fmeffects: Model-Agnostic Interpretations with Forward Marginal Effects

Create local, regional, and global explanations for any machine learning model with forward marginal effects. You provide a model and data, and 'fmeffects' computes feature effects. The package is based on the theory in: C. A. Scholbeck, G. Casalicchio, C. Molnar, B. Bischl, and C. Heumann (2022) <arXiv:2201.08837>.

Version: 0.1.1
Depends: R (≥ 3.5.0)
Imports: checkmate, data.table, partykit, ggparty, ggplot2, cowplot, R6, testthat
Suggests: caret, knitr, mlr3verse, ranger, rmarkdown, rpart
Published: 2023-09-26
Author: Holger Löwe [cre, aut], Christian Scholbeck [aut], Christian Heumann [rev], Bernd Bischl [rev], Giuseppe Casalicchio [rev]
Maintainer: Holger Löwe <hbj.loewe at gmail.com>
BugReports: https://github.com/holgstr/fmeffects/issues
License: LGPL-3
URL: https://github.com/holgstr/fmeffects
NeedsCompilation: no
Materials: README NEWS
CRAN checks: fmeffects results

Documentation:

Reference manual: fmeffects.pdf
Vignettes: Why FMEs?
Get started

Downloads:

Package source: fmeffects_0.1.1.tar.gz
Windows binaries: r-devel: fmeffects_0.1.1.zip, r-release: fmeffects_0.1.1.zip, r-oldrel: fmeffects_0.1.1.zip
macOS binaries: r-release (arm64): fmeffects_0.1.1.tgz, r-oldrel (arm64): fmeffects_0.1.1.tgz, r-release (x86_64): fmeffects_0.1.1.tgz
Old sources: fmeffects archive

Reverse dependencies:

Reverse suggests: marginaleffects

Linking:

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