What documentation style does this skill follow?

It follows the official PyTorch conventions as seen in torch/_tensor_docs.py and torch/nn/functional.py, utilizing Sphinx and reStructuredText (reST) formatting.

Can I use this for non-PyTorch projects?

Yes, while it is optimized for PyTorch, it is an excellent tool for any Python project that uses Sphinx for documentation and requires high-quality, structured docstrings.

Does it support cross-referencing?

Absolutely. It automatically suggests and formats Sphinx roles like :class:, :func:, and :meth: to link related components within your documentation.

How does it handle mathematical formulas?

The skill uses Sphinx math directives and inline math roles (:math:) to properly render LaTeX equations and tensor shape notation.

Why are docstrings generated as raw strings?

Docstrings use the r""" prefix to ensure that backslashes used in LaTeX math or special characters are treated literally and don't cause Python syntax issues.

PyTorch Docstring Expert

Name: PyTorch Docstring Expert
Author: Activer007

byActiver007

0•

Learning & Documentation

Generates standardized, high-quality docstrings for PyTorch functions and methods following official Sphinx and reStructuredText conventions.

The PyTorch Docstring skill automates the creation of professional, publication-ready documentation for Python codebases, specifically optimized for the PyTorch ecosystem. It ensures that all functions and methods include essential elements such as raw string formatting, Sphinx cross-references, LaTeX mathematical formulas for tensor shapes, and structured sections for arguments, return values, and examples. By strictly adhering to the patterns found in the core PyTorch library, this skill helps developers maintain consistency, improve code readability, and ensure that documentation renders perfectly in Sphinx-based web portals.

Key Features

01Inline LaTeX math support for complex tensor shape descriptions

020 GitHub stars

03Automatic raw string (r""") formatting to support LaTeX/math backslashes

04Structured 'Args', 'Returns', and 'Examples' block generation

05Intelligent Sphinx cross-referencing for classes, methods, and functions

06Sphinx and reStructuredText (reST) compliance for high-quality rendering

Use Cases

01Standardizing documentation across a large-scale deep learning library

02Generating math-heavy descriptions for mathematical functions in PyTorch

03Documenting new custom neural network layers and operators

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add activer007/ordinary-claude-skills docstring

For use in Claude.ai and ChatGPT

Download Skill