This service allows JupyterHub users to create multiple workspace environments, called “projects.” Each project allows users to independently install extensions, libraries, data files and other dependencies necessary for their workflows.
Users should be able to leverage notebook projects to keep independent work separate, to manage mutually-conflicting dependencies and to support a diverse array of computational tools.
- Copy the
projects
directory to your Python path. - Configure JupyterHub to run projects as a managed service. Look at
jupyterhub_config.py
for example configuration. - Place the
start-projects.py
andprojects_config.py
files in the directory from which the service will be run. You may need to changeprojects_config.py
point to the directories where you want to store user data. - Place the contents of the
static
directory in your Jupyter share directory, this will allow the custom templates to load the necessary js and css for the frontend.