Neuromatch Course Template

A course template built with JupyterBook v2 (MyST).

Live book: https://neuromatch.github.io/course-template/

For course authors

Fork this repository and follow the instructions in the book itself. Each day of this template teaches you how to build and publish your course. For scrollytelling pages (a pinned stage of figures and live Python beside scrolling text), see W1D2 Tutorial 6, "Scrollytelling Pages".

After forking

Update these things before adding content:

1. In myst.yml: change project.title, project.authors, and project.github
2. In README.md: change the Live book URL to match your GitHub Pages URL
3. In GitHub: go to Settings → Pages and set Source to GitHub Actions (set your custom domain there too, if you have one)

The first push to main publishes the site and creates notebooks-branch for the generated notebooks. Pull requests are checked and built, but not published.

The course kit

The MyST plugin, the nma-kit command (notebooks, checks, site preparation) and the CI workflows come from the published Neuromatch course kit (neuromatch/course-kit): the npm package @neuromatch/course-kit (referenced by extend: in myst.yml) and the PyPI package neuromatch-course-kit (the nma-kit command). This template holds no kit code, only content and configuration; kit fixes reach your course on its next build (@v0.1 in .github/workflows/course.yml).

Local preview

Requires Node.js >= 20 (use nvm use if you have .nvmrc support) and Python 3.10+.

uv venv
uv pip install -r requirements.txt neuromatch-course-kit
uv run nma-kit check                                # content rules, as in CI
uv run nma-kit sync-assets                          # branding files (_nma/), re-run after kit upgrades
uv run myst start

The book will be available at http://localhost:3000. For scrollytelling pages, run uv run nma-kit render-scrolly first so their panels show real outputs.