Dreambooth Training (JoePenna) On Vast.ai

A Guide To Running Joe Penna’s Dreambooth Stable Diffusion 1.5 On Vast.ai

Note : This is now a member-only story. However if you join the Stable Diffusion Training Discord Server, and ask me there (@yushan.777), I will give you a Friends link to view with free access.

Table of Contents

Training Images
Setup Vast.ai
Launch The Instance & Jupyter
Cell : Build Environment
Cell : Download The 1.x Model
Cell : Dreambooth Training Environment Setup
Cell : Training
Delete and Destroy your GPU Instance
Other Notes
— Max Training Steps
— Learning Rate
— Flip
— Token
— Using Your Own Regularization Images
— How Do I Change Repeats?
— Who is Joe Penna?

Update : Guide updated for repo commit 847eda0 (April 16 2023)

If you are reading this then you are probably wanting to train a face, object or style into a Stable Diffusion model to make…