rules is a “parsnip-adjacent” packages with model definitions for different rule-based models, including:

  • cubist models that have discrete rule sets that contain linear models with an ensemble method similar to boosting
  • classification rules where a ruleset is derived from an initial tree fit
  • rule-fit models that begin with rules extracted from a tree ensemble which are then added to a regularized linear or logistic regression.


Th package is not yet on CRAN and can be installed via:

# install.packages("devtools")

Code of Conduct

Please note that the rules project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.