L0Learn: Fast Algorithms for Best Subset Selection

Highly optimized toolkit for (approximately) solving L0-regularized learning problems. The algorithms are based on coordinate descent and local combinatorial search. For more details, check the paper by Hazimeh and Mazumder (2018) <arXiv:1803.01454>; the link is provided in the URL field below.

Version: 1.0.7
Depends: R (≥ 3.3.0)
Imports: Rcpp (≥ 0.12.13), Matrix, methods, ggplot2, reshape2
LinkingTo: Rcpp, RcppArmadillo
Suggests: knitr, rmarkdown
Published: 2018-09-11
Author: Hussein Hazimeh, Rahul Mazumder
Maintainer: Hussein Hazimeh <hazimeh at mit.edu>
License: MIT + file LICENSE
URL: https://arxiv.org/abs/1803.01454
NeedsCompilation: yes
CRAN checks: L0Learn results

Downloads:

Reference manual: L0Learn.pdf
Vignettes: L0Learn Vignette
Package source: L0Learn_1.0.7.tar.gz
Windows binaries: r-devel: L0Learn_1.0.7.zip, r-release: L0Learn_1.0.7.zip, r-oldrel: L0Learn_1.0.7.zip
OS X binaries: r-release: L0Learn_1.0.7.tgz, r-oldrel: L0Learn_1.0.7.tgz
Old sources: L0Learn archive

Linking:

Please use the canonical form https://CRAN.R-project.org/package=L0Learn to link to this page.