https://github.com/jo-phil/hopkins-statistic

A Python package for computing the Hopkins statistic to assess clustering tendency.
https://github.com/jo-phil/hopkins-statistic

cluster-analysis cluster-tendency clustering data-analysis data-science hopkins-statistic machine-learning python statistical-analysis statistics

Last synced: 21 days ago
JSON representation

A Python package for computing the Hopkins statistic to assess clustering tendency.

• Host: GitHub
• URL: https://github.com/jo-phil/hopkins-statistic
• Owner: jo-phil
• License: mit
• Created: 2025-12-02T20:21:09.000Z (6 months ago)
• Default Branch: main
• Last Pushed: 2026-05-07T21:31:49.000Z (27 days ago)
• Last Synced: 2026-05-07T23:41:11.031Z (27 days ago)
• Topics: cluster-analysis, cluster-tendency, clustering, data-analysis, data-science, hopkins-statistic, machine-learning, python, statistical-analysis, statistics
• Language: Python
• Homepage: https://jo-phil.github.io/hopkins-statistic/
• Size: 109 KB
• Stars: 1
• Watchers: 0
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • Changelog: CHANGELOG.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# hopkins-statistic

[![CI](https://github.com/jo-phil/hopkins-statistic/actions/workflows/ci.yml/badge.svg)](https://github.com/jo-phil/hopkins-statistic/actions/workflows/ci.yml "View workflow runs")

[![Coverage](https://img.shields.io/endpoint?url=https://gist.githubusercontent.com/jo-phil/dd436ee381602e21922433dfcdd0b9ec/raw/coverage.json)](https://github.com/jo-phil/hopkins-statistic/actions/workflows/ci.yml?query=branch%3Amain+event%3Apush "View coverage report")

[![PyPI - Version](https://img.shields.io/pypi/v/hopkins-statistic)](https://pypi.org/project/hopkins-statistic/ "View latest release on PyPI")

[![PyPI - Python Version](https://img.shields.io/pypi/pyversions/hopkins-statistic)](https://pypi.org/project/hopkins-statistic/ "View supported Python versions on PyPI")

[![Docs](https://github.com/jo-phil/hopkins-statistic/actions/workflows/docs.yml/badge.svg)](https://jo-phil.github.io/hopkins-statistic/ "Read the documentation")

[![DOI](https://img.shields.io/badge/DOI-10.5281%2Fzenodo.18172791-blue)](https://doi.org/10.5281/zenodo.18172791 "View on Zenodo")

[//]: # (--8<-- [start:opener])

hopkins-statistic is a library for computing the Hopkins statistic to assess

clustering tendency (also known as clusterability) by testing for departure from

complete spatial randomness (CSR) in point patterns.

This implementation defaults to the [Cross and Jain (1982)] formulation, raising

distances to the power of the data dimension. In two dimensions, this matches

the original definition by [Hopkins and Skellam (1954)]. Because the resulting

statistic follows a beta distribution under the null hypothesis, *p*-values can

be computed analytically, avoiding the need for Monte Carlo simulations.

[Cross and Jain (1982)]: https://doi.org/10.1016/S1474-6670(17)63365-2 "Measurement of Clustering Tendency"

[Hopkins and Skellam (1954)]: https://doi.org/10.1093/oxfordjournals.aob.a083391 "A New Method for determining the Type of Distribution of Plant Individuals"

[//]: # (--8<-- [end:opener])

## Installation

```bash

pip install hopkins-statistic

```

## Usage

```python

import numpy as np

from hopkins_statistic import hopkins

rng = np.random.default_rng(42)

X = rng.uniform(size=(100, 2))

statistic = hopkins(X, rng=rng)

print(f"{statistic:.3f}")

#> 0.514

```

## License

This project is licensed under the terms of the MIT license.