Core Principles

py3plex keeps multilayer and multiplex network analysis accessible with minimal setup, clear APIs, and sensible defaults.

• Practical first: built-in loaders, generators, validators, and a CLI reduce boilerplate.

• Composable workflows: chainable methods, pipelines, and a SQL-like DSL keep analyses repeatable and auditable.

• Interoperable: built on NetworkX so existing algorithms and tooling carry over.

• Visualization-ready: layer-aware layouts and rendering designed for multiplex views.

• Reproducible: configuration-driven workflows and deterministic utilities (set seeds when randomness is involved).

Key Features

• Multilayer API: Modular, chainable interface for constructing, inspecting, and validating multilayer graphs quickly.

• SQL-like DSL: SELECT-style queries to filter nodes or edges and compute measures without verbose code.

• Pipeline support: Scikit-learn–style pipelines for reproducible workflows, benchmarking, and rapid iteration.

• Visualization: Built-in layouts and rendering utilities tuned for multiplex graphs, including layer-specific views.

• Integrations: NetworkX compatibility plus loaders, datasets, and a plugin-friendly architecture for extension.

Learn More

Explore the resources below for deeper context, runnable examples, and the research background behind py3plex.

• Paper: https://link.springer.com/article/10.1007/s41109-019-0203-7

• Examples: https://github.com/SkBlaz/Py3Plex/tree/master/examples

• Repository: https://github.com/SkBlaz/Py3Plex