Dreambooth Training SDXL Using Kohya_SS (Windows)
Table of Contents
Introduction
Pre-requisites
Initial Setup
Preparing Your Dataset
The Model
Start Training
Using Captions
Config-Based Training
Aspect Ratio / Resolution Bucketing
Resume Training
Batches, Epochs, Steps…
Introduction
I will skip what SDXL is since I’ve already covered that in my vast.ai guide so I’ll just jump right in. If you want a more in-depth read about SDXL then I recommend The Arrival of SDXL by Ertuğrul Demir
kohya_ss supports training for LoRA, Textual Inversion but this guide will just focus on the Dreambooth method. Much of the following still also applies to training on top of the older SD1.5 and SD2.x models.
There are two ways to go about training the Dreambooth method:
Token+class Method: Trains to associate the subject or concept with a specific token word (identifier)+class. i.e. “ohwx person”
No need to provide captions. For example, if you use it to…