Jean Zay: Python and Python libraries

The Python ecosystem has many optimised tools for scientific computations and data processing. The most well known are available on Jean Zay via the modules.

The Python modules encapsulate Conda environments and contain libraries installed via conda or pip.

We still offer Python 2.7 even though the support and updates have been stopped since 01/01/2020. As new versions of Python are published, we create the modules and their names indicate the version installed (the associated libraries are also the latest ones compatible at the moment of module creation).

• To see the list of modules: module avail python
• To load a module (and, therefore, indirectly a conda environment): module load <name>

If these modules are incomplete for your usage, you have the following options:

• Request the addition of a library via a ticket from the IDRIS support team (assist@idris.fr).
• Add the missing package yourself locally via pip install --user --no-cache-dir <package>.
• Create a new conda environment via conda create.

The first option is much preferred for the reasons of performance and compatibility with the Jean Zay environment. This also enables us to provide other users with the benefit of these tools.

To know more about the advantages and risks of the other options, refer to the page: Python personal environment.

If you wish to use the mpi4py package, you should also load one of the available MPI modules (module avail openmpi) via the command module. No MPI module is loaded by default in the Python environments (however, the mpi4pi library is available).

The module purge, module switch and module unload commands are not functional with the conda environments. If you wish to use another module, it is preferable to launch another session.