Package: T4transport 0.2.0.9002

T4transport: Tools for Computational Optimal Transport

Computational optimal transport for weighted finite measures, Gaussian distributions, and finite dissimilarity spaces. Provides exact and regularized transport, sliced and alignment-based comparisons, Gromov-Wasserstein calculations, barycenters, medians, and interpolation. Explicit representations and result inspection distinguish mathematical targets, numerical termination, and approximation. See Peyre and Cuturi (2019) <doi:10.1561/2200000073> for a general exposition.

Authors:Kisung You [aut, cre]

T4transport_0.2.0.9002.tar.gz
T4transport_0.2.0.9002.zip(r-4.7-x86_64)T4transport_0.2.0.9002.zip(r-4.7-arm64)T4transport_0.2.0.9002.zip(r-4.6-x86_64)T4transport_0.2.0.9002.zip(r-4.6-arm64)T4transport_0.2.0.9002.zip(r-4.5-x86_64)
T4transport_0.2.0.9002.tgz(r-4.6-x86_64)T4transport_0.2.0.9002.tgz(r-4.6-arm64)T4transport_0.2.0.9002.tgz(r-4.5-x86_64)T4transport_0.2.0.9002.tgz(r-4.5-arm64)
T4transport_0.2.0.9002.tar.gz(r-4.7-arm64)T4transport_0.2.0.9002.tar.gz(r-4.7-x86_64)T4transport_0.2.0.9002.tar.gz(r-4.6-arm64)T4transport_0.2.0.9002.tar.gz(r-4.6-x86_64)
T4transport_0.2.0.9002.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
T4transport/json (API)

# Install 'T4transport' in R:
install.packages('T4transport', repos = c('https://kisungyou.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/kisungyou/t4transport/issues

Pkgdown/docs site:https://www.kisungyou.com

Uses libs:
  • openblas– Optimized BLAS
  • c++– GNU Standard C++ Library v3
  • openmp– GCC OpenMP (GOMP) support library
Datasets:
  • digit3 - MNIST Images of Digit 3
  • digits - MNIST Images of All Digits

On CRAN:

Conda:

openblascppopenmp

7.08 score 7 stars 1 packages 19 scripts 383 downloads 53 exports 4 dependencies

Last updated from:1d56c5a0a4. Checks:15 OK. Indexed: yes.

TargetResultTimeFilesSyslog
linux-devel-arm64OK180
linux-devel-x86_64OK208
source / vignettesOK415
linux-release-arm64OK202
linux-release-x86_64OK236
macos-release-arm64OK212
macos-release-x86_64OK290
macos-oldrel-arm64OK145
macos-oldrel-x86_64OK460
windows-devel-arm64OK413
windows-devel-x86_64OK284
windows-release-arm64OK297
windows-release-x86_64OK248
windows-oldrel-x86_64OK255
wasm-releaseOK142

Exports:as_ot_measureecdfbaryecdfmedfbary14Cfbary14Cdistfbary15Bfbary15Bdistfiedlergaussbary1dgaussbarypdgaussmed1dgaussmedpdgaussvis2dgwbarygwdisthistbaryhistbary14Chistbary15Bhistdisthistinterphistmedimagebaryimagebary14Cimagebary15Bimagedistimageinterpimagemedimg2measureipotipotDot_collectionot_costot_densityot_distanceot_gaussianot_gaussian_distanceot_gaussian_summaryot_imageot_inspectot_measureot_summarypwbarypwdistrbary23LrbaryGDrmedIRLSrmedWBsinkhornsinkhornDswdistwassbootwassersteinwassersteinD

Dependencies:rbibutilsRcppRcppArmadilloRdpack

Choosing a transport geometry and inspecting results
Represent observations and choose their geometry | Alignment and relational geometry | Gaussian parameters and distribution summaries | Read the available evidence
Rendered fromtoolbox-workflow.Rmdusingknitr::rmarkdown

Last update: 2026-10-02
Started: 2026-10-02

Computing the Wasserstein distance
Transport cost and Wasserstein distance | A small two-dimensional example | Supply cross-distances explicitly | Numerical scale and further workflows | References
Rendered frombasic_compute_distance.Rmdusingknitr::rmarkdown

Last update: 2026-10-02
Started: 2025-07-19

Distribution representations and transport objectives
One observation is a probability distribution | A cost and a distance have different units | Entropy regularization changes the objective | Scalar barycenters and medians | Choose a conversion deliberately | Work directly with Gaussian parameters
Rendered fromdistribution-workflow.Rmdusingknitr::rmarkdown

Last update: 2026-10-02
Started: 2026-10-02

Migrating analyses to the 0.2.0 development interface
Numerical audit repairs in 0.2.0.9002 | Additive toolbox interfaces in 0.2.0.9001 | Recheck the target and numerical outcome | ECDF summaries preserve weighted finite distributions | Histogram displays and estimates are separate | Historical regularized names now share one checked implementation | Image methods have explicit grids and projection error | Gaussian means, variances, and covariance domains
Rendered frommigration.Rmdusingknitr::rmarkdown

Last update: 2026-10-02
Started: 2026-10-02

Ordinary and robust summaries with explicit numerical status
Compare targets on the same observations | State what is allowed to change | Initialization and support-size sensitivity | Read failures and stopping criteria | Fixed-grid summaries
Rendered fromrobust-summary-workflow.Rmdusingknitr::rmarkdown

Last update: 2026-10-02
Started: 2026-10-02

Transport plans and image interpolation
Rendered fromtransport-interpolation.Rmdusingknitr::rmarkdown

Last update: 2026-10-02
Started: 2026-10-02

Readme and manuals

Help Manual

Help pageTopics
MNIST Images of Digit 3digit3
MNIST Images of All Digitsdigits
Barycenter of Empirical CDFsecdfbary
Wasserstein Median of Empirical CDFsecdfmed
Fixed-Support Barycenter by Cuturi & Doucet (2014)fbary14C fbary14Cdist
Fixed-Support Barycenter by Benamou et al. (2015)fbary15B fbary15Bdist
Compute the fiedler vector of a point cloudfiedler
Barycenter of Gaussian Distributions in \mathbb{R}gaussbary1d
Barycenter of Gaussian Distributions in \mathbb{R}^pgaussbarypd
Wasserstein Median of Gaussian Distributions in \mathbb{R}gaussmed1d
Wasserstein Median of Gaussian Distributions in \mathbb{R}^pgaussmedpd
Draw a Bivariate Gaussian Covariance Ellipsegaussvis2d
Gromov-Wasserstein Barycentergwbary
Gromov-Wasserstein Distancegwdist
Barycenter of Histogramshistbary
Barycenter of Histograms by Cuturi and Doucet (2014)histbary14C
Barycenter of Histograms by Benamou et al. (2015)histbary15B
Distance between Histogramshistdist
Interpolation between Histogramshistinterp
Wasserstein Median of Histogramshistmed
Barycenter of Imagesimagebary
Barycenter of Images according to Cuturi & Doucet (2014)imagebary14C
Barycenter of Images according to Benamou et al. (2015)imagebary15B
Wasserstein Distance between Two Imagesimagedist
Interpolation between Imagesimageinterp
Median on a Prescribed Image Gridimagemed
Extract a discrete measure from a gray-scale image matriximg2measure
Wasserstein Distance via Inexact Proximal Point Methodipot ipotD
Declare supplied transport costs or ground distancesot_cost
Compare finite probability measuresot_distance
Parametric Gaussian probability measuresot_gaussian
Exact Wasserstein-2 distance between Gaussian parameter objectsot_gaussian_distance
Summaries of Gaussian parameter objectsot_gaussian_summary
Explicit image and density conversionsot_density ot_image
Inspect a transport computation without recomputing itot_inspect
Finite probability measures and collectionsas_ot_measure ot_collection ot_measure
Barycenters and medians of finite probability measuresot_summary
Procrustes-Wasserstein Barycenterpwbary
Procrustes-Wasserstein Distancepwdist
Generalized inverse of a T4transport weighted CDFquantile.t4transport_cdf
Free-Support Barycenter by von Lindheim (2023)rbary23L
Free-Support Wasserstein Barycenter by Barycentric-Projection UpdatesrbaryGD
Free-Support Median by IRLSrmedIRLS
Free-Support Median by Weiszfeld Update with Barycentric ProjectionrmedWB
Wasserstein Distance via Entropic Regularization and Sinkhorn Algorithmsinkhorn sinkhornD
Sliced Wasserstein Distanceswdist
Wasserstein Distance Estimation with Boostrappingwassboot
Wasserstein Distance via Linear Programmingwasserstein wassersteinD