Learn R Programming

boostr (version 1.0.0)

A modular framework to bag or boost any estimation procedure.

Description

boostr provides a modular framework that return the focus of ensemble learning back to 'learning' (instead of programming).

Copy Link

Version

Install

install.packages('boostr')

Monthly Downloads

9

Version

1.0.0

License

GPL-2

Maintainer

Steven Pollack

Last Published

May 17th, 2014

Functions in boostr (1.0.0)

isClassConstructor

check if a function is a(n S3) class constructor
vanillaBagger

Standard (vanilla) bagging procedure.
boost

Boost an Estimation Procedure with a Reweighter and an Aggregator.
arcfsReweighter

Reweighter for the arc-fs algorithm.
addDots

Extend a function's signature to include '...'
buildEstimationProcedure

Build a boostr compatible estimation procedure.
arcx4Reweighter

Reweighter for the arc-x4 algorithm.
defaultOOBPerformanceAnalysis

Perform generic out-of-bag error analysis.
kFoldCV

Generic k-fold Cross Validation wrapper
predictResponseFromWeightedAverage

Predict a numeric response using (un)weighted averaging.
ensembleEstimators

Extraction functions for boostr object attributes
predictClassFromWeightedVote

Predict a class using (un)weighted voting.
boostBackend

Boost an estimation procedure with a reweighter and aggregator.
arcx4Aggregator

Stock aggregators
wrapPerformanceAnalyzer

Create a boostr compatible wrapper for a performance analyzer.
boostWithArcFs

Boostr implemented versions of arc-fs, arc-x4 and AdaBoost.
wrapAggregator

Create a boostr compatible wrapper for an aggregator.
arcfsAggregator

Aggregator for the arc-fs algorithm.
adaboostReweighter

Reweighter function for the Adaboost.M1 algorithm
boostr

Boost (or bag) an estimation procedure with any reweighter or aggregator.
wrapProcedure

Create a boostr compatible wrapper for an estimation procedure.
makePredictions

Gather predictions from an ensemble of estimators.
wrapReweighter

Create a boostr compatible wrapper for a reweighter.
adaboostAggregator

Aggregator for the Adaboost.M1 algorithm