jupyter-container

jupyter-container

Adds a single-user Jupyter notebook service to your Ansible Container project. Designed for use with JupyterHub.

Run the following commands to install the service:

# Set the working directory to your Ansible Container project root
$ cd myproject

# Install the service
$ ansible-container install marcusianlevine.jupyter-container

Requirements

  • Ansible Container
  • An existing Ansible Container project. To create a project, simply run the following:
    # Create an empty project directory
    $ mkdir myproject
    
    # Set the working directory to the new directory
    $ cd myproject
    
    # Initialize the project
    $ ansible-contiainer init
    

Role Variables

  • py2_env
    • Name of the Python 2 virtualenv added to conda
  • jupyterhub_pip_version
    • Must match the version of JupyterHub installed on the Hub spawning from this image
  • cran_packages
    • List of R packages to install
  • conda2_packages
    • List of conda Python2 paackages to install
  • pip2_packages
    • List of pip Python2 packages to install
  • extra_conda3_packages
    • List of extra conda Python3 packages to install in addition to conda2_packages
  • extra_pip3_packages
    • List of extra pip Python3 packages to install in addition to all listed in pip2_packages
  • extra_lab_extensions
    • List of extra Jupyter Lab extensions (npm packages) to install
  • install_nbextensions
    • Install jupyter_contrib_nbextensions with conda (and associated configurator)

Dependencies

Note: this role is designed to build from a single-user Jupyter notebook base image

License

BSD

Author Information

Written by Marcus Levine for CKM Advisors.

About

Build a custom single-user Jupyter Notebook container for use with JupyterHub

Install
ansible-galaxy install marcusianlevine/ansible-role-jupyter-container
GitHub repository
License
bsd-3-clause
Downloads
82
Owner
Data Science + DevOps = DataEng