msgl: High Dimensional Multiclass Classification Using Sparse Group Lasso

Multinomial logistic regression with sparse group lasso penalty. Suitable for high dimensional multiclass classification with many classes. The algorithm finds the sparse group lasso penalized maximum likelihood estimator. This result in feature and parameter selection, and parameter estimation. Use of multiple processors for cross validation and subsampling is supported through OpenMP. Development version is on github.

Version: 2.2.0
Depends: R (≥ 3.0.0), Matrix, sglOptim (== 1.2.0)
Imports: methods, utils, stats
LinkingTo: Rcpp, RcppProgress, RcppArmadillo, BH, sglOptim
Published: 2015-09-19
Author: Martin Vincent
Maintainer: Martin Vincent <vincent at>
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation: yes
Citation: msgl citation info
Materials: NEWS
CRAN checks: msgl results


Reference manual: msgl.pdf
Package source: msgl_2.2.0.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
OS X Snow Leopard binaries: r-release: msgl_2.0.125.1.tgz, r-oldrel: msgl_2.0.125.1.tgz
OS X Mavericks binaries: r-release: msgl_2.2.0.tgz
Old sources: msgl archive