wtf is numpy-doc?

This repository stores the published documentation for every past version of NumPy, the widely used Python library for numerical computing. The benefit is straightforward: when someone needs to look up how a function worked in an older release of NumPy, the docs for that specific version are archived here and available to browse. The repo is essentially a collection of static web pages organized into folders by version number. When a new NumPy release comes out, a maintainer generates the documentation files from the source code, places them into a new folder named after the release (like 1.18), and updates a central index.html page to include a link to the new set of pages. The end result is a browsable archive of documentation snapshots spanning multiple years of releases. The primary audience is developers and data scientists who are working with older codebases that depend on a specific version of NumPy. For example, if you are maintaining a scientific application built on NumPy 1.18 and need to verify the exact arguments a function accepted at that time, you would visit the archive to find the docs matching your version rather than the current ones, which might reflect newer changes or removed features. The project is purely an archive, not a tool or application. It does not contain the NumPy software itself, just the reference materials. The README doesn't go into detail about how the documentation generation process works beyond pointing to a separate build guide, but the workflow described is a manual process where maintainers build the docs locally and then commit the results to preserve them.