https://github.com/gbroques/dbscan

DBSCAN density-based clustering algorithm in Python.
https://github.com/gbroques/dbscan

clustering-algorithm dbscan dbscan-clustering density-based-clustering eps python python-3

Last synced: 7 months ago
JSON representation

DBSCAN density-based clustering algorithm in Python.

• Host: GitHub
• URL: https://github.com/gbroques/dbscan
• Owner: gbroques
• License: mit
• Created: 2018-04-22T23:30:40.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2022-12-08T00:59:24.000Z (almost 3 years ago)
• Last Synced: 2025-04-08T01:51:17.299Z (8 months ago)
• Topics: clustering-algorithm, dbscan, dbscan-clustering, density-based-clustering, eps, python, python-3
• Language: Jupyter Notebook
• Size: 290 KB
• Stars: 10
• Watchers: 2
• Forks: 1
• Open Issues: 4
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# DBSCAN

[![Build Status](https://travis-ci.org/gbroques/dbscan.svg?branch=master)](https://travis-ci.org/gbroques/dbscan)

[![Coverage Status](https://coveralls.io/repos/github/gbroques/dbscan/badge.svg?branch=master)](https://coveralls.io/github/gbroques/dbscan?branch=master)

Density-Based Spatial Clustering of Applications with Noise (DBSCAN) implementation in Python.

API inspired by Scikit-learn.

## Usage

```python

import numpy as np

from dataviz import generate_clusters

from dataviz import plot_clusters

from dbscan import DBSCAN

def generate_data(num_clusters: int, seed=None) -> np.ndarray:

    num_points = 20

    spread = 7

    bounds = (1, 100)

    clusters = generate_clusters(num_clusters, num_points, spread, bounds, bounds, seed)

    return np.array(clusters)

num_clusters = 4

clusters = generate_data(num_clusters, seed=1)

dbscan = DBSCAN(eps=7, min_samples=5)

dbscan.fit(clusters)

plot_clusters(clusters, dbscan.labels_, dbscan.components_)

```

![dbscan plot](dbscan-plot.png)

* Red crosses denote **core points**

**Reference:**  *Introduction to Data Mining* (1st Edition) by Pang-Ning Tan

Section 8.4, Page 526

**Original Paper:** Ester, Martin, Hans-Peter Kriegel, Jörg Sander, and Xiaowei Xu. "A density-based algorithm for discovering clusters in large spatial databases with noise." In Kdd, vol. 96, no. 34, pp. 226-231. 1996.