Package: bcv 1.0.2

Michal Burdukiewicz

bcv: Cross-Validation for the SVD (Bi-Cross-Validation)

Methods for choosing the rank of an SVD (singular value decomposition) approximation via cross validation. The package provides both Gabriel-style "block" holdouts and Wold-style "speckled" holdouts. It also includes an implementation of the SVDImpute algorithm. For more information about Bi-cross-validation, see Owen & Perry's 2009 AoAS article (at <arxiv:0908.2062>) and Perry's 2009 PhD thesis (at <arxiv:0909.3052>).

Authors:Michal Burdukiewicz [cre, ctb], Patrick O. Perry [aut, cph], Seyed Morteza Najibi [ctb]

bcv_1.0.2.tar.gz
bcv_1.0.2.tar.gz(r-4.5-noble)bcv_1.0.2.tar.gz(r-4.4-noble)
bcv_1.0.2.tgz(r-4.4-emscripten)bcv_1.0.2.tgz(r-4.3-emscripten)
bcv.pdf |bcv.html✨
bcv/json (API)

# Install 'bcv' in R:
install.packages('bcv', repos = 'https://cloud.r-project.org')

Bug tracker:https://github.com/michbur/bcv/issues

Uses libs:
  • openblas– Optimized BLAS

On CRAN:

Conda:

openblas

2.00 score 358 downloads 1 mentions 3 exports 0 dependencies

Last updated 2 years agofrom:4d6cfd919c. Checks:3 OK. Indexed: no.

TargetResultLatest binary
Doc / VignettesOKMar 18 2025
R-4.5-linux-x86_64OKMar 18 2025
R-4.4-linux-x86_64OKMar 18 2025

Exports:cv.svd.gabrielcv.svd.woldimpute.svd

Dependencies:

Citation

To cite package ’bcv’ in publications use:

Patrick O. Perry, Seyed Mortez Najibi, Michal Burdukiewicz, (2023). bcv: Cross-Validation for the SVD (Bi-Cross-Validation). R package version 1.0.2.

Corresponding BibTeX entry: 

  @Manual{,
    title = {bcv: Cross-Validation for the SVD (Bi-Cross-Validation).},
    author = {Patrick O. Perry and Seyed Mortez Najibi and Michal
      Burdukiewicz},
    year = {2023},
    note = {R package version 1.0.2},
  }

Readme and manuals

About bcv

Methods for choosing the rank of an SVD approximation via cross validation. The package provides both Gabriel-style "block" holdouts and Wold-style "speckled" holdouts. It also includes an implementation of the SVDImpute algorithm. For more information about Bi-cross-validation, see Owen & Perry's 2009 AoAS article and Perry's 2009 PhD thesis.

Installation

bcv is available on CRAN, so installation is as simple as:

install.packages("bcv")

You can install the latest development version of the code using the devtools R package.

# Install devtools, if you haven't already.
install.packages("devtools")

library(devtools)
install_github("michbur/bcv")