Crandore Hub

hstats

Interaction Statistics

Fast, model-agnostic implementation of different H-statistics introduced by Jerome H. Friedman and Bogdan E. Popescu (2008) <doi:10.1214/07-AOAS148>. These statistics quantify interaction strength per feature, feature pair, and feature triple. The package supports multi-output predictions and can account for case weights. In addition, several variants of the original statistics are provided. The shape of the interactions can be explored through partial dependence plots or individual conditional expectation plots. 'DALEX' explainers, meta learners ('mlr3', 'tidymodels', 'caret') and most other models work out-of-the-box.

Versions across snapshots

VersionRepositoryFileSize
1.2.2 rolling linux/jammy R-4.5 hstats_1.2.2.tar.gz 399.7 KiB
1.2.2 rolling linux/noble R-4.5 hstats_1.2.2.tar.gz 399.5 KiB
1.2.2 rolling source/ R- hstats_1.2.2.tar.gz 250.5 KiB
1.2.2 latest linux/jammy R-4.5 hstats_1.2.2.tar.gz 399.7 KiB
1.2.2 latest linux/noble R-4.5 hstats_1.2.2.tar.gz 399.5 KiB
1.2.2 latest source/ R- hstats_1.2.2.tar.gz 250.5 KiB
1.2.2 2026-04-26 source/ R- hstats_1.2.2.tar.gz 250.5 KiB
1.2.2 2026-04-23 source/ R- hstats_1.2.2.tar.gz 250.5 KiB
1.2.2 2026-04-09 windows/windows R-4.5 hstats_1.2.2.zip 401.0 KiB
1.2.1 2025-04-20 source/ R- hstats_1.2.1.tar.gz 198.9 KiB

Dependencies (latest)

Imports

Suggests