Solving the linear SVM problem with coordinate descent is very efficient and is implemented in one of the most often used packages, 'LIBLINEAR' (available at http://www.csie.ntu.edu.tw/~cjlin/liblinear). It has been shown that the uniform selection of coordinates can be accelerated by using an online adaptation of coordinate frequencies (ACF). This package implements ACF and is based on 'LIBLINEAR' as well as the 'LiblineaR' package (). It currently supports L2-regularized L1-loss as well as L2-loss linear SVM. Similar to 'LIBLINEAR' multi-class classification (one-vs-the rest, and Crammer & Singer method) and cross validation for model selection is supported. The training of the models based on ACF is much faster than standard 'LIBLINEAR' on many problems.

Documentation

Manual: LiblineaR.ACF.pdf
Vignette: None available.

Maintainer: Aydin Demircioglu <aydin.demircioglu at ini.rub.de>

Author(s): Aydin Demircioglu <aydin.demircioglu at ini.rub.de>; Tobias Glasmachers <tobias.glasmachers at ini.rub.de>; Urun Dogan <urundogan at gmail.com>

Install package and any missing dependencies by running this line in your R console:

install.packages("LiblineaR.ACF")

Depends
Imports
Suggests SparseM, testthat
Enhances
Linking to
Reverse
depends
Reverse
imports
Reverse
suggests
Reverse
enhances
Reverse
linking to

Package LiblineaR.ACF
Materials
URL http://github.com/aydindemircioglu/liblineaR.ACF/
Task Views
Version 1.94-2
Published 2016-01-04
License GPL-2
BugReports
SystemRequirements
NeedsCompilation yes
Citation
CRAN checks LiblineaR.ACF check results
Package source LiblineaR.ACF_1.94-2.tar.gz