Skip to content

DOCKER_SETUP.md

Latest commit

 

History

History
73 lines (48 loc) · 1.56 KB

DOCKER_SETUP.md

File metadata and controls

73 lines (48 loc) · 1.56 KB

Docker Setup Guide

For End Users (Quick Start)

Just use the pre-built images - no building required!

Option 1: Docker Compose (Recommended)

# 1. Clone the repository
git clone https://github.com/ls3-lab/QueryGym.git
cd QueryGym

# 2. Set your API key
export OPENAI_API_KEY="sk-..."

# 3. Run with Docker Compose
docker compose run --rm querygym

# Or start Jupyter
docker compose up jupyter

Option 2: Direct Docker Run

docker pull ghcr.io/ls3-lab/querygym:latest
docker run -it --rm --gpus all ghcr.io/ls3-lab/querygym:latest

See the full Docker guide for more examples.

For Developers (Building Locally)

If you need to build images locally for development:

# Build GPU image
make build

# Build CPU image  
make build-cpu

# Test builds
make test

# Clean up
make clean

The Makefile is for local development only. End users should use pre-built images.

Files Overview

  • compose.yml / docker-compose.yml - Uses pre-built GHCR images (for end users)
  • Dockerfile - GPU image definition (for building)
  • Dockerfile.cpu - CPU image definition (for building)
  • .dockerignore - Build optimization
  • Makefile - Development commands (for building locally)
  • examples/docker/ - Ready-to-use examples with notebooks

Images

  • GPU (default): ghcr.io/ls3-lab/querygym:latest
  • CPU (lightweight): ghcr.io/ls3-lab/querygym:cpu

Images are automatically built and published on every release.