Skip to content

Train Dreambooth models and LoRAs locally with nvidia-docker

Notifications You must be signed in to change notification settings

jchook/kohya-trainer-docker

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Kohya Trainer Docker

Dockerized Linaqruf/kohya-trainer so you can train dreambooth LoRAs and fine-tune Stable Diffusion models on your own hardware running Linux.

Features

  • Easy set-up - One command to both install and launch.
  • Easy config - Manage your project on the host filesystem.
  • Straightforward - Nothing fancy. Default settings. Simple Dockerfile.
  • Non-root - Files are owned by your user. Nothing runs as root.
  • Fast - Start-up and shutdown are snappy.

Requirements

  • A Linux OS
  • A recent Nvidia GPU
  • docker
  • docker-compose
  • nvidia-docker

How to Use

Enjoy a streamlined experience with just:

just build
just up

Otherwise, you can use the standard docker tools as expected:

docker compose up

See Also

About

Train Dreambooth models and LoRAs locally with nvidia-docker

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published