How are experiment stages invoked?

Stages are triggered via uv run using a specific stage naming convention and JSON configuration files located in the configs/ directory.

What coding standards does it prioritize?

It emphasizes best practices for ML code, specifically pure functions, detailed docstrings covering shapes and dtypes, and clear documentation of side effects.

How does this skill handle project structure?

It enforces a specific layout where experiments live in src/viterbo/experiments/ and shared helpers are kept in src/viterbo/common/ to prevent premature abstraction.

How does it manage plots for LaTeX?

It generates layout and styles within Python while storing the resulting assets in designated LaTeX-compatible directories for easy inclusion in documents.

What testing tools are supported by this skill?

The skill includes pre-defined scripts for Ruff formatting and linting, Pyright type checking, and Pytest for smoke and targeted testing.

Python Experiment Conventions

Name: Python Experiment Conventions
Author: JoernStoehler

byJoernStoehler

0•

Data Science & ML

Standardizes Python experiment layouts, stage entrypoints, and asset handling for consistent data science workflows.

This skill provides a structured framework for managing Python-based experiments within the Viterbo ecosystem. It enforces specific directory layouts for experiments and shared helpers, defines standardized stage entrypoints, and streamlines asset generation for LaTeX integration. By providing pre-configured linting, smoke testing, and data artifact management commands, it ensures that machine learning and data science projects remain maintainable, reproducible, and aligned with best practices like pure functions and detailed docstrings.

Key Features

01Enforces standardized project layouts for experiments and configurations

02Streamlines stage execution using uv and specific entrypoint patterns

03Integrates automated linting and smoke testing with Ruff and Pyright

04Manages data artifacts and Git LFS paths for reproducible research

05Automates plot and asset generation for seamless LaTeX document integration

060 GitHub stars

Use Cases

01Setting up a new machine learning experiment with consistent directory structures

02Running multi-stage data processing pipelines with standardized config handling

03Generating publication-ready plots and assets for LaTeX research papers

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add joernstoehler/msc-viterbo python-conventions

For use in Claude.ai and ChatGPT

Download Skill