https://github.com/vaibhavkarve/multihypergraph

A simple python package for graph theory that supports multi-edges, hyper-edges, looped-edges and every other combination of these.
https://github.com/vaibhavkarve/multihypergraph

functional-programming graph graph-theory graphs hypergraph hypergraphs multigraph python python3 typing

Last synced: 2 months ago
JSON representation

A simple python package for graph theory that supports multi-edges, hyper-edges, looped-edges and every other combination of these.

• Host: GitHub
• URL: https://github.com/vaibhavkarve/multihypergraph
• Owner: vaibhavkarve
• License: gpl-3.0
• Created: 2019-02-07T18:48:05.000Z (about 7 years ago)
• Default Branch: master
• Last Pushed: 2019-03-17T21:21:11.000Z (about 7 years ago)
• Last Synced: 2025-09-22T17:24:43.069Z (6 months ago)
• Topics: functional-programming, graph, graph-theory, graphs, hypergraph, hypergraphs, multigraph, python, python3, typing
• Language: Python
• Size: 115 KB
• Stars: 14
• Watchers: 2
• Forks: 2
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# multihypergraph

A simple python package for graph theory that supports multi-edges, hyper-edges, looped-edges and every other combination of these. This package is an exercise in defining statically-typed, functional implementations of graphs that have the following features:

  - Each vertex of the graph is a characted. For example `'ab,bc,ac'"` is a triangle graph with vertices `{'a', 'b', 'c'}`.

  - Multi-edges are allowed. `'ab,ab'` is a valid graph.

  - Hyper-edges are allowed. `'abc'` is a valid graph.

  - Self-loops are allowed. `'aa'` is a valid graph.

  - Collapsed edges are allowed. `'aab'` is a valid graph

Check out the wikipedia entries for [Hypergraph](https://en.wikipedia.org/wiki/Hypergraph) and [Multigraph](https://en.wikipedia.org/wiki/Multigraph). Think of this package as happy marriage between the two.

## Installation

  

    pip install multihypergraph

## Features

  - Almost all the code is functional.

  - Mutability of data types is never used.

  - All types are explicitly mentioned using static-typing (and checked courtesy `mypy`).

  - The above mentioned feature makes it easier to reason mathematically about the code.

  - The emphasis is on mathematical understanding and soundness rather than algorithmic efficiency.

  - The top-level scripts are divided into `objects/` and `morphisms/` to emphasize the category-theoretic structure of multihypergraphs.

  - Functional programming allows us to write easy-to-decipher tests using `pytest`.