RVowpalWabbit: R interface to the Vowpal Wabbit

R interface to Vowpal Wabbit fast out-of-core learning system The Vowpal Wabbit (VW) project is a fast out-of-core learning system sponsored by Yahoo! Research and written by John Langford along with a number of contributors. There are two ways to have a fast learning algorithm: (a) start with a slow algorithm and speed it up, or (b) build an intrinsically fast learning algorithm. This project is about approach (b), and it has reached a state where it may be useful to others as a platform for research and experimentation. There are several optimization algorithms available with the baseline being sparse gradient descent (GD) on a loss function (several are available). The code should be easily usable. Its only external dependence is on the Boost library, which is often installed by default. This R package does not include the distributed computing implementation of the cluster/ directory of the upstream sources. Use of the software as a network servie is also not directly supported as the aim is a simpler direct call from R for validation and comparison.

Version: 0.0.6
Depends: R (≥ 2.12.0), Rcpp (≥ 0.9.6)
LinkingTo: Rcpp
OS_type: unix
Published: 2014-01-07
Author: Dirk Eddelbuettel
Maintainer: Dirk Eddelbuettel <edd at debian.org>
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
URL: https://github.com/JohnLangford/vowpal_wabbit/ http://dirk.eddelbuettel.com/code/rcpp.html
NeedsCompilation: yes
SystemRequirements: The Boost 'program_options' library (http://boost.org) is required.
Materials: README ChangeLog
CRAN checks: RVowpalWabbit results


Reference manual: RVowpalWabbit.pdf
Package source: RVowpalWabbit_0.0.6.tar.gz
Windows binaries: r-devel: not available, r-release: not available, r-oldrel: not available
OS X Snow Leopard binaries: r-release: not available, r-oldrel: RVowpalWabbit_0.0.5.tgz
OS X Mavericks binaries: r-release: not available
Old sources: RVowpalWabbit archive