-ss/sd-scripts:

代码块

git clone https://github.com/cocktailpeanut/fluxgym​
cd fluxgym​
git clone -b sd3 https://github.com/kohya-ss/sd-scripts​

Your folder structure will look like this:

代码块

/fluxgym​
  app.py​
  requirements.txt​
  /sd-scripts​

Now activate a venv from the root fluxgym folder:

If you're on Windows:

代码块

python -m venv env​
env\Scripts\activate​

If your're on Linux:

代码块

python -m venv env​
source env/bin/activate​

This will create an env folder right below the fluxgym folder:

代码块

/fluxgym​
  app.py​
  requirements.txt​
  /sd-scripts​
  /env​

Now go to the sd-scripts folder and install dependencies to the activated environment:​

代码块

cd sd-scripts​
pip install -r requirements.txt​

本地训练LORA​

本地训练LORA