Package: genpca 0.2.2

Brad Buchsbaum

genpca: Generalized Principal Component Analysis

Generalized PCA and related matrix decompositions in weighted inner-product spaces. Methods are based on Allen, G. I., Grosenick, L., and Taylor, J. (2014) <doi:10.1080/01621459.2013.852978>, "A generalized least-square matrix decomposition", Journal of the American Statistical Association, 109(505), 145-159; and Abdi, H. (2007), "Singular value decomposition (SVD) and generalized singular value decomposition" <https://personal.utdallas.edu/~herve/Abdi-SVD2007-pretty.pdf>, in "Encyclopedia of Measurement and Statistics", 907-912.

Authors:Brad Buchsbaum [aut, cre, cph]

genpca_0.2.2.tar.gz
genpca_0.2.2.zip(r-4.7-x86_64)genpca_0.2.2.zip(r-4.7-arm64)genpca_0.2.2.zip(r-4.6-x86_64)genpca_0.2.2.zip(r-4.6-arm64)
genpca_0.2.2.tar.gz(r-4.7-arm64)genpca_0.2.2.tar.gz(r-4.7-x86_64)genpca_0.2.2.tar.gz(r-4.6-arm64)genpca_0.2.2.tar.gz(r-4.6-x86_64)
genpca_0.2.2.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
genpca/json (API)

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

Bug tracker:https://github.com/bbuchsbaum/genpca/issues

Pkgdown/docs site:https://bbuchsbaum.github.io

Uses libs:
  • openblas– Optimized BLAS
  • c++– GNU Standard C++ Library v3

On CRAN:

Conda:

openblascpp

4.12 score 22 scripts 339 downloads 16 exports 23 dependencies

Last updated from:a63dfc9ec7. Checks:10 OK. Indexed: no.

TargetResultTimeFilesSyslog
linux-devel-arm64OK193
linux-devel-x86_64OK259
source / vignettesOK259
linux-release-arm64OK190
linux-release-x86_64OK183
windows-devel-arm64OK280
windows-devel-x86_64OK222
windows-release-arm64OK287
windows-release-x86_64OK208
wasm-releaseOK111

Exports:componentsgeigen_covgenpcagenpca_covgenplsgenplscgmd_clear_cachegpca_mlegplssvd_opmnpca_mrlreconstructrepair_metricrplssfpcatransfertruncate

Dependencies:assertthatchkclicorpcordigesteigencoreFNNgeigengluelatticelifecyclemagrittrMatrixmultivariouspillarpkgconfigRcppRcppArmadilloRcppEigenrlangtibbleutf8vctrs

Generalized PLS-SVD: Explicit Whitening Reference
Quick practical use | Reference for contributors | Notation | Reference implementation | Cross-checking the operator | Where next
Rendered fromgplssvd-reference.Rmdusingknitr::rmarkdown

Last update: 2026-09-15
Started: 2026-09-15

Getting Started with genpca
Fit a weighted analysis | Use the fitted object | Add observation weights | Choose the number of components | Where next | The decomposition | Two metrics, one weighted objective | The problem it solves | The solution: whiten, decompose, unwhiten | What the metrics mean | Choosing structured metrics | Choosing a backend | What are ou, ov, u, and v? | Relation to the French school | References
Rendered fromgenpca.Rmdusingknitr::rmarkdown

Last update: 2026-09-15
Started: 2026-09-15

GPCA at Scale and Special Cases
Backend selection | Backends on the same problem | Sparse workflow (spectra) | What stays sparse | Covariance-only GPCA | Out-of-sample projection | Performance tips | Where next
Rendered fromgpca-scale.Rmdusingknitr::rmarkdown

Last update: 2026-09-15
Started: 2026-09-15

GPCA Metrics: Building M and A
Why metrics | Heteroscedastic diagonals | Inverse column variance is scaled PCA | Weighting both margins at once | Which way does a metric point? | Recipes | AR(1) row metric (whitens temporal autocorrelation) | Spatial RBF kernel (whitens smooth spatial noise) | Graph Laplacian (emphasises contrast across edges) | A note on sfpca() | Learning metrics with gpca_mle() | SPD requirements and remedies | Why the requirement exists | What happens when a metric falls short | Where next
Rendered fromgpca-metrics.Rmdusingknitr::rmarkdown

Last update: 2026-09-15
Started: 2026-09-15

Modelling Structured Noise
The problem | A metric is a spectral filter | Building these matrices in practice | What a metric can and cannot separate | Whiten by the noise, not by the signal | Compare bounded and linearly increasing weights | Use both margins | When one metric is not enough | Caveats worth carrying | Where next
Rendered fromstructured-noise.Rmdusingknitr::rmarkdown

Last update: 2026-09-15
Started: 2026-09-15

Sparse and Functional PCA with sfpca()
When to reach for it | A worked example | What the two penalties do | Choosing the penalties | Reading the output | Metric form versus constraint form | Practical notes | Where next | Reference
Rendered fromsfpca.Rmdusingknitr::rmarkdown

Last update: 2026-09-15
Started: 2026-09-15