Package: xplainfi 1.2.0

Lukas Burk

xplainfi: Feature Importance Methods for Global Explanations

Provides a consistent interface for common feature importance methods as described in Ewald et al. (2024) <doi:10.1007/978-3-031-63797-1_22>, including permutation feature importance (PFI), conditional and relative feature importance (CFI, RFI), leave one covariate out (LOCO), and Shapley additive global importance (SAGE), as well as feature sampling mechanisms to support conditional importance methods.

Authors:Lukas Burk [aut, cre, cph]

xplainfi_1.2.0.tar.gz
xplainfi_1.2.0.zip(r-4.7-any)xplainfi_1.2.0.zip(r-4.6-any)xplainfi_1.2.0.zip(r-4.5-any)
xplainfi_1.2.0.tgz(r-4.6-any)xplainfi_1.2.0.tgz(r-4.5-any)
xplainfi_1.2.0.tar.gz(r-4.7-any)xplainfi_1.2.0.tar.gz(r-4.6-any)
xplainfi_1.2.0.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
xplainfi/json (API)

# Install 'xplainfi' in R:
install.packages('xplainfi', repos = c('https://mlr-org.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/mlr-org/xplainfi/issues

Pkgdown/docs site:https://mlr-org.github.io

On CRAN:

Conda:

feature-importanceinterpretable-machine-learningmachine-learningmlr3statistical-inferencevariable-importance

6.47 score 6 stars 13 scripts 234 downloads 31 exports 35 dependencies

Last updated from:f9aeb38e21 (on v1.2.0). Checks:9 OK. Indexed: yes.

TargetResultTimeFilesSyslog
linux-develOK207
source / vignettesOK241
linux-releaseOK215
macos-releaseOK168
macos-oldrelOK164
windows-develOK164
windows-releaseOK140
windows-oldrelOK170
wasm-releaseOK145

Exports:CFIcheck_groupsConditionalARFSamplerConditionalCtreeSamplerConditionalGaussianSamplerConditionalKNNSamplerConditionalSAGEConditionalSamplerFeatureImportanceMethodFeatureSamplerKnockoffGaussianSamplerKnockoffSamplerLOCOMarginalPermutationSamplerMarginalReferenceSamplerMarginalSAGEMarginalSamplerPerturbationImportancePFIRFIrsmp_all_testSAGEsim_dgp_confoundedsim_dgp_correlatedsim_dgp_ewaldsim_dgp_independentsim_dgp_interactionssim_dgp_mediatedWVIMwvim_design_matrixxplain_opt

Dependencies:backportsbbotkcheckmateclicodetoolsdata.tabledigestevaluatefuturefuture.applyglobalslatticelgrlistenvMatrixmatrixStatsmiraimlbenchmlr3mlr3fselectmlr3measuresmlr3miscmoocoremvtnormnanonextpalmerpenguinsparadoxparallellyPRROCR6rbibutilsRdpackrlangstabmuuid

Getting Started with xplainfi
Core Concepts | Basic Example | Permutation Feature Importance (PFI) | Leave-One-Covariate-Out (LOCO) | Feature Samplers | Detailed Scoring Information | Observation-wise losses and importances | Statistical Inference | Using Pre-trained Learners | Parallelization | Example with future | Example with mirai
Rendered fromxplainfi.Rmdusingknitr::rmarkdown

Last update: 2026-07-14
Started: 2024-10-14

Feature Samplers
Setup | Base Class: FeatureSampler | Key Properties | Permutation Sampler | Marginal Reference Sampler | Conditional Samplers | Gaussian Conditional Sampler | ARF Sampler | Ctree Conditional Sampler | kNN Conditional Sampler | Example 1: All-numeric conditioning (Euclidean distance) | Example 2: Mixed-type conditioning (Gower distance) | Knockoff Samplers | Gaussian Knockoffs | Summary
Rendered fromfeature-samplers.Rmdusingknitr::rmarkdown

Last update: 2026-02-26
Started: 2025-10-27

Simulation Settings for Feature Importance Methods
Introduction | Overview of Simulation Settings | 1. Correlated Features DGP | Mathematical Model | Causal Structure | Usage Example | 2. Mediated Effects DGP | 3. Confounding DGP | 4. Interaction Effects DGP | 5. Independent Features DGP (Baseline) | Expected Behavior | 6. Ewald et al. (2024) DGP
Rendered fromsimulation-settings.Rmdusingknitr::rmarkdown

Last update: 2026-01-29
Started: 2025-10-09