parboost: Distributed Model-Based Boosting

Distributed gradient boosting based on the mboost package. The parboost package is designed to scale up component-wise functional gradient boosting in a distributed memory environment by splitting the observations into disjoint subsets, or alternatively using bootstrap samples (bagging). Each cluster node then fits a boosting model to its subset of the data. These boosting models are combined in an ensemble, either with equal weights, or by fitting a (penalized) regression model on the predictions of the individual models on the complete data.

Version: 0.1.3
Depends: R (≥ 3.0.1), parallel, mboost, party, iterators
Imports: plyr, caret, glmnet, doParallel
Published: 2014-03-03
Author: Ronert Obst
Maintainer: Ronert Obst <ronert.obst at gmail.com>
License: GPL-2
NeedsCompilation: no
Materials: README NEWS
CRAN checks: parboost results

Downloads:

Reference manual: parboost.pdf
Package source: parboost_0.1.3.tar.gz
Windows binaries: r-devel: parboost_0.1.3.zip, r-release: parboost_0.1.3.zip, r-oldrel: parboost_0.1.3.zip
OS X Snow Leopard binaries: r-release: parboost_0.1.3.tgz, r-oldrel: parboost_0.1.3.tgz
OS X Mavericks binaries: r-release: parboost_0.1.3.tgz
Old sources: parboost archive