fastshap: Fast Approximate Shapley Values

Computes fast (relative to other implementations) approximate Shapley values for any supervised learning model. Shapley values help to explain the predictions from any black box model using ideas from game theory; see Strumbel and Kononenko (2014) <doi:10.1007/s10115-013-0679-x> for details.

Version: 0.0.7
Imports: abind, ggplot2 (≥ 3.3.4), gridExtra, matrixStats, plyr, Rcpp (≥ 1.0.1), tibble
LinkingTo: Rcpp, RcppArmadillo
Suggests: AmesHousing, covr, earth, htmltools, knitr, lightgbm, ranger, reticulate (≥ 1.14), rmarkdown, rstudioapi, tinytest, titanic, xgboost
Published: 2021-12-06
Author: Brandon Greenwell ORCID iD [aut, cre]
Maintainer: Brandon Greenwell <greenwell.brandon at>
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation: yes
Materials: NEWS
CRAN checks: fastshap results


Reference manual: fastshap.pdf


Package source: fastshap_0.0.7.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release (arm64): fastshap_0.0.7.tgz, r-release (x86_64): fastshap_0.0.7.tgz, r-oldrel: fastshap_0.0.7.tgz
Old sources: fastshap archive

Reverse dependencies:

Reverse imports: itsdm
Reverse suggests: ENMTools, vip


Please use the canonical form to link to this page.