Preinstalled libraries and adding your own

Workspaces come with the scientific Python stack already installed. Check what you have before installing anything:

import numpy, pandas, scipy, sklearn, torch
for m in (numpy, pandas, scipy, sklearn, torch):
    print(f"{m.__name__:12s} {m.__version__}")

What is included

NumPy, pandas, SciPy, scikit-learn, and PyTorch, plus common geospatial and imaging libraries. Between them these cover most data analysis, scientific computing, and model training work.

Runtime is Python 3.12.

What is not supported

warning

Eugo does not support TensorFlow, spaCy, or JAX.

This is the constraint most likely to derail a migration, so establish it early. If a pipeline depends on one of them, that dependency needs replacing before the rest of the move is worth planning. PyTorch is the supported path for deep learning work.

Adding a package

%pip install some-package

%pip rather than !pip: the magic installs into the kernel actually running your notebook, whereas the shell version can install into a different interpreter and leave you with a package that imports in the terminal but not the notebook.

Session-scoped versus workspace-scoped

Installing from a notebook affects that environment. For something the whole team needs, install into the workspace so it persists and everyone shares it.

If each person installs into their own session instead, every new session pays the setup cost, and environments quietly diverge until someone's code works and someone else's does not.

Workers need it too

A package your task body imports must be available on the compute nodes, not only in your session. Workspace-level installation covers both. A session-only install does not, which produces an ImportError inside a task while the same import succeeds in the notebook.

Pin what matters

For reproducible work, record versions:

%pip freeze > /workspace/requirements-lock.txt

Keep that file in version control alongside the code. "It worked in August" is not a recoverable state without it.