Just like any other Hugo site, you first need to create a new Hugo project. Follow Hugo’s documentation for more information.
The gist of how pyhugodocs works: you list the objects to be documented in a “reference directory” (usually called _api-reference). This folder is set to be ignored by Hugo (you have to add that to your config.toml!)
pyhugodoc then reads these objects, gets their docstrings, and generates a api-reference folder automatically.
This is the folder that Hugo will render.
Check out pyhugodoc’s GitHub repository for a sample!
Once you have installed pyhugodoc, you need to create a config file to tell it where to look for
content.
The config file should be called pyhugodoc.yaml and should be in the root of your repository.
Sample:
site_dir: docs # the name of the folder in which the Hugo site is
reference_dir: content/_api-reference # path to reference dir, relative to site_dir
docstring_style: google # the docstring style used; can be google or numpy