vip: Variable Importance Plots

A general framework for constructing variable importance plots from various types of machine learning models in R. Aside from some standard model- specific variable importance measures, this package also provides model- agnostic approaches that can be applied to any supervised learning algorithm. These include an efficient permutation-based variable importance measure as well as novel approaches based on partial dependence plots (PDPs) and individual conditional expectation (ICE) curves which are described in Greenwell et al. (2018) <arXiv:1805.04755>. An experimental method for quantifying the relative strength of interaction effects is also included (see the previous reference for details).

Version: 0.1.3
Imports: ggplot2 (≥ 0.9.0), gridExtra, magrittr, ModelMetrics, pdp, plyr, stats, tibble, utils
Suggests: DT, C50, caret, Ckmeans.1d.dp, covr, Cubist, doParallel, dplyr, earth, gbm, glmnet, h2o, htmlwidgets, keras, knitr, lattice, mlbench, neuralnet, NeuralNetTools, nnet, party, partykit, randomForest, ranger, rmarkdown, rpart, RSNNS, sparkline, sparklyr, testthat, varImp, xgboost
Published: 2019-07-03
Author: Brandon Greenwell ORCID iD [aut, cre], Brad Boehmke ORCID iD [aut], Bernie Gray ORCID iD [aut]
Maintainer: Brandon Greenwell <greenwell.brandon at>
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation: no
Materials: README NEWS
CRAN checks: vip results


Reference manual: vip.pdf
Package source: vip_0.1.3.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
OS X binaries: r-release: vip_0.1.3.tgz, r-oldrel: vip_0.1.3.tgz
Old sources: vip archive

Reverse dependencies:

Reverse suggests: pdp


Please use the canonical form to link to this page.