R Observatory: Metrics for dgpsi version 2.6.0

Balamuta, James Joseph

doi:10.5281/zenodo.21843040

dgpsi

2.6.0Package record

Interface to 'dgpsi' for Deep and Linked Gaussian Process Emulations

0packages depend

2.9Kdownloads / year

0.1%test coverage

13/13flavors OK

Overview

About

Maintained by Deyu MingCRAN page ↗GitHub ↗Website ↗Issues ↗

Interface to the 'python' package 'dgpsi' for Gaussian process, deep Gaussian process, and linked deep Gaussian process emulations of computer models and networks using stochastic imputation (SI). The implementations follow Ming & Guillas (2021) doi:10.1137/20M1323771 and Ming, Williamson, & Guillas (2023) doi:10.1080/00401706.2022.2124311 and Ming & Williamson (2023) doi:10.48550/arXiv.2306.01212. To get started with the package, see https://mingdeyu.github.io/dgpsi-R/.

Changes in 2.6.0

Prediction speed with predict() enhanced for small testing data sets by reducing overhead caused by the multi-threading implementation.
The Python environment now installs packages exclusively from conda-forge whenever possible. Packages from other channels will only be used if they are unavailable on conda-forge.
A bug in vigf(), affecting a bundle of emulators that includes GP emulators, has now been fixed.
The column names from the training input and output provided to gp() and dgp() are retained in the relevant slots of the returned objects, as well as in any updated objects produced by the downstream functions that operate on them.
The column names from the testing input and output supplied to validate() and design() are retained in the relevant slots of the returned objects.
Improved numerical stability and achieved ~30x faster speed for DGP emulators using heteroskedastic likelihoods with replicates, with or without the Vecchia approximation.
Enhanced initialization of DGP emulators with heteroskedastic and categorical likelihoods for improved performance.
Removed the mode argument from predict() for DGP emulators with categorical likelihoods. Predictions of class probabilities can now be obtained using either the "mean_var" or "sampling" method.
Set the default method for predict(), validate(), and plot() to "mean_var".
Redesigned the output of validate() for dgp objects with likelihood = "Categorical". See validate() documentation for details.
Added support for the nugget_est argument in dgp() to control whether nuggets of GP nodes feeding into the likelihood node are estimated when likelihood is not NULL.
Updated initial nugget values when nugget_est = TRUE in dgp(). If likelihood = NULL, all initial GP nuggets default to 1e-6; otherwise, GP nodes feeding into the likelihood node default to 1e-4 and all others to 1e-6.

Show all changes in 2.6.0Hide the rest of the changes in 2.6.0

Added the accuracy metric to the figures produced by plot() for DGP emulators with categorical likelihoods.
Fixed the confusion matrix visualization (style = 2 in plot()) so that the diagonal is drawn from top-left to bottom-right.
Updated init_py() to handle errors related to TOS acceptance when installing Miniconda, and to automate TOS acceptance for required channels.
Enabled use of the newaccelerate BLAS library on Apple Silicon machines running macOS > 13.3.
Added the decouple argument to dgp() to allow likelihood parameters to be modeled using separate deep Gaussian process hierarchies when depth > 2.
Added the link argument to dgp() to support binary classification using either logit or probit link function when likelihood = "Categorical".
Inference for (D)GPs with homogeneous noise and replicates in the training data has been significantly enhanced, achieving over 10× speed-up.

From NEWS.md in the release

Install

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

install.packages("dgpsi", 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("mingdeyu/dgpsi-R")

Development version from mingdeyu/dgpsi-R. 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-dgpsi

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-dgpsi

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-dgpsi

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: 2.8 min · r-oldrel-windows-x86_64

Status changes

OK13 OK2026-08-05
NOTE12 OK, 1 NOTE2026-08-01
OK12 OK2026-04-25
ERROR13 OK, 1 ERROR2026-03-10

Advisories

none known

How this is measured·How advisories are found

Documentation

In release 2.6.0

READMEYes · 484 wordsVignettesYes · 1 · staticNEWSYes · 100% structured · mentions 2.6.0Citation fileYes · cites a publicationWebsiteYes · pkgdown

Help pages: 32 (1 internal)includes a package overview page
Examples that run: 0%10 never run · 17 hold no code
Exported functions with examples: 86%
Documented parameters: 100%
Return-value docs: 100%
Help pages citing references: 13%

Vignettes (1)

A Quick Guide to dgpsi HTML

How these are measured

Downloads

2.9K

CRAN downloads in the past year

Rank #13,421 · ~8/day · ~245/mo

Releases R versions

Daily download trend is not available in this view yet.

34330 days

86190 days

2.9K1 year

Compare downloads with other packages →

Also on89 r2u8 autocran

How downloads are counted

Repository

mingdeyu/dgpsi-r ↗Account: mingdeyu

1Stars

5Forks

2Open issues

1Open PRs

7Releases

321Commits

3Commit identities

deep-gaussian-processesemulationgaussian-processespackagersurrogate-models

321 commits · Last activity 2026-03-07 · 0% stars, 30d · Default branch at 2.6.0-9000, ahead of CRAN 2.6.0 (checked 2026-10-18)

Responsiveness

How these are sampled

PRs merged (all time): 80%
Median days to close, 50 most recently updated closed issues: 4
Median days to close, 50 most recently updated closed PRs: 0

Stars over time

2026-03-30 · 12026-07-07 · 1

How these are measured

PracticesBeta

7 development-tooling and community-health practices detected across 4 families in the upstream repository

Checks run against github.com/mingdeyu/dgpsi-r on 2026-10-18.

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 (3)

pkgdown

GitHub Pages site last deployed 2026-03-07 · The repository has a site deploy workflow

pkgdown site last built 2025-10-17

CRAN release process (1)

cran-comments.md

Git structural (1)

.gitattributes

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

Dependencies

Declared dependencies

20 external dependencies (excludes base and recommended)

Installing dgpsi adds 95 packages (58 with compiled code, 87 MB of source tarballs). Heaviest: stringi 11 MB, sf 9.3 MB, ggplot2 6.0 MB.

Depends (1)

R >= 4.0

Imports (20)

reticulate benchmarkmeutilsggplot2 ggforce reshape2 patchwork lhsmethodsstatsclhs dplyr uuid tidyr rlang lifecycle magrittr visNetworkparallelkableExtra

LinkingTo (0)

none

Suggests (5)

knitr rmarkdownMASSR.utils spelling

Enhances (0)

none

Dependency license breakdown →

Reverse dependencies

0direct

0indirect

Nothing depends on this yet.

How these are counted

Code & Tests

People & History

People (2)

Deyu Ming

Daniel Williamson

Author

Releases (7)

Pick two releases to compare their code metrics.

R
R 4.6.0 released · 2026-04-24
2.6.0Latest
record
2025-10-15 · current release · diff ↗ ·
removed 2 exports
combine, set_linked_idx
R
R 4.5.0 released · 2025-04-11
2.5.0
record
2024-12-15 · diff ↗ ·
removed 5 exports
Hetero, NegBin, Poisson, kernel, pei

added 6 exports
deserialize, get_thread_num, serialize, set_id, set_thread_num, set_vecchia
R
R 4.4.0 released · 2024-04-24
2.4.0
record
2024-01-14 · diff ↗ ·
added 1 export
prune
2.3.0
record
2023-09-03 · diff ↗ ·
added 1 export
vigf
2.2.0
record
2023-06-05 · diff ↗ ·
added 1 export
set_seed
R
R 4.3.0 released · 2023-04-21
2.1.6
record
2023-02-08 · diff ↗ ·
added 10 exports
alm, design, draw, mice, pack, pei, set_imp, unpack, update, window
2.1.5
record
2022-09-29
R
R 4.2.0 released · 2022-04-22

Package metadata

First published: 2022-09-29
Total releases: 7 / 4 yrs
License: MIT + file LICENSE OSI
Minimum R: ≥ 4.0
Download size: 703 KB

Cite

Cite this package

Run in R for the authors' entry:

citation("dgpsi")

The package provides a citation file, so citation() returns the authors' own entry.

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 dgpsi version 2.6.0 [Data set]. HJJB, LLC. Data release v2026-10-22. https://doi.org/10.5281/zenodo.21843040

From data release v2026-10-22, 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 →