Python

Python is a popular programming language used for data analysis, simulation, machine learning, and much more.

On our HPC systems, we do not recommend using the clusters’ version of Python. Instead, we suggest creating your own Python environment. This gives you control over which packages and versions you install and avoids interfering with others.

There are a few ways to set up Python environments, depending on your needs:

Which method should I use?

We support different ways of using Python on our clusters. Follow the links below to learn how to set up and manage Python for your work:

Method	When to use it
Conda environments	Recommended for most users. Easy package management, especially for scientific packages. Simplifies environment sharing and reproducibility.
Virtualenv	A lightweight alternative to Conda. Useful if you only need standard Python packages. Does not rely on the conda package manager.
Containers (Apptainer)	Required on GPU clusters such as AISurrey; optional elsewhere. Best for advanced use cases or custom environments. Excellent for portability and reproducibility.

Need help?

If you are not sure which method to use, or something is not working, please reach out, see Available Support.

Note

The instructions on these pages assume you already know some basic Python. If you’re new to Python as a language, we suggest starting with an online course such as W3Schools Python Tutorial or The Python Tutorial.

We have also compiled a list of suitable courses on LinkedIn Learning, which you can access for free with your university account.