vip: Variable Importance Plots

A general framework for constructing variable importance plots from various types machine learning models in R. Aside from some standard model- based variable importance measures, this package also provides a novel approach based on partial dependence plots (PDPs) and individual conditional expectation (ICE) curves as described in Greenwell et al. (2018) <arXiv:1805.04755>.

Version: 0.1.0
Imports: dplyr, ggplot2 (≥ 0.9.0), gridExtra, magrittr, pdp, plyr, stats, tibble, tidyr, utils
Suggests: C50, caret, earth, gbm, h2o, knitr, party, partykit, ranger, rpart, randomForest, rmarkdown, xgboost, glmnet, testthat
Published: 2018-06-15
Author: Brandon Greenwell ORCID iD [aut, cre], Brad Boehmke [aut]
Maintainer: Brandon Greenwell <greenwell.brandon at gmail.com>
BugReports: https://github.com/koalaverse/vip/issues
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
URL: https://github.com/koalaverse/vip
NeedsCompilation: no
Materials: README NEWS
CRAN checks: vip results

Downloads:

Reference manual: vip.pdf
Package source: vip_0.1.0.tar.gz
Windows binaries: r-devel: vip_0.1.0.zip, r-release: vip_0.1.0.zip, r-oldrel: vip_0.1.0.zip
OS X binaries: r-release: vip_0.1.0.tgz, r-oldrel: vip_0.1.0.tgz

Linking:

Please use the canonical form https://CRAN.R-project.org/package=vip to link to this page.