NewYour coding agent can read the release notes before it upgrades.Set up the MCP server →
PyPI · #2759 most downloaded on PyPI
Clustering based on density with variable density clusters
Last release 4 months ago
01 Jun 2026
Release timing varies
gaps range from 2 weeks to 13 months
Some releases are documented
notes for 27 of 60 stable releases
Nothing withdrawn
no release was ever pulled
11 years old
61 releases · first in 2015
Optimize get_stability_scores: O(n) bincount replaces O(k*n) loop by @sebastianbreguel in #709
Full Changelog: release-0.8.43...release-0.8.44
One column per quarter.
Remove unused cython imports in dist_metrics.pxd by @muitiiifruckt in #704
Full Changelog: release-0.8.42...release-0.8.43
Fix outlier scores by @JelmerBot in #697
Full Changelog: release-0.8.41...release-0.8.42
Deprecated argument names in hdbscan_.py by @joaquindas in #673
Full Changelog: release-0.8.40...release-0.8.41
Adds cythonize() call to extensions in setup.py() by @axiak in https://github.com/scikit-learn-contrib/hdbscan/pull/658
Full Changelog: https://github.com/scikit-learn-contrib/hdbscan/compare/release-0.8.39...release-0.8.40
Full Changelog: release-0.8.39...release-0.8.40
Add property setters for class HDBSCAN by @divyegala in #657
HDBSCAN by @divyegala in #657Full Changelog: release-0.8.38-2...release-0.8.39
Build for python 3.12, deprecate python 3.7
Full Changelog: release-0.8.38-1...release-0.8.38.post2
Full Changelog: release-0.8.38-1...release-0.8.38-2
Nothing published for this version
Pin numpy<2 for now since this breaks builds.
Pin numpy<2 for now since this breaks builds.
Full Changelog: https://github.com/scikit-learn-contrib/hdbscan/compare/release-0.8.35...release-0.8.36
Full Changelog: https://github.com/scikit-learn-contrib/hdbscan/compare/release-0.8.35...release-0.8.36
Fixed DeprecationWarning in hdbscan/hdbscan_.py by @gerald-ftk in https://github.com/scikit-learn-contrib/hdbscan/pull/612
Full Changelog: https://github.com/scikit-learn-contrib/hdbscan/compare/0.8.33...release-0.8.34-rc1
Hopefully the last round of patches to work around recent issues introduced by a newer release of Cython
Hopefully the last round of patches to work around recent issues introduced by a newer release of Cython
Cython 3 introduces a number of issues; prevent using Cython 3 for compilation until we've worked through all the changes required.
Cython 3 introduces a number of issues; prevent using Cython 3 for compilation until we've worked through all the changes required.
Resolve recent build issues with full module name resolution
Resolve recent build issues with full module name resolution
Full Changelog: https://github.com/scikit-learn-contrib/hdbscan/compare/0.8.30...0.8.31
Changes to support scikit-learn 1.3
Full Changelog: https://github.com/scikit-learn-contrib/hdbscan/compare/0.8.29...0.8.30
Fix malformed list in docs by @dmlls in https://github.com/scikit-learn-contrib/hdbscan/pull/535
Full Changelog: https://github.com/scikit-learn-contrib/hdbscan/compare/0.8.28...0.8.29
Nothing published for this version
fix issue-436 unused deprecated joblib option by @neontty in https://github.com/scikit-learn-contrib/hdbscan/pull/438
Full Changelog: https://github.com/scikit-learn-contrib/hdbscan/compare/0.8.26...0.8.27
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
We introduce the density based clustering validity index from Density Based Clustering Validation by Moulavi, Jaskowiak, Campello, Zimek and Sander, a
We introduce the density based clustering validity index from Density Based Clustering Validation by Moulavi, Jaskowiak, Campello, Zimek and Sander, as well as some minor bug fixes, and the option to match the reference implementation of HDBSCAN*
Nothing published for this version
Nothing published for this version
Nothing published for this version
Support for cluster strengths and general bug fixes.
Support for cluster strengths and general bug fixes.
A visual refresh of the plotting routines, along with a series of bug fixes for various corner cases.
A visual refresh of the plotting routines, along with a series of bug fixes for various corner cases.
Nothing published for this version
A very minor performance regression to get this fixed, but it eliminates segfaults on Linux, so is necessary.
A very minor performance regression to get this fixed, but it eliminates segfaults on Linux, so is necessary.
Boruvka scales poorly with dimension; by approximating the minimal spanning tree we can achieve far better scaling with dimension at the cost of sligh
Boruvka scales poorly with dimension; by approximating the minimal spanning tree we can achieve far better scaling with dimension at the cost of slightly less accurate clustering. In testing the loss of accuracy seems very small, so I'm pushing it out with that defaulted to on.
Nothing published for this version
Fix bugs introduced when working around numpy performance regression.
Fix bugs introduced when working around numpy performance regression.
Fix issues with 32 bit Windows installs Fix issues caused by a performance regression in bumpy structured array accesses
Fix issues with 32 bit Windows installs Fix issues caused by a performance regression in bumpy structured array accesses
Nothing published for this version
Better performance Fixes for Robust Single Linkage Faster cluster extraction from trees New Notebooks
Better performance Fixes for Robust Single Linkage Faster cluster extraction from trees New Notebooks
Further optimization of routines that became bottlenecks due to the new faster minimum spanning tree algorithm. Reworking the calling of query on tree
Further optimization of routines that became bottlenecks due to the new faster minimum spanning tree algorithm. Reworking the calling of query on trees to optimize core distance computation in higher dimensional spaces.
Bug fixes for pip installing 0.4
Bug fixes for pip installing 0.4
Nothing published for this version
New algorithms bring dramatic speed improvements, particularly for low dimensional data.
New algorithms bring dramatic speed improvements, particularly for low dimensional data.
Nothing published for this version
Support low memory; Robust single linkage.
Support low memory; Robust single linkage.
Nothing published for this version
Your coding agent can read these notes before it upgrades. Set up the MCP server →