sirus: Stable and Interpretable RUle Set

A regression and classification algorithm based on random forests, which takes the form of a short list of rules. SIRUS combines the simplicity of decision trees with a predictivity close to random forests. The core aggregation principle of random forests is kept, but instead of aggregating predictions, SIRUS aggregates the forest structure: the most frequent nodes of the forest are selected to form a stable rule ensemble model. The algorithm is fully described in the following articles: Benard C., Biau G., da Veiga S., Scornet E. (2021), Electron. J. Statist., 15:427-505 <doi:10.1214/20-EJS1792> for classification, and Benard C., Biau G., da Veiga S., Scornet E. (2021), AISTATS, PMLR 130:937-945 <>, for regression. This R package is a fork from the project ranger (<>).

Version: 0.3.3
Depends: R (≥ 3.6)
Imports: Rcpp (≥ 0.11.2), Matrix, ROCR, ggplot2, glmnet
LinkingTo: Rcpp, RcppEigen
Suggests: survival, testthat, ranger
Published: 2022-06-13
DOI: 10.32614/CRAN.package.sirus
Author: Clement Benard [aut, cre], Marvin N. Wright [ctb, cph]
Maintainer: Clement Benard <clement.benard5 at>
License: GPL-3
NeedsCompilation: yes
Materials: README
CRAN checks: sirus results


Reference manual: sirus.pdf


Package source: sirus_0.3.3.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release (arm64): sirus_0.3.3.tgz, r-oldrel (arm64): sirus_0.3.3.tgz, r-release (x86_64): sirus_0.3.3.tgz, r-oldrel (x86_64): sirus_0.3.3.tgz
Old sources: sirus archive


Please use the canonical form to link to this page.