Package: glex 0.7.0.9000

Marvin N. Wright

glex: Global Explanations for Tree-Based Models

Global explanations for tree-based models by decomposing regression or classification functions into the sum of main components and interaction components of arbitrary order. Calculates SHAP values and q-interaction SHAP for all values of q for tree-based models such as xgboost.

Authors:Marvin N. Wright [aut, cre], Joseph Theo Meyer [aut], Munir Hiabu [aut], Lukas Burk [aut], Jinyang Liu [aut]

glex_0.7.0.9000.tar.gz
glex_0.7.0.9000.zip(r-4.7-x86_64)glex_0.7.0.9000.zip(r-4.7-arm64)glex_0.7.0.9000.zip(r-4.6-x86_64)glex_0.7.0.9000.zip(r-4.6-arm64)glex_0.7.0.9000.zip(r-4.5-x86_64)
glex_0.7.0.9000.tgz(r-4.6-x86_64)glex_0.7.0.9000.tgz(r-4.6-arm64)glex_0.7.0.9000.tgz(r-4.5-x86_64)glex_0.7.0.9000.tgz(r-4.5-arm64)
glex_0.7.0.9000.tar.gz(r-4.7-arm64)glex_0.7.0.9000.tar.gz(r-4.7-x86_64)glex_0.7.0.9000.tar.gz(r-4.6-arm64)glex_0.7.0.9000.tar.gz(r-4.6-x86_64)
glex_0.7.0.9000.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
glex/json (API)

# Install 'glex' in R:

install.packages('glex', repos = c('https://plantedml.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/plantedml/glex/issues

Pkgdown/docs site:https://plantedml.com

Uses libs:

c++– GNU Standard C++ Library v3

Datasets:

bike - Bikesharing data

On CRAN:

cpp

5.82 score 8 stars 39 scripts 15 exports 31 dependencies

Last updated from:c30470ed21. Checks:15 OK. Indexed: yes.

Target	Result	Time
linux-devel-arm64	OK	201
linux-devel-x86_64	OK	222
source / vignettes	OK	337
linux-release-arm64	OK	190
linux-release-x86_64	OK	207
macos-release-arm64	OK	213
macos-release-x86_64	OK	475
macos-oldrel-arm64	OK	188
macos-oldrel-x86_64	OK	593
windows-devel-arm64	OK	336
windows-devel-x86_64	OK	369
windows-release-arm64	OK	329
windows-release-x86_64	OK	346
windows-oldrel-x86_64	OK	339
wasm-release	OK	100

Exports:dummy_groups glex glex_explain glex_vi group_components plot_force plot_main_effect plot_pdp plot_shap_decomposition plot_threeway_effects plot_twoway_effects plot_waterfall subset_component_names subset_components theme_glex

Dependencies:backports checkmate cli codetools cpp11 crayon data.table farver foreach ggplot2 glue gtable hms isoband iterators jsonlite labeling lifecycle pkgconfig prettyunits progress R6 RColorBrewer Rcpp rlang S7 scales scico vctrs viridisLite withr

Categorical features: native factors, one-hot groups and semantic groups

Native factors | Dummy encoding fragments the decomposition | Re-assembling the factor with group_components() | Other encoding schemes | Semantic groups

Rendered fromfeature-grouping.Rmdusingknitr::rmarkdown

Last update: 2026-11-11
Started: 2026-10-03

glex

Rendered fromglex.Rmdusingknitr::rmarkdown

Last update: 2023-02-24
Started: 2023-02-24

Citation

Development and contributors

Readme and manuals

Help Manual

Help page	Topics
Bikesharing data	bike
Derive feature groups from the factors behind a dummy encoding	dummy_groups
Global explanations for tree-based models.	glex glex.default glex.ranger glex.rpf glex.rpf_pruned glex.xgb.Booster
Package options for glex plots	glex_options
Variable Importance for Main and Interaction Terms	glex_vi
Group features of a decomposition	group_components
Force plot of a single prediction	plot_force
Partial Dependence Plot	plot_pdp
Plot how a feature's SHAP value is assembled from components	glex_explain plot_shap_decomposition
Waterfall plot of a single prediction	plot_waterfall
Plot Prediction Components	plot.glex plot_main_effect plot_threeway_effects plot_twoway_effects
Plot glex Variable Importances	plot.glex_vi
Print glex objects	print.glex
Subset components	subset_components subset_component_names
A ggplot2 theme for glex plots	theme_glex