R Observatory: Metrics for mlr3cluster version 0.7.0

Balamuta, James Joseph

doi:10.5281/zenodo.21843040

mlr3cluster

0.7.0Package record

Cluster Extension for 'mlr3'

1packages dependsuggested by 2

63.5Kdownloads / year

91.0%test coverage

13/13flavors OK

Overview

About

Maintained by Maximilian MückeCRAN page ↗GitHub ↗Website ↗Issues ↗

Extends the 'mlr3' package with cluster analysis.

Changes in 0.7.0

Breaking changes

clust.ap, clust.dbscan, clust.dbscan_fpc, clust.hdbscan, clust.meanshift, clust.optics, and clust.stdbscan now store $model as a list. Use $native_model for the fitted upstream object (#129).
clust.dbscan_fpc no longer offers method = "dist", which treated the features as a distance matrix and either failed or returned wrong clusters.
clust.diana no longer exposes stop.at.k, which cluster::diana() ignores since only the default FALSE is implemented.
clust.flexmix now uses the upstream default nrep = 3, which runs the EM algorithm three times and keeps the best fit. Set nrep = 1 to restore the previous behavior.
clust.kkmeans now only accepts p with alg = "kerninghan", and clust.kmeans_rcpp and clust.MBatchKMeans now only accept tol_optimal_init with initializer = "optimal_init", since the upstream functions ignore them otherwise.
clust.kproto now treats logical features as categorical, so a vector-valued lambda must include an entry for each logical feature.
tgen("moons") returns the classification task generator from mlr3 again. mlr3cluster 0.6.0 had replaced it with a clustering version, which has been removed. Use as_task_clust(tgen("moons")$generate(n)) to create a clustering task (#128).

New features

as_task_clust() now converts a TaskClassif or TaskRegr to a TaskClust by dropping the target from the features. This creates cluster tasks from any task generator in mlr3, e.g. as_task_clust(tgen("spirals")$generate(n)).

Other improvements

clust.cobweb, clust.em, clust.ff, and clust.SimpleKMeans now support factor and ordered features, which Weka handles as nominal attributes.

Show all changes in 0.7.0Hide the rest of the changes in 0.7.0

clust.flexmix no longer requires k, so unsetting it infers the number of components from cluster.
clust.kmeans_rcpp and clust.MBatchKMeans now check that CENTROIDS is a matrix when it is set, instead of failing in ClusterR during training.
clust.tclust now supports the Gaussian parsimonious clustering models via restr = "GPCM" and the new pars parameter.
LearnerClust$save_assignments now checks that it is set to a single TRUE or FALSE.

Bug fixes

clust.ap now predicts with the similarity function s used for training, which it stores in the model.
clust.ap can now predict with a similarity function s that has no sel argument, e.g. function(x) -as.matrix(dist(x))^2. Previously, training worked but predicting failed.
clust.clara (metric, medoids.x, and stand), clust.dbscan_fpc (scale and seeds), and clust.pam (stand) now use the training values for prediction checks. Changing these parameters after training no longer bypasses the checks or incorrectly prevents prediction.
clust.cmeans now accepts zero weights, as long as at least one weight is positive, matching e1071::cmeans().
clust.cmeans, clust.kkmeans, clust.kmeans, clust.kmeans_rcpp, clust.kproto, and clust.MBatchKMeans now match initial cluster centers to the task features by column name instead of by position.
clust.cobweb, clust.em, clust.ff, clust.kmodes, clust.kproto, and clust.SimpleKMeans now treat logical features as categorical. Previously, they could be ignored or encoded differently in training and prediction.
clust.dbscan_fpc now accepts whole-number doubles such as c(10, 20) for countmode. Previously, only integer vectors were accepted.
clust.featureless now predicts with the num_clusters used for training.
clust.meanshift now predicts with the convergence parameters thr and iter used for training, which it stores in the model.
clust.optics, clust.skmeans, clust.stdbscan, and clust.tclust now reject hyperparameter values that always fail upstream: minPts = 0 for clust.optics, negative approx for clust.optics and clust.stdbscan, method = "kmndirs" for clust.skmeans, and trace values above 1 for clust.tclust.
clust.protoclust now predicts with the distance parameters method, diag, upper, and p used for training, which it stores in the model.
clust.silhouette now returns NaN instead of an error when every observation is in its own cluster, e.g. for small test sets in resampling.
clust.specc now uses the selected kernel when no kernel parameters are set. Previously, it always used an RBF kernel with an estimated sigma in this case.
clust.xmeans now initializes C to 0.5 so that it is always passed to Weka. Previously, Weka reset C to 0 whenever any other parameter was set.

From NEWS.md in the release

Install

R (CRAN)pak (dev)Ubuntu (r2u)Fedora (COPR)openSUSE (OBS)

install.packages("mlr3cluster", repos = "https://cloud.r-project.org/")

Installs the current CRAN release. Prebuilt binaries on Windows and macOS.

if (!requireNamespace("pak", quietly = TRUE))
  install.packages("pak", repos = "https://cloud.r-project.org/")
pak::pak("mlr-org/mlr3cluster")

Development version from mlr-org/mlr3cluster. Installs pak first if it is not already available.

Release

# enable cranapt (r2u) if it is not already set up, then install
grep -rqs r2u.stat.illinois.edu /etc/apt/sources.list /etc/apt/sources.list.d/ \
  || curl -fsSL https://raw.githubusercontent.com/eddelbuettel/r2u/master/inst/scripts/add_cranapt_noble.sh | sudo bash
sudo apt install -y r-cran-mlr3cluster

Fast Ubuntu binaries via r2u (no compilation). The cranapt setup runs once and is skipped when it is already enabled. We list this channel from recorded r2u download activity rather than from a live index of the repository, so treat it as last known rather than guaranteed.

bspm (used by r2u) also needs make: sudo apt install make

# enable the iucar/cran COPR if it is not already present, then install
ls /etc/yum.repos.d/*iucar*cran*.repo >/dev/null 2>&1 \
  || sudo dnf copr enable -y iucar/cran
sudo dnf install -y R-CRAN-mlr3cluster

RPMs from the iucar/cran COPR, rebuilt from CRAN.

For install.packages() integration in R, also: sudo dnf install R-CoprManager

# add the autoCRAN OBS repo (Tumbleweed) if it is missing, then install
zypper -q lr | grep -qi autoCRAN \
  || sudo zypper addrepo https://download.opensuse.org/repositories/devel:/languages:/R:/autoCRAN/openSUSE_Tumbleweed/devel:languages:R:autoCRAN.repo
sudo zypper install -y R-mlr3cluster

RPMs auto-built from CRAN on openSUSE OBS (Tumbleweed). We list this channel from recorded autoCRAN download activity rather than from a live index of the repository, so treat it as last known rather than guaranteed.

Health

CRAN checks

13OK

CRAN check page ↗

Slowest check: 5.8 min · r-oldrel-windows-x86_64

Status changes

OK13 OK2026-10-15
ERROR12 OK, 1 ERROR2026-09-25
OK13 OK2026-09-21
ERROR12 OK, 1 ERROR2026-09-20
OK13 OK2026-08-05

Show 4 earlier changesHide 4 earlier changes

NOTE12 OK, 1 NOTE2026-08-01
OK13 OK2026-06-08
NOTE12 OK, 1 NOTE2026-06-07
OK14 OK2026-03-10

Advisories

none known

How this is measured·How advisories are found

Documentation

In release 0.7.0

READMEYes · 618 wordsVignettesNoNEWSYes · 100% structured · mentions 0.7.0Citation fileNoWebsiteYes · pkgdown

Help pages: 61includes a package overview page
Examples that run: 100%45 run in a plain check (40 of them under a condition)
Exported functions with examples: 98%
Documented parameters: 75%
Return-value docs: 4%
Help pages citing references: 74%

How these are measured

Downloads

63.5K

CRAN downloads in the past year

Rank #1,733 · ~174/day · ~5.3K/mo

Releases R versions

Daily download trend is not available in this view yet.

4K30 days

12.2K90 days

63.5K1 year

Compare downloads with other packages →

Also on2.9K r2u7 autocran3.4K conda_forge

How downloads are counted

Repository

mlr-org/mlr3cluster ↗Organization: mlr-org

27Stars

6Forks

2Open issues

1Open PRs

17Releases

784Commits

14Commit identities

mlr3cluster-analysisclusteringrr-packagemachine-learning

License file reads as LGPL-3.0 · 784 commits · Last activity 2026-10-18 · 0% stars, 30d · Default branch at 0.7.0.9000, ahead of CRAN 0.7.0 (checked 2026-10-21)

Responsiveness

How these are sampled

PRs merged (all time): 92%
Median days to close, 50 most recently updated closed issues: 29
Median days to close, 50 most recently updated closed PRs: 1

Stars over time

2025-07-28 · 252026-08-31 · 27

How these are measured

PracticesBeta

16 development-tooling and community-health practices detected across 6 families in the upstream repository

Checks run against github.com/mlr-org/mlr3cluster on 2026-10-21.

Continuous integration (2)

GitHub ActionsR CMD check workflow

The R CMD check workflow runs on Linux, and on R-devel

Governance and community (3)

Contributing guide (inherited from mlr-org)

Funding links: GitHub Sponsors

The repository owner accepts GitHub Sponsors

Docs source (4)

README.Rmdpkgdown

GitHub Pages site last deployed 2026-10-18 · The repository has a site deploy workflow

pkgdown site last built 2026-10-18

CRAN release process (1)

cran-comments.md

Show all practicesHide 2 practices

Automation and maintenance (2)

Dependabotpre-commit

Lint, format, editor (4)

lintrAirEditorConfigRStudio project

How this is detected·Detection ruleset v4 (2026-10-20)

AI coding toolsBeta

This repository shows signs of AI coding tools.

Claude Codefirst seen 2026-02-04
Named in .Rbuildignore: .claude, CLAUDE.md
Named in .gitignore: .claude
Pull request it wrote: opened by a project member
Commits: at least 1 crediting it in the message
Gemini CLIfirst seen on or before 2026-07-18
Config: a file or ignore line, not recorded which
Cursorfirst seen 2026-08-17
Named in .Rbuildignore: .cursor

Shared conventions any agent may read, so they name no particular tool.

Seen in the repository: AGENTS.md

Models named in its commits

Claude Opus 51 commit · Sep 2026

Counts are commits, not lines of code. Commits by a tool's accounts count once across all of its accounts. Commits crediting it are the largest count among its credit lines, a minimum.How this is detected·Detection ruleset v4 (updated 2026-09-27)

Dependencies

Declared dependencies

38 external dependencies (excludes base and recommended)

Installing mlr3cluster adds 24 packages (11 with compiled code, 16 MB of source tarballs). Heaviest: data.table 5.7 MB, palmerpenguins 2.9 MB, mlbench 967 KB.

Depends (2)

R >= 3.5.0mlr3

Imports (11)

backports checkmateclusterdata.tablegraphicsmethodsmlr3misc (>= 0.21.0)paradox R6statsutils

LinkingTo (0)

none

Enhances (0)

none

Dependency license breakdown →

Reverse dependencies

1direct

1indirect

Imports 1

Suggested by 2

Depends, Imports or LinkingTo, most downloaded first

mlr3verse

Suggests, most downloaded first

mlr3viz MapperAlgo

How these are counted

Code & Tests

People & History

People (4)

Maximilian Mücke

Author, Maintainer since 0.1.11 (2025) · added in 0.1.10

Michel Lang

Author

Damir Pulatov

Author

Marc Becker

Contributor · added in 0.1.11

Previously maintained by Damir Pulatov (0.1.0 to 0.1.10).

Releases (20)

Pick two releases to compare their code metrics.

0.7.0Latest
record
2026-10-17 · current release · diff ↗ ·
removed 1 export
TaskGeneratorMoons
0.6.0
record
2026-09-17 · diff ↗ ·
added 3 exports
TaskGeneratorBlobs, TaskGeneratorMoons, as_tasks_clust
0.5.0
record
2026-08-21 · diff ↗ ·
added 3 exports
LearnerClustGMeans, LearnerClustKMeansRcpp, LearnerClustKModes
0.4.1
record
2026-07-10 · diff ↗
0.4.0
record
2026-06-11 · diff ↗ ·
added 8 exports
LearnerClustFlexmix, LearnerClustGenie, LearnerClustKCCA, LearnerClustMovMF, LearnerClustSKMeans, LearnerClustSOM, LearnerClustSTDBSCAN, LearnerClustTclust
R
R 4.6.0 released · 2026-04-24
0.3.0
record
2026-03-01 · diff ↗ ·
added 3 exports
LearnerClustCLARA, LearnerClustKProto, LearnerClustSpectral
0.2.0
record
2026-02-04 · diff ↗ ·
added 1 export
LearnerClustProtoclust
0.1.12
record
2025-11-19 · diff ↗
R
R 4.5.0 released · 2025-04-11
0.1.11
record
2025-02-19 · diff ↗
0.1.10
record
2024-10-04 · diff ↗ ·
added 2 exports
LearnerClustBICO, LearnerClustBIRCH
R
R 4.4.0 released · 2024-04-24
0.1.9
record
2024-03-18 · diff ↗ ·
added 3 exports
LearnerClustDBSCANfpc, LearnerClustHDBSCAN, LearnerClustOPTICS
R
R 4.3.0 released · 2023-04-21
0.1.8
record
2023-03-12 · diff ↗

Show 11 earlier eventsHide 11 earlier events

0.1.7
record
2023-03-10 · diff ↗
0.1.6
record
2022-12-22 · diff ↗
0.1.5
record
2022-11-01 · diff ↗ ·
added 1 export
LearnerClustMclust
0.1.4
record
2022-08-14 · diff ↗
R
R 4.2.0 released · 2022-04-22
0.1.3
record
2022-04-06 · diff ↗
0.1.2
record
2021-09-03 · diff ↗ ·
added 3 exports
LearnerClustHclust, as_prediction_clust, as_task_clust
R
R 4.1.0 released · 2021-05-18
0.1.1
record
2020-11-16 · diff ↗ ·
added 8 exports
LearnerClustAP, LearnerClustCobweb, LearnerClustEM, LearnerClustFarthestFirst, LearnerClustKKMeans, LearnerClustMeanShift, LearnerClustMiniBatchKMeans, LearnerClustSimpleKMeans
0.1.0
record
2020-10-01
R
R 4.0.0 released · 2020-04-24

Package metadata

First published: 2020-10-01
Total releases: 20 / 6 yrs
License: LGPL-3 OSI
Minimum R: ≥ 3.5.0
Download size: 274 KB

Appears in task views

Cluster Analysis & Finite Mixture Models

Cite

Cite this package

Run in R for the authors' preferred citation:

citation("mlr3cluster")

Format

Mücke, M., Becker, M., Lang, M., & Pulatov, D. (2026). mlr3cluster: Cluster Extension for 'mlr3' (Version 0.7.0) [Computer software]. https://doi.org/10.32614/CRAN.package.mlr3cluster

This package has no citation file, so citation() returns this entry built from DESCRIPTION.

Cite the R Observatory

For a number measured here: a download total, a coverage figure, an archival date.

Format

APA

Balamuta, J. J. (2026). R Observatory: Metrics for mlr3cluster version 0.7.0 [Data set]. HJJB, LLC. Data release v2026-10-25. https://doi.org/10.5281/zenodo.21843040

From data release v2026-10-25, which the citation names so these numbers can be found later. More on citing and the projects behind them.

Report a problem with this page →