Skip to content

Jupyter for courses

The Jupyter for courses app is an app which makes using a custom Python environment simple when hosting or participating in courses.

Python environments can be defined as modules in the project persistent storage. Default resources for the course Python environment can also be defined in the same directory.

Using the app

In the form for launching the application:

  • Select the project and reservation used for the course. The reservation field will only be visible if you have an active reservation you have access to use.
  • Select the course module
  • Launch the application

Creating a course environment

The files for course environments (modules) can be created under /projappl/project_<project-number>/www_lumi_modules. The directories can be created if they do not exist.

The course environment is only visible for the project that it was created for. Note that you may need to Restart Web Server in the Help menu in the web interface if the course environment is not visible in the form after creating the files and selecting the correct project.

Two files are needed for the course modules to be visible in the web interface:

  • a <course>.lua file that defines the module that sets up the Python environment. Only files containing the text Jupyter will be visible in the app. For more information about writing these module files, check the documentation for creating and using your own modules. This module file will be loaded using module load when launching Jupyter.
  • a <course>-resources.yml that defines the default resources used for Jupyter. If this file is omitted, the resources must explicitly be defined in the resource settings in the form.


Module (/projappl/project_1234567/www_lumi_modules/some-course.lua):

-- Jupyter
-- Relative to the course dir
-- Anything valid for checkout
-- lab / notebook / empty (defaults to jupyter)
Resources (/projappl/project_1234567/www_lumi_modules/some-course-resources.yml):
cores: 4
time: "02:00:00"
partition: "interactive"
mem: "16GB"

The Python environment for the course also needs to be created. It is recommended that you use the LUMI container wrapper for creating the Python environment. After creating the environment using the LUMI container wrapper, the full path to the bin directory needs to be added inside the prepend_path in the Lua file for the course above.