What Python versions are supported in the CI config?

The default CI matrix is configured to support Python 3.11, 3.12, and 3.13, matching the modern requirements of the BayesFlow ecosystem.

Does it support optional dependencies?

Yes, it provides both the pyproject.toml configuration for optional extras and the Python code patterns required to safely guard imports for those features.

Why does this skill enforce the src-layout convention?

The src-layout is a best practice that prevents accidental imports of the local package during development, ensuring that tests run against the actual installed version of the package.

How does this skill help with IDE auto-completion?

By strictly managing the __all__ list in __init__.py, the skill ensures that all public functions and classes are discoverable by IDEs and visible to users using star imports.

BayesFlow Package Architect

Name: BayesFlow Package Architect
Author: matthiaskloft

bymatthiaskloft

0•

Ciencia de Datos y ML

Standardizes the creation and management of BayesFlow extension packages with precise export handling and CI configuration.

This skill provides a structured framework for developing BayesFlow extensions, ensuring consistent package architecture through the src-layout convention. It automates critical packaging tasks such as managing public API exports via the __all__ list, configuring pyproject.toml with specific dependency groups, and implementing robust version detection via importlib.metadata. By following these established patterns, developers can avoid common pitfalls like invisible symbols or IDE auto-completion failures, while maintaining high code quality with pre-configured Ruff and Mypy settings tailored specifically for the BayesFlow ecosystem.

Características Principales

01Implements src-layout convention for clean, professional Python package structures

02Provides guarded import patterns for handling optional 'extra' dependencies

03Automates __all__ export management to ensure public API visibility and IDE support

04Generates standardized pyproject.toml templates with optional dependency support

05Sets up GitHub Actions CI matrices for Python versions 3.11 through 3.13

060 GitHub stars

Casos de Uso

01Configuring CI/CD pipelines and linting rules for BayesFlow-compatible projects

02Initializing a new BayesFlow extension with a production-ready file structure

03Managing complex public API exports and versioning for shared ML libraries

Características Principales

01Implements src-layout convention for clean, professional Python package structures

02Provides guarded import patterns for handling optional 'extra' dependencies

03Automates __all__ export management to ensure public API visibility and IDE support

04Generates standardized pyproject.toml templates with optional dependency support

05Sets up GitHub Actions CI matrices for Python versions 3.11 through 3.13

060 GitHub stars

Casos de Uso

01Configuring CI/CD pipelines and linting rules for BayesFlow-compatible projects

02Initializing a new BayesFlow extension with a production-ready file structure

03Managing complex public API exports and versioning for shared ML libraries