How does this skill handle floating-point comparisons?

It utilizes pytest.approx and numpy.testing.assert_allclose to handle precision issues inherent in scientific computing, ensuring tests don't fail due to minor rounding differences.

How do I organize my test files using this skill?

The skill follows the standard recommended layout, placing tests in a separate /tests directory to keep your source code clean and ensure the package is tested in its installed state.

Can it help with CI/CD for my research project?

Yes, it provides templates and guidance for setting up GitHub Actions and other CI tools to automate your testing workflows and ensure code quality.

Does it support testing with NumPy and Pandas?

Absolutely, it includes specific patterns and best practices for verifying multidimensional arrays, dataframes, and other complex data structures common in scientific Python.

Scientific Python Testing

Name: Scientific Python Testing
Author: uw-ssec

byuw-ssec

•

セキュリティとテスト

Implements robust testing suites for scientific Python projects using pytest and industry-standard numerical validation patterns.

This skill provides specialized guidance for testing research software and scientific Python packages, adhering to Scientific Python Community standards. It assists users in configuring pytest, designing reusable fixtures, and implementing complex numerical tests using tools like numpy.testing and pytest.approx. Whether you're validating floating-point calculations, testing stochastic algorithms, or setting up continuous integration for data science pipelines, this skill ensures your scientific code remains reproducible, correct, and well-documented.

主な機能

01Comprehensive fixture design for complex research data and models

0218 GitHub stars

03Integration of code coverage and CI/CD workflows for scientific packages

04Automated pytest configuration and directory structure setup

05Advanced numerical validation using floating-point tolerances and NumPy utilities

06Parametrized testing patterns for multiple scientific input scenarios

ユースケース

01Validating the accuracy of numerical algorithms and scientific simulations

02Implementing automated testing for NumPy, SciPy, and Pandas-based libraries

03Refactoring legacy research code with a fail-fast test-driven approach

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add uw-ssec/rse-plugins python-testing

For use in Claude.ai and ChatGPT

Download Skill