Practical Data Science

Repo for www.practialdatascience.org

This respository uses the sphinx document builder together with nbconvert to (mostly) convert jupyter notebooks into a static HTML website.

It reads source material from the source folder (mostly jupyter notebooks, with a few .rst restructured text files (like the home page in index.rst), and when you run sphinx uses those to build an HTML site in docs/html, which is then hosted by github pages.

Hello world!

To Build

• Clone repository
• Install all relevant sphinx packages for converting notebooks to HTML: conda install -c conda-forge sphinx nbsphinx recommonmark pip install sphinx_markdown_tables
• cd to the root directory of repository.
• Run sphinx-build source docs/html

The site will be built into docs/html.

You can also install sphinx-autobuild (pip install sphinx-autobuild), and from the root directory run python3 -m sphinx_autobuild source docs/html. This will create a process that watched the source folder, and any time it see changes, re-build the repo. If you're doing iterative development, it can be helpful to leave this running to auto-build after ever change so you can look at how your changes are working in your local browser.

You can open the site locally by double-clicking on docs/index.html (which just redirects to docs/html/index.html).

Copyright Nicholas Eubank, 2021