Stellars JupyterHub for Data Science Platform

Multi-user JupyterHub 4 with Miniforge, Data Science stack, and NativeAuthenticator.

This platform is built to support multiple data scientists on a shared environment with isolated sessions. Powered by JupyterHub, it ensures secure, user-specific access via the NativeAuthenticator plugin. It includes a full data science stack with GPU support (optional), and integrates seamlessly into modern Docker-based workflows.

This deployment provides access to a centralized JupyterHub instance for managing user sessions. Optional integrations such as TensorBoard, MLFlow, or Optuna can be added manually via service extensions.

References

This project spawns user environments using docker image: stellars/stellars-jupyterlab-ds

Visit the project page for stellars-jupyterlab-ds: https://github.com/stellarshenson/stellars-jupyterlab-ds

Quickstart

Docker Compose

Download compose.yml and config/jupyterhub_config.py config file
Run: docker compose up --no-build
Open https://localhost/jupyterhub in your browser
Add admin user through self-sign-in (user will be authorised automatically)
Log in as admin

Start Scripts

start.sh or start.bat – standard startup for the environment
scripts/build.sh alternatively make build – builds required Docker containers

Authentication

This stack uses NativeAuthenticator for user management. Admins can whitelist users or allow self-registration. Passwords are stored securely.

Deployment Notes

Ensure config/jupyterhub_config.py is correctly set for your environment (e.g., TLS, admin list).
Optional volume mounts and configuration can be modified in jupyterhub_config.py for shared storage.

Customisation

You should customise the deployment by creating a compose_override.yml file.
Example below introduces custom config file jupyterhub_config_override.py to use for your deployment:

# Copyright (c) Jupyter Development Team.
# Distributed under the terms of the Modified BSD License.

services:

  # service for management of a series of users and their jupyterlab
  # environments. it is internally managed via a proxy that redirects
  # users to their dedicated environments
  jupyterhub:
    volumes:
      - ./config/jupyterhub_config_override.py:/srv/jupyterhub/jupyterhub_config.py:ro # config file (read only)

2.5 KiB Executable File Raw Blame History Unescape Escape