flashlight: Shed Light on Black Box Machine Learning Models

Shed light on black box machine learning models by the help of model performance, permutation variable importance (Fisher et al. (2018) <arxiv:1801.01489>), ICE profiles, partial dependence (Friedman J. H. (2001) <doi:10.1214/aos/1013203451>), accumulated local effects (Apley D. W. (2016) <arXiv:1612.08468>), further effects plots, overall interaction strength, and variable contribution breakdown for single observations (Gosiewska and Biecek (2019) <arxiv:1903.11420>). All tools are implemented to work with case weights and allow for stratified analysis. Furthermore, multiple flashlights can be combined and analyzed together.

Version: 0.4.0
Depends: R (≥ 3.5.0)
Imports: stats, utils, dplyr, tidyr, rlang, ggplot2, ggpubr, MetricsWeighted (≥ 0.3.0)
Suggests: knitr, lubridate, ranger, xgboost, caret, moderndive
Published: 2019-11-10
Author: Michael Mayer [aut, cre, cph]
Maintainer: Michael Mayer <mayermichael79 at gmail.com>
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation: no
Materials: README NEWS
CRAN checks: flashlight results

Downloads:

Reference manual: flashlight.pdf
Vignettes: flashlight
Package source: flashlight_0.4.0.tar.gz
Windows binaries: r-devel: flashlight_0.4.0.zip, r-release: flashlight_0.4.0.zip, r-oldrel: flashlight_0.4.0.zip
OS X binaries: r-release: flashlight_0.4.0.tgz, r-oldrel: flashlight_0.4.0.tgz
Old sources: flashlight archive

Linking:

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