# scikit-learn-contrib/hdbscan

A high performance implementation of HDBSCAN clustering.

Repository: https://github.com/scikit-learn-contrib/hdbscan
Canonical: https://ross.abutalabs.com/products/hdbscan
Homepage: http://hdbscan.readthedocs.io/en/latest/
Language: Jupyter Notebook
License: BSD-3-Clause
License Family: permissive
Topics: machine-learning, machine-learning-algorithms, clustering, clustering-algorithm, cluster-analysis, clustering-evaluation
Last push: 2026-06-12T16:59:29+00:00

## Health v2 (maintenance only)
Score: 85/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 87, release rhythm 74, longevity 100
- inputs: {"age_days": 4151, "days_push": 82, "days_rel": 93, "gap_med": 37, "n_releases_24m": 8}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3136, forks 534 (observed 2026-08-28T04:07:45.700759+00:00)

## What it is
A high-performance Python implementation of the HDBSCAN hierarchical density-based clustering algorithm, part of the scikit-learn-contrib ecosystem. It finds clusters of varying densities with minimal parameter tuning, making it well suited to exploratory data analysis.

## Use cases
- cluster data with varying densities without tuning epsilon
- find meaningful clusters in exploratory data analysis
- replace DBSCAN with a more robust density-based clustering algorithm
- cluster high-dimensional embeddings like sentence or image vectors
- identify noise and outliers in unlabeled datasets
- perform hierarchical density-based clustering in Python

## When to choose
- you need clustering that works well with little or no parameter tuning
- your data contains clusters of varying densities that DBSCAN handles poorly
- you want a scikit-learn-compatible, well-documented clustering library

## When to avoid
- you need to specify an exact number of clusters, as HDBSCAN determines them automatically
- you need extremely fast clustering on very large datasets where simpler algorithms like k-means suffice
- you require a non-Python environment

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, data-science
- domain: machine-learning, data-science, analytics
- platform: python, cross-platform
- tags: clustering, hdbscan, dbscan, density-based-clustering, unsupervised-learning, scikit-learn

## Member repositories
- scikit-learn-contrib/hdbscan (main) score 85

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:45.700759+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:26:05.728046+00:00, confidence not recorded.
  - readme: https://github.com/scikit-learn-contrib/hdbscan (fetched 2026-08-28T04:07:45.700759+00:00, sha 5f5dacf1583a)
  - registry_pypi: https://pypi.org/pypi/hdbscan/json (fetched 2026-08-29T09:41:21.695306+00:00, sha 3eee94c96313)
- Data as of 2026-08-30T08:39:29.467469+00:00.