Foreword:

????Better read my last document before reading this. Lora is a plugin for Stable_Diffusion, allowing trainings on personal computer. Your customized training file should be put here:

????stable-diffusion-webui\models\Lora

????Here I recommend this webUI for training:

????https://github.com/bmaltais/kohya_ss

Dependences:
??? a.Python 3.10.x
????b.Git(Optional)
??? c.Pro access to Internet

????d.Visual Studio 2015, 2017, 2019, and 2022 redistributable

????????Link: https://aka.ms/vs/17/release/vc_redist.x64.exe

Getting started:

????You get the repository from the Github page above.

Deployment:

????Attention, better create a virtual environment of Python before starting. If not, warnings of version mismatching should occur.

????Open "setup.bat". Choose "Install kohya_ss gui" and then "Torch 2". Then the time-costing download will begin.

??? After downloading, you will see "Configuring accelerate...". Then, several preferences should be made.

????For me, I chose "This machine", "No distributed training", "no", "no", "no", "all", "no". You may change them for yourself.

????Then the deployment is done.

References:

????[1]https://github.com/bmaltais/kohya_ss

This document is still being completed.