R Observatory: Metrics for MoEClust version 1.6.0

Balamuta, James Joseph

doi:10.5281/zenodo.21843040

MoEClust

1.6.0Package record

Gaussian Parsimonious Clustering Models with Covariates and a Noise Component

1packages depend

5.3Kdownloads / year

–test coverage

13/13flavors OK

Overview

About

Maintained by Keefe MurphyCRAN page ↗GitHub ↗Issues ↗

Clustering via parsimonious Gaussian Mixtures of Experts using the MoEClust models introduced by Murphy and Murphy (2020) doi:10.1007/s11634-019-00373-8. This package fits finite Gaussian mixture models with a formula interface for supplying gating and/or expert network covariates using a range of parsimonious covariance parameterisations from the GPCM family via the EM/CEM algorithm. Visualisation of the results of such models using generalised pairs plots and the inclusion of an additional noise component is also facilitated. A greedy forward stepwise search algorithm is provided for identifying the optimal model in terms of the number of components, the GPCM covariance parameterisation, and the subsets of gating/expert network covariates.

Changes in 1.6.0

New Features & Improvements

Various improvements to MoE_gpairs (also see additional Bug Fixes below):
_Significant_ fixes when there are expert covariates and diag.pars$show.dens=TRUE &/or

response.type="density" by properly using log average density instead of average log density.

Marginal densities when diag.pars$show.dens=TRUE are now always evaluated over evenly-spaced

grids, the size of which can now be modified via diag.pars$diag.grid (equal to 100, by default):

previously, the grids were formed using the observed values, which led to strange behaviour.

Added density.pars$dens.points=FALSE for overlaying points when response.type="density".
Various improvements in relation to subset args.:
data.ind & cov.ind can now be character strings / variable names (previously numeric indices only).
Added submat for showing only "upper"/"lower" triangular & "diagonal" plot panels.
When submat="all", the slowness of response.type="density" plots is now offset

Show all changes in 1.6.0Hide the rest of the changes in 1.6.0

by using densities pre-calculated from upper-triangular panels for lower-triangular panels.

MoE_Uncertainty gains two new arguments:
col: default of "cluster" colours according to cluster-membership, but

old behaviour of highlighting uncertain observations can be recovered via col="uncertain".

rug1d (TRUE, by default) for use with univariate models, which puts

the actual observed values along the x axis when type="barplot".

MoE_control gains new init.z option "soft.random": the "random" option has been

renamed to "random.hard", but init.z="random" will work as before due to partial matching.

tau0 can now always be supplied as a vector (previously allowed only with gating covariates & noise.gate=TRUE).
Speed improvements by replacing matrixStats::rowLogSumExps with new logsumexp & softmax

functions from mclust (w/ mclust (>= 6.1) now ensured in Imports:) where appropriate throughout.

stats::lm.wfit-related speed-ups from previous update now extend to MoE_gpairs with scatter.type="lm".
Further related minor speed-ups for models with G == 0 and G == 1.

Bug Fixes & Miscellaneous Edits

Additional minor fixes to MoE_gpairs:
Subsetting is now allowed to result in only one single panel.
Additional edits in relation to diag.pars$show.dens:
show.dens=TRUE now works properly when subset$data.ind is used.
The expert.covar arg. is no longer invoked when show.dens=TRUE.
Fixed (i.e. increased) height of diagonal panels when show.dens=TRUE &/or show.hist=TRUE.
Additional edits in relation to barcode panels:
Partially fixed dimensions of vertical panels when conditional="barcode"

(caution still advised when using RStudio's "Plots" pane if non-square).

Barcode panels now have colour throughout (previously only MAP-related panels),

with related minor fixes when barcode.pars$use.points=TRUE.

Minor label-related adjustments:
Outer labels of mosaic panels are now correct when diagonal=FALSE.
Cosmetic adjustments to default orientation of labels matching categorical variable levels.
density.pars$show.labels="mixed" now works properly.
Many documentation improvements & clarifications.
Minor fixes in relation to MoECriterion objects and MoE_plotCrit:
Bug fix in relation to non-finite values for direct plots of MoECriterion objects, e.g. plot(x$BIC).
crit="loglik" formerly erroneously produced the same plot as crit="aic".
New crit options "df" & "iters" added.
Fixed bugs when a 'soft' z.list is supplied when algo != "EM".
MoE_estep & MoE_cstep now work when there is only one observation, with a related

fix to predict.MoEClust(..., use.y=TRUE) when predicting only one observation.

Fixed extremely rare bug in MoE_clust & associated predict, fitted, & residuals methods

when algo="CEM" and a model has only one observation/prediction assigned to its noise component.

Fixed minor bug when as.Mclust is used with expert.covar=TRUE for multivariate models

with expert network covariates and subsequently used to produce density-related plots.

Additional minor documentation improvements.

From inst/NEWS.md in the release

Install

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

install.packages("MoEClust", 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("Keefe-Murphy/MoEClust")

Development version from Keefe-Murphy/MoEClust. 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-moeclust

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-MoEClust

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-MoEClust

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: 3.5 min · r-oldrel-windows-x86_64

Status changes

OK13 OK2026-09-08
ERROR12 OK, 1 ERROR2026-09-07
OK13 OK2026-08-05
NOTE12 OK, 1 NOTE2026-08-01
OK14 OK2026-04-10

Show 2 earlier changesHide 2 earlier changes

ERROR13 OK, 1 ERROR2026-04-09
OK14 OK2026-03-10

Advisories

none known

How this is measured·How advisories are found

Documentation

In release 1.6.0

READMEYes · 521 wordsVignettesYes · 1 · dynamicNEWSYes · 100% structured · mentions 1.6.0Citation fileYes · cites a publicationWebsitenot declared in DESCRIPTION

Help pages: 34includes a package overview page
Examples that run: 76%26 run in a plain check · 6 only when donttest examples are enabled (R CMD check --as-cran) · 2 hold no code
Exported functions with examples: 100%
Documented parameters: 84%
Return-value docs: 100%
Help pages citing references: 53%

Vignettes (1)

MoEClust HTML

How these are measured

Downloads

5.3K

CRAN downloads in the past year

Rank #6,999 · ~14/day · ~439/mo

Releases R versions

Daily download trend is not available in this view yet.

54630 days

1.2K90 days

5.3K1 year

Compare downloads with other packages →

Also on123 r2u8 autocran167 c2d4u

How downloads are counted

Repository

keefe-murphy/moeclust ↗Account: Keefe-Murphy

7Stars

0Forks

0Open issues

0Open PRs

18Releases

115Commits

2Commit identities

model-based-clusteringgaussian-mixture-modelsmixture-of-experts

License file reads as GPL-3.0 · 115 commits · Last activity 2025-03-05 · 0% stars, 30d

Responsiveness

How these are sampled

PRs merged (all time): not measured
Median days to close, 50 most recently updated closed issues: 10
Median days to close, 50 most recently updated closed PRs: not measured

Stars over time

2023-02-13 · 62026-07-07 · 7

How these are measured

PracticesBeta

Checks run against github.com/keefe-murphy/moeclust on 2026-10-11.

No development-tooling practices detected in the upstream repository.

How this is detected·Detection ruleset v3 (2026-09-26)

Dependencies

Declared dependencies

9 external dependencies (excludes base and recommended)

Installing MoEClust adds 10 packages (8 with compiled code, 24 MB of source tarballs). Heaviest: BH 14 MB, mclust 2.7 MB, colorspace 2.0 MB.

Depends (1)

R >= 4.0.0

Imports (6)

latticematrixStats mclust mvnfastnnetvcd

LinkingTo (0)

none

Suggests (6)

clusterclustMD geometry knitr rmarkdown snow

Enhances (0)

none

Dependency license breakdown →

Reverse dependencies

1direct

0indirect

Imports 1

Depends, Imports or LinkingTo, most downloaded first

clustTMB

How these are counted

Code & Tests

Datasets

People & History

People (2)

Keefe Murphy

Author, Maintainer since 1.0.0 (2017)

Thomas Brendan Murphy

Contributor

Releases (18)

Pick two releases to compare their code metrics.

R
R 4.6.0 released · 2026-04-24
R
R 4.5.0 released · 2025-04-11
1.6.0Latest
record
2025-03-05 · current release · diff ↗
R
R 4.4.0 released · 2024-04-24
1.5.2
record
2023-12-11 · diff ↗ ·
added 1 export
FARI
R
R 4.3.0 released · 2023-04-21
1.5.1
record
2022-12-20 · diff ↗ ·
added 2 exports
MoE_AvePP, MoE_Similarity
R
R 4.2.0 released · 2022-04-22
1.5.0
record
2022-03-28 · diff ↗ ·
added 1 export
MoE_entropy
1.4.2
record
2021-12-20 · diff ↗
1.4.1
record
2021-10-12 · diff ↗
1.4.0
record
2021-06-22 · diff ↗ ·
removed 1 export
as.Mclust
R
R 4.1.0 released · 2021-05-18
1.3.3
record
2021-01-06 · diff ↗
1.3.2
record
2020-11-18 · diff ↗
1.3.1
record
2020-05-12 · diff ↗

Show 12 earlier eventsHide 12 earlier events

R
R 4.0.0 released · 2020-04-24
1.3.0
record
2020-03-31 · diff ↗
1.2.4
record
2019-12-11 · diff ↗
1.2.3
record
2019-07-31 · diff ↗ ·
added 1 export
MoE_stepwise
1.2.2
record
2019-05-15 · diff ↗
R
R 3.6.0 released · 2019-04-26
1.2.1
record
2018-12-11 · diff ↗ ·
added 1 export
MoE_cstep
1.2.0
record
2018-08-24 · diff ↗ ·
removed 2 exports
MoE_aitken, MoE_qclass

added 5 exports
MoE_Uncertainty, MoE_news, aitken, force_posiDiag, quant_clust
R
R 3.5.0 released · 2018-04-23
1.1.0
record
2018-02-06 · diff ↗ ·
added 1 export
expert_covar
1.0.0
record
2017-11-28
R
R 3.4.0 released · 2017-04-21

Package metadata

First published: 2017-11-28
Total releases: 18 / 9 yrs
License: GPL (>= 3) OSI
Minimum R: ≥ 4.0.0
Bundled data: 4.8 KB / 2 files
Download size: 1.4 MB

Appears in task views

Cluster Analysis & Finite Mixture Models

Cite

Cite this package

Run in R for the authors' preferred citation:

citation("MoEClust")

The package provides a citation file that builds its entry when R runs it. Run citation() to see it.

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 MoEClust version 1.6.0 [Data set]. HJJB, LLC. Data release v2026-10-18. https://doi.org/10.5281/zenodo.21843040

From data release v2026-10-18, 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 →