scikit-learn-contrib/hdbscan
A high performance implementation of HDBSCAN clustering.
This tool helps data analysts and researchers find natural groupings (clusters) within their complex datasets. You input your raw data, and it outputs labels indicating which cluster each data point belongs to, even if clusters are of different densities or have noise. It's designed for anyone working with data who needs to identify underlying structures or patterns without extensive trial-and-error.
3,080 stars. Used by 16 other packages. Actively maintained with 4 commits in the last 30 days. Available on PyPI.
Use this if you need to quickly and reliably find clusters in your data, especially when you suspect clusters might have varying densities or your data contains significant noise, without extensive parameter tuning.
Not ideal if you require every single data point to be assigned to a cluster, as this algorithm is designed to identify and leave out noisy data points.
Stars
3,080
Forks
531
Language
Jupyter Notebook
License
BSD-3-Clause
Category
Last pushed
Jan 26, 2026
Commits (30d)
4
Dependencies
4
Reverse dependents
16
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/scikit-learn-contrib/hdbscan"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
annoviko/pyclustering
pyclustering is a Python, C++ data mining library.
panagiotisanagnostou/HiPart
Hierarchical divisive clustering algorithm execution, visualization and Interactive visualization.
erdogant/clusteval
Clusteval provides methods for unsupervised cluster validation
mqcomplab/MDANCE
MDANCE: O(N) clustering for molecular dynamics. Process 1.5M frames in 40min. 8 specialized algorithms.
wq2012/SpectralCluster
Python re-implementation of the (constrained) spectral clustering algorithms used in Google's...