iterative Random Forests

Iteratively grows feature weighted random forests and finds high-order feature interactions in a stable fashion.


The R package iRF implements iterative Random Forests, a method for iteratively growing ensemble of weighted decision trees, and detecting high-order feature interactions by analyzing feature usage on decision paths. This version uses source codes from the R package randomForest by Andy Liaw and Matthew Weiner and the original Fortran codes by Leo Breiman and Adele Cutler.

To download and install the package, use devtools

library(devtools)
devtools::install_github("sumbose/iRF")

You can subsequently load the package with the usual R commands:

library(iRF)

OSX users may need to intall gfortran to compile. This can be done with the following commands:

curl -O http://r.research.att.com/libs/gfortran-4.8.2-darwin13.tar.bz2
sudo tar fvxz gfortran-4.8.2-darwin13.tar.bz2 -C /

Binaries are available for OSX and linux in the binaries directory and can be installed using the command:

R CMD INSTALL <filename>

For a detailed description on the usage of iRF, see the vignette.

News

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.

install.packages("iRF")

2.0.0 by Karl Kumbier, a year ago


https://arxiv.org/abs/1706.08457


Browse source code at https://github.com/cran/iRF


Authors: Sumanta Basu and Karl Kumbier (based on source codes from the R packages FSInteract by Hyun Jik Kim and Rajen D. Shah, randomForest by Andy Liaw and Matthew Wiener, and the original Fortran codes by Leo Breiman and Adele Cutler)


Documentation:   PDF Manual  


GPL-2 license


Imports AUC, Matrix, data.table, dplyr, Rcpp, methods, foreach, doParallel, RColorBrewer

Suggests MASS, rgl

Linking to Rcpp

System requirements: C++11


See at CRAN