Skip to content

gee-gendo/diffusers-dev-container

 
 

Repository files navigation

Gendo diffuser training devcontainer

Setup devcontainers in a fresh VM and train :hugging_face: Hugging Face sdxl example.

## Overview

Files added from the examples:

  1. .devcontainer/devcontainer.json is the key file that describes the devcontainer for VS code.
  2. setup-env.sh creates the python env with venv and installs the python dependencies
  3. examples/text_to_image/train-sdxl.sh just calls the instructions from the original examples/text_to_image/README_sdxl.md

Setup

  1. Download repo and cd into it git clone github.com/gee-gendo/diffusers-dev-container
  2. In VS code, Reopen with dev container. Install the extension if needed.
cmd + shift + P
Reopen in container
You can click show logs to follow the build
  1. Run setup script
# Setup python env and dependencies
./setup-env.sh
  1. Select python interpreter (in .venv folder).
  2. Run training script
cd examples/text_to_image
./train-sdxl.sh

You can then check the logs to confirm everything is running. GPU usage available as usual with the nvidia-smi cli.

About

🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch and FLAX.

Resources

License

Code of conduct

Contributing

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 99.9%
  • Other 0.1%