nilearn/nilearn

Nilearn – Machine‑learning for brain imaging

What it is – Nilearn is a Python library that makes it easy to analyse 3‑D brain MRI volumes and cortical surface data. It wraps statistical tools (general linear models) and the scikit‑learn machine‑learning toolbox so you can do things like classification, decoding, and connectivity analysis on neuro‑imaging data.

Why it matters – Brain‑imaging researchers often need to apply sophisticated statistical or ML techniques to large, high‑dimensional datasets. Nilearn provides a high‑level, well‑documented API that abstracts away low‑level neuro‑imaging file handling while still giving access to powerful scikit‑learn algorithms. This lowers the barrier for reproducible, data‑driven neuroscience.

Key features (as described in the README)

• Supports GLM‑based analyses and multivariate statistics via scikit‑learn.
• Handles both volumetric (voxel) and surface (mesh) neuro‑imaging data.
• Includes plotting utilities (matplotlib, optional plotly/kaleido) for visualising results.
• Comprehensive, tutorial‑style documentation and an active community (drop‑in hours, Discord, Mastodon, Bluesky, YouTube).

How to get started

python -m pip install nilearn

The README points to an install page for more detailed setup, and a development guide for contributors.

Community & support

• Weekly online drop‑in hours (Wednesday 4‑5 pm UTC) on Jitsi Meet.
• Public chat on Discord, plus presence on Mastodon, Bluesky, and a YouTube channel.
• Issue tracker and contribution guide linked from the docs.

Where to learn more

• Source code: https://github.com/nilearn/nilearn/
• Stable documentation: https://nilearn.github.io/
• Development docs (contributing, testing, etc.) are also hosted on the same site.

---

All statements are taken directly from the repository’s README; no additional features have been inferred.