R Observatory: Metrics for babelmixr2 version 0.1.12

Balamuta, James Joseph

doi:10.5281/zenodo.21843040

CRAN requires the check problems on this package to be fixed before Oct 23 or it will be archived. See the check details

babelmixr2 needs 3 packages that CRAN will archive unless their check problems are fixed, the earliest by Oct 18. See which.

babelmixr2

0.1.12

Use 'nlmixr2' to Interact with Open Source and Commercial Software

0packages dependsuggested by 4

7.1Kdownloads / year

–test coverage

10/13flavors OK

Overview

About

Maintained by Matthew FidlerCRAN page ↗GitHub ↗Website ↗

Run other estimation and simulation software via the 'nlmixr2' (Fidler et al (2019) doi:10.1002/psp4.12445) interface including 'PKNCA', 'NONMEM' and 'Monolix'. It also converts 'nlmixr2'/'rxode2' models to 'PopED' optimal design databases (Fidler et al (2024) doi:10.70534/XUMG6226). While not required, you can get/install the 'lixoftConnectors' package in the 'Monolix' installation, as described at the following url https://monolixsuite.slp-software.com/r-functions/2024R1/installation-and-initialization. When 'lixoftConnectors' is available, 'Monolix' can be run directly instead of setting up command line usage.

Changes in 0.1.12

est="pknca" unit conversion scales the whole DV expression, so an additive term (like a baseline) is converted too (#236)
NONMEM: names that differ only by case get one NONMEM name each for the whole control stream, so the data columns, $INPUT, $MODEL, $PK, $DES and $ERROR agree. Covariates ignore case (like rxode2), so wt and WT are one data column; a variable or compartment that clashes is renamed RXC001, ... (#237)
Monolix: a modeled rate() with f() infuses over f*amt/rate, the rxode2 (and NONMEM) duration, instead of amt/rate (#239)
as.nlmixr2() of a nonmem2rx import solves with rxControl(nonmem = TRUE, addlKeepsCov = FALSE), the way nonmem2rx validates the import, when rxode2 supports it (nlmixr2/rxode2#1429) and nlmixr2est keeps it through its fits (nlmixr2/nlmixr2est#1167). Statements that do not depend on a state (NONMEM's $PK) then read TIME as the time of the record ending each interval, like NONMEM, instead of the continuous time; for TIME in $PK the IPRED difference to NONMEM goes from 21.8% to 0.004% (#252). Models with delay() keep the continuous time.

Show all changes in 0.1.12Hide the rest of the changes in 0.1.12

An empty Monolix administration table builds its type factor with factor(levels=) instead of structure(.Label=), which R-devel's R CMD check flags as deprecated.
Monolix imports (as.nlmixr2() of a monolix2rx model) no longer fail when the project's steady-state dose count (nbdoses) is below 6: minSS/maxSS come from monolix2rx's .getSsLimits(), raised to rxode2's floor of 5 and 7.
simplifyUnit() and modelUnitConversion() work without the suggested packages testthat and units installed: simplifyUnit() checked its arguments with checkmate's expect_*() (testthat expectations) instead of assert_*(), and both now ask for units with rxode2::rxReq(). Their examples run only when units is installed, and the PKNCA and saemix tests skip when those packages are not installed (CRAN's noSuggests check).
Monolix (fourth stress kit run): EQUATION: only has assignments and conditions, so PK macros there were rejected. The model lines before the ODEs (with the compartment properties the macros use) are now written in the PK: block before the macros, and the ODEs stay in EQUATION:. Compartment properties set after the ODEs are moved before them, and a property set in an if is set in every branch (rxode2's default in the others) instead of from a default assigned before the if, which Monolix cannot reassign. A missing LIMIT is written as missing instead of -Inf.
Monolix projects Monolix could not load (third stress kit run, with Monolix's reasons):
"Conflicting variable definition": Monolix assigns a variable once, so a reassigned variable (like if (SEX == 1) cl <- cl * 1.2, or the pruned else if) gets a new name (cl_rx1); an if/else that changes a defined variable is pruned first.
"Undefined variable": compartment properties set in the model (f(), alag(), dur(), rate()) are defined in EQUATION:, so the PK macros using them are now written in EQUATION: after them (not in a PK: block before). A modeled rate uses Tk0=rx_tk0_<state> with rx_tk0_<state> = amtDose/rate (macro arguments cannot be calculations).
monolixControl(stiff=TRUE) wrote odeType = stiff where Monolix expects an input definition; it is now in EQUATION:.
A LIMIT on uncensored observations (M2) is refused: Monolix only uses LIMIT with censored values.
Monolix fits use the number of observations Monolix used for the objective function, like NONMEM fits; it was off for data with evid=2 records or time not starting at zero.
Monolix: a mu-referenced parameter that stays inside exp() in the model (like cl <- exp(tcl + eta.cl) * (CRCL/100)^cl.crcl, or f(depot) <- exp(lfdepot)) was given a log-normal distribution while the model still took exp() of it, so exp() was applied twice. It is now a normal parameter on nlmixr2's scale (its initial value, estimate and covariance too).
When lixoftConnectors cannot load or run a Monolix project, the error now includes Monolix's reason (its [ERROR] lines) instead of "see Monolix's [ERROR] above".
Fixes found by running the NONMEM/Monolix stress kit with NONMEM 7.4:
nonmemControl(est="imp") and est="its" fits were always reported as unsuccessful; NONMEM ends them with "OPTIMIZATION WAS NOT TESTED FOR CONVERGENCE", which is now read as finished.
nonmemControl(est="posthoc") fits could not be read (no #TERM: block and no parameter history).
Box-Cox, Yeo-Johnson and logit + Yeo-Johnson residual errors were rejected by NM-TRAN (IPRED redefined in nested/ELSE IF structures); the transformations are now written without nested IFs. They are still transform-both-sides (DV in CCONTR, IPRED in $ERROR). Predictions at or below sqrt(DBL_EPSILON) are floored like rxode2 instead of being set to -1000000000 (which gave lnorm() fits a huge objective function), and Box-Cox no longer divides by zero when lambda is 0.
The IPRED/PRED comparison with NONMEM now undoes the residual transformation first (NONMEM's predictions are on the transformed scale).
Censored (CENS, CENS + LIMIT, LIMIT only) models were rejected by NM-TRAN ("random variable is defined in a nested IF structure").
A $PK parameter changed later in the model (like if (SEX == 1) cl <- cl * 1.2) was changed in $DES, which NONMEM does not allow, and used undefined in $ERROR; it is now defined as RXPK_<name> in $PK.
Models without random effects are refused (NONMEM treats them as single-subject data, where METHOD=COND is invalid).
Monolix fixes found by running the NONMEM/Monolix stress kit:
Infusions given by rate (RATE) or duration (TINF) were written to the data but not declared in [CONTENT], so Monolix gave them as bolus doses.
Endpoints with dotted names (like concentration.in.plasma) kept the dots in the Monolix project (OUTPUT, observation names, predictions), which Monolix cannot load.
Reading a fit with a fixed parameter (like tfdepot <- fix(0.8)) failed with "subscript out of bounds"; Monolix's covariance has only the estimated parameters.
A monolixControl(runCommand=) command that finished without writing Monolix's output now stops with an error instead of waiting for it forever.
nonmemControl(cov="") now skips the covariance step as documented; before it gave an error because match.arg() cannot match "".
The NONMEM/Monolix stress test (inst/stress) is now a kit to run on a machine with NONMEM and/or Monolix: install-kit.R installs the versions to test, run-stress.R --check finds NONMEM/Monolix, and run-stress.R --kit runs every case end to end, checks that rxode2 reproduces the NONMEM/Monolix predictions and that a second fit reads the saved output, and zips the results to send back. New cases cover missing observations, evid=2, character IDs, time not starting at zero, extra data columns, a single subject, oral plus iv doses, ODE infusions (rate, steady state, modeled duration), two endpoints, bounded thetas, several covariates, models without random effects and the NONMEM/Monolix estimation options.
NONMEM control streams now use $ABBR PROTECT so NM-TRAN replaces LOG, EXP, SQRT, division and powers with NONMEM's protected functions (NONMEM 7.4 or later). Turn it off with nonmemControl(protect=FALSE) or options(babelmixr2.nmProtect=FALSE). Since NM-TRAN writes B**E as PEXP(E*PLOG(B)), integer powers are written as products (or powers of x*x, which is never negative) and powers of a positive number with DEXP(). Functions of numbers (like exp(0), log(2*pi) or expit(0)) and divisions of numbers are written as numbers, and model variables named like a protected function (like plog) are renamed. babelmixr2's own zero protection (protectZeros) is now only used with protect=FALSE, like for NONMEM before 7.4 (#62).
NONMEM control streams now explain the zero-protection code babelmixr2 adds: each RXDZ### IF block is preceded by a comment saying what it keeps the variable away from and why, as is the IF (W1 .EQ. 0.0) residual variance protection (#91). The protection for lfactorial()/lgamma1p() now keeps its argument above -1 (it previously clamped it just below -1, making x+1 negative), log(x) and 1/x no longer share one protected variable, so 1/x keeps the sign of a negative x, and a variable reassigned in the model is protected again instead of reusing the protection of its old value. Zero protection needed by f(), alag(), rate() or dur() (written in $PK) is no longer shared with other lines, which could use it before (or without) it being calculated.
est="pknca" now works with covariates in the data and with a mix of intravascular and extravascular doses (#102). Doses into a compartment that the observations are calculated from are intravascular. With both routes, ka is estimated from the extravascular doses and vc and cl from the intravascular doses. Intravascular bolus doses have the concentration at the time of dosing back-extrapolated (replacing a predose concentration at the first dose), and other doses have it imputed (as the predose concentration, or zero for the first dose). When no doses are only extravascular, ka is not updated. Multiple-dose data no longer need a concentration at each dose time; each dose until the next (with at least 2 concentrations) is used, with vc from the first dose of each route and cl from dosing intervals mostly covered by concentrations. Doses at the same time are combined.
When a NONMEM run fails, est="nonmem" now says why and where to look instead of failing with an unclear error: a run command that was not found or wrote no output, a NONMEM license problem, an NM-TRAN error in the control stream or the data, NONMEM not starting (for example a compiler problem), NONMEM crashing or stopping during estimation, and output babelmixr2 cannot read (#46).
est="pknca" now updates the initial estimates of models that are not mu-referenced, like ka <- tka * exp(eta.ka), instead of failing with "Must have names" (#101). When there is no vc, a central volume named v, V, Vc, VC, v1 or V1 (only one of them) now gets the NCA central volume estimate. Parameters defined as expit() of a theta are now transformed back correctly. A message lists the parameters that could not be updated.
NONMEM models can now use nested if/else if/else statements (and ifelse()): their branches are pruned with rxode2's branch pruning before the model is translated to NONMEM (#11). With the default nonmemControl(prune="auto") a model whose if blocks are simple is still written with NONMEM IF blocks and only a model that needs it is pruned; prune=TRUE always prunes and prune=FALSE never prunes (the error for unsupported if/else statements suggests prune=TRUE).
monolixControl(prune=) has the same option for Monolix. Monolix writes if/elseif/else (and nested if) statements directly, so prune="auto" (the default) only prunes a model that uses ifelse(), which Monolix cannot write; prune=TRUE always prunes and prune=FALSE never prunes. A logical expression used as a number in a Monolix model is written as a 0/1 indicator variable (#11).
A logical expression used as a number in a NONMEM model (like cl <- tcl * (WT > 70)) is now written as a 0/1 indicator variable, since NONMEM cannot use a logical expression as a number. A numeric if () condition is written as not equal to zero.
est="nonmem" and est="monolix" fits with table=tableControl(cwres=TRUE) no longer fail with "objective function 'FOCEi' already present". The fit keeps both the NONMEM (or Monolix) objective, which stays in use, and nlmixr2's FOCEi objective (#94). as.nlmixr2() of a nonmem2rx or monolix2rx model now also keeps the imported objective in use with cwres=TRUE, and no longer fails after an earlier import with cwres=TRUE (it started from that fit's etas).
est="monolix" now translates compartment properties (f(), alag(), rate() and dur()) that are expressions, like f(depot) <- exp(lfdepot), instead of erroring with "the complex F is not supported by babelmixr2" (#115). The expression is calculated in a new variable (like rx_f_depot) that the PK: macro uses. A property set only inside an if now keeps rxode2's default otherwise (1 for f(), 0 for alag()) instead of being applied unconditionally.
The ACoP 2024 babelmixr2/PopED abstract (doi:10.70534/XUMG6226) is now in citation("babelmixr2"), the package description and the PopED article (#155).
as.nlmixr2() of a nonmem2rx model now stops with an informative error when the model still contains untranslated NONMEM residual variables (eps# or err#) instead of failing inside the estimation routine (#95).
popedControl(sigdig=) now sets the ODE solver tolerances itself (atol = rtol = 0.5*10^(-sigdig-2), the same values as before) instead of using rxode2::rxControl(sigdig=). rxode2 5.1.5 loosened the tolerances that rxControl(sigdig=) gives, which made the finite-difference FIM, and so the design OFV and RSEs, less accurate (#223).
est="nonmem" now fits censored data the way nlmixr2 does (#92). M3 (CENS), M4 (CENS with a finite LIMIT) and M2 (CENS=0 with a finite LIMIT, including data with a LIMIT but no CENS column) use F_FLAG likelihoods in $ERROR with LAPLACIAN estimation. A missing LIMIT is now written as NONMEM's infinity instead of 0, CENS/LIMIT columns that do not censor anything are dropped, the objective function is adjusted so the log-likelihood includes the censored observations correctly, and censoring with a transformed endpoint (like lnorm()) is refused instead of giving the wrong likelihood. The censored observations are left out of the NONMEM PRED comparison since NONMEM's PRED is their likelihood.
est="nonmem" and est="monolix" now fit linCmt() models. A pure linCmt() model uses NONMEM's closed-form solutions (ADVAN1-ADVAN4, ADVAN11 or ADVAN12 with TRANS1 micro-constants) or Monolix's pkmodel(); a model the closed form cannot represent (other ODEs, parameters that change with time, and for Monolix modeled rates or durations, doses into more than one compartment or amounts used in the model) is translated to ODEs with rxode2::linToOde(). nonmemControl(linCmt="ode") and monolixControl(linCmt="ode") always use ODEs. The closed-form solutions need an rxode2 with linCmtMicro(); with an older rxode2, linCmt() models use ODEs.
est="poped" now designs linCmt() models (translated to ODEs). est="nlmer", est="fmeMcmc" and est="pseudoOptim" are now tested with linCmt() models.
New NONMEM/Monolix stress test (inst/stress/), shared by the package tests and a command line runner (run-stress.R). The runner can also fit every case with NONMEM and/or Monolix end to end on a machine that has them, and optionally translate the nlmixr2lib models; see inst/stress/README.md.
The NONMEM data dropped the RATE, SS and II items: which items were written was decided from settings made only after the data were converted, so infusions, steady state doses and modeled rates or durations were written as plain bolus doses. The items are now kept when the data use them.
A steady state dose with a lag time (which rxode2 splits in two) is now written as one SS dose for NONMEM and Monolix, not as two bolus doses.
NONMEM now writes a modeled duration as Dn (it was DURn, which NONMEM does not know) and a modeled rate as Rn (it was dropped).
NONMEM now uses the NONMEM name of a mu-reference

The NEWS section is longer than shown here.

From NEWS.md in the release

Install

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

install.packages("babelmixr2", 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("nlmixr2/babelmixr2")

Development version from nlmixr2/babelmixr2. 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-babelmixr2

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-babelmixr2

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

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

Health

CRAN checks

3NOTE10OK

Check messages

NOTER code for possible problemsr-devel-linux-x86_64-debian-clang, r-devel-linux-x86_64-debian-gcc, r-devel-windows-x86_64

Found calls to structure() using deprecated special names:
  babelmixr2/R/monolixNlmixr2est.R (.Label: 1)
'.Label' should be changed to 'levels'.

CRAN check page ↗

Slowest check: 18.5 min · r-oldrel-windows-x86_64

Status changes

NOTE10 OK, 3 NOTENow
NOTE8 OK, 5 NOTE2026-10-14
ERROR8 OK, 4 NOTE, 1 ERROR2026-09-15
NOTE8 OK, 5 NOTE2026-08-12
ERROR6 OK, 5 NOTE, 2 ERROR2026-08-04
NOTE11 OK, 2 NOTE2026-07-16

Show 6 earlier changesHide 6 earlier changes

OK13 OK2026-06-09
ERROR12 OK, 1 ERROR2026-06-08
OK12 OK2026-04-25
NOTE11 OK, 3 NOTE2026-04-22
ERROR10 OK, 3 NOTE, 1 ERROR2026-04-18
NOTE11 OK, 3 NOTE2026-03-10

Advisories

none known

How this is measured·How advisories are found

Documentation

In release 0.1.12

READMEYes · 260 wordsVignettesNoNEWSYes · 100% structured · mentions 0.1.12Citation fileYes · cites a publicationWebsiteYes · pkgdown

The repository also keeps web-only articles.

Help pages: 49 (14 internal)
Examples that run: 50%7 run in a plain check (3 of them under a condition) · 7 only when donttest examples are enabled (R CMD check --as-cran)
Exported functions with examples: 42%
Documented parameters: 91%
Return-value docs: 97%
Help pages citing references: 4%

How these are measured

Downloads

7.1K

CRAN downloads in the past year

Rank #4,197 · ~19/day · ~588/mo

Releases R versions

Daily download trend is not available in this view yet.

73430 days

2.1K90 days

7.1K1 year

Compare downloads with other packages →

Also on468 r2u35 c2d4u

How downloads are counted

Repository

nlmixr2/babelmixr2 ↗Organization: nlmixr2

16Stars

2Forks

10Open issues

2Open PRs

12Releases

1,239Commits

5Commit identities

nonmempharmacometricsrmonolix

1,239 commits · Last activity 2026-10-16 · 0% stars, 30d

Responsiveness

How these are sampled

PRs merged (all time): 98%
Median days to close, 50 most recently updated closed issues: 42
Median days to close, 50 most recently updated closed PRs: 0

Stars over time

2025-03-21 · 92026-08-31 · 16

How these are measured

PracticesBeta

10 development-tooling and community-health practices detected across 5 families in the upstream repository

Checks run against github.com/nlmixr2/babelmixr2 on 2026-10-11.

Continuous integration (2)

GitHub ActionsR CMD check workflow

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

Docs source (4)

README.Rmdpkgdown

GitHub Pages site last deployed 2026-10-09 · The repository has a site deploy workflow

pkgdown site last built 2026-10-09, before release 0.1.12 (2026-10-16)

Coverage (1)

Coverage reporting (Codecov)

CRAN release process (2)

cran-comments.mdCRAN-SUBMISSION

Show all practicesHide 1 practice

Lint, format, editor (1)

Air

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

AI coding toolsBeta

This repository shows signs of AI coding tools.

Claude Codefirst seen 2026-04-02
Pull request it wrote: opened by a project member
Commits: at least 38 crediting it in the message

From outside contributors

Claude Code
Pull requests: 1 from outside contributors, first on 2026-06-18

Models named in its commits

Claude Opus 4.814 commits · Jun to Jul 2026
Claude Fable 58 commits · Jul 2026
Claude Sonnet 4.64 commits · Apr to Jun 2026
Claude Opus 4.72 commits · Jun 2026
Claude Opus 51 commit · Jul 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

35 external dependencies (excludes base and recommended)

Installing babelmixr2 adds 140 packages (90 with compiled code, 114 MB of source tarballs). Heaviest: BH 14 MB, stringi 11 MB, ggplot2 6.0 MB.

Depends (1)

R >= 3.5

Imports (14)

checkmate cli digest lotri nlmixr2data nlmixr2extra nlmixr2plot magrittr nlmixr2est nonmem2rx (>= 0.1.5)monolix2rxmethodsrex rxode2

LinkingTo (4)

Rcpp rxode2 RcppArmadillo RcppEigen

Suggests (19)

qs2 testthat withr lixoftConnectors PKNCA rmarkdown spelling PopED unitsnlmedplyr devtools memoise FME coda crayon saemix lme4 nlmixr2lib

Enhances (0)

none

Dependencies facing archival

CRAN will archive these unless their check problems are fixed by the date shown. This package needs each one to install.

nonmem2rxfix by Oct 18
nlmixr2plotfix by Oct 22
nlmixr2estfix by Oct 23

Dependency license breakdown →

Reverse dependencies

0direct

0indirect

Suggested by 4

Suggests, most downloaded first

nlmixr2 nlmixr2save ruminate ubiquity

How these are counted

Code & Tests

People & History

People (4)

Matthew Fidler

Author, Maintainer since 0.1.0 (2022)

Bill Denney

Author

Nook Fulloption

Contributor · added in 0.1.1

goldfish art

Theodoros Papathanasiou

Contributor · added in 0.1.6

Releases (12)

Pick two releases to compare their code metrics.

0.1.12Latest
record
2026-10-16 · current release · diff ↗ ·
added 6 exports
.nlmerCurrentFormula, .nlmerNlmerFun, iterPrintControl, nlmerControl, nlmixr2NlmerControl, saemixControl
·
listed once in CRAN's queue
Seen in one reading of CRAN's incoming queue, at 2026-10-16 17:07 UTC, so it stayed less than the time between readings.
R
R 4.6.0 released · 2026-04-24
0.1.11
record
2025-11-28 · diff ↗
0.1.10
record
2025-10-14 · diff ↗
0.1.9
record
2025-08-31 · diff ↗ ·
added 4 exports
.fmeMcmcF, .pseudoOptimF, fmeMcmcControl, pseudoOptimControl
0.1.8
record
2025-07-15 · diff ↗ ·
added 97 exports
%>%, ini<-, model<-, .nlmixrNlmeFun, ACF, RxODE, VarCorr, add.dosing, add.sampling, addCwres, addNpde, addTable, augPred, bobyqaControl, bootplot, bootstrapFit, et, etExpand, eventTable, expit, fixed.effects, fixef, foceiControl, geom_amt, geom_cens, getData, getVarCov, groupedData, ini, lbfgsb3cControl and 67 more
0.1.7
record
2025-05-07 · diff ↗
R
R 4.5.0 released · 2025-04-11
0.1.6
record
2025-01-20 · diff ↗
0.1.5
record
2024-11-06 · diff ↗ ·
removed 1 export
.popedSetup

added 2 exports
.popedCluster, babelBpopIdx
0.1.4
record
2024-09-23 · diff ↗ ·
added 20 exports
.popedF, .popedFree, .popedRxRunSetup, .popedSetup, .popedSolveIdME, .popedSolveIdME2, .popedW, babel.poped.database, mlxtran, monolix2rx, nmcov, nmext, nminfo, nmtab, nmxml, popedControl, popedGetMultipleEndpointModelingTimes, popedMultipleEndpointIndexDataFrame, popedMultipleEndpointParam, popedMultipleEndpointResetTimeIndex
R
R 4.4.0 released · 2024-04-24
0.1.2
record
2023-12-12 · diff ↗
0.1.1
record
2023-05-28 · diff ↗ ·
added 4 exports
as.nlmixr, as.nlmixr2, as.nonmem2rx, nonmem2rx
R
R 4.3.0 released · 2023-04-21
0.1.0
record
2022-10-28

Show 1 earlier eventHide 1 earlier event

R
R 4.2.0 released · 2022-04-22

Across CRAN, the median accepted release of the past year spent 19 hours in the queue, 90% under 11 days.How queue time is measured

Package metadata

First published: 2022-10-28
Total releases: 12 / 4 yrs
License: GPL (>= 3) OSI
Minimum R: ≥ 3.5
Download size: 4.2 MB

Appears in task views

Mixed, Multilevel, and Hierarchical Models in R

Cite

Cite this package

Run in R for the authors' preferred citation:

citation("babelmixr2")

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 babelmixr2 version 0.1.12 [Data set]. HJJB, LLC. Data release v2026-10-17. https://doi.org/10.5281/zenodo.21843040

From data release v2026-10-17, 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 →