stacks: Tidy Model Stacking

Model stacking is an ensemble technique that involves training a model to combine the outputs of many diverse statistical models, and has been shown to improve predictive performance in a variety of settings. 'stacks' implements a grammar for 'tidymodels'-aligned model stacking.

Version: 1.0.1
Depends: R (≥ 2.10)
Imports: tune (≥ 0.1.3), dplyr (≥ 1.0.0), rlang (≥ 0.4.0), tibble (≥ 2.1.3), purrr (≥ 0.3.2), parsnip (≥ 1.0.2), workflows (≥ 0.2.3), recipes (≥ 0.2.0), rsample (≥ 0.1.1), workflowsets (≥ 0.1.0), butcher (≥ 0.1.3), yardstick, tidyr, glue, ggplot2, glmnet, cli, stats, foreach
Suggests: testthat (≥ 3.0.0), covr, kknn, ranger, knitr, modeldata, rmarkdown, nnet, kernlab, mockr, h2o, SuperLearner
Published: 2022-12-14
Author: Simon Couch [aut, cre], Max Kuhn [aut], RStudio [cph, fnd]
Maintainer: Simon Couch <simonpatrickcouch at>
License: MIT + file LICENSE
NeedsCompilation: no
Materials: README NEWS
CRAN checks: stacks results


Reference manual: stacks.pdf
Vignettes: Getting Started With stacks
Classification Models With stacks


Package source: stacks_1.0.1.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release (arm64): stacks_1.0.1.tgz, r-oldrel (arm64): stacks_1.0.1.tgz, r-release (x86_64): stacks_1.0.1.tgz, r-oldrel (x86_64): stacks_1.0.1.tgz
Old sources: stacks archive

Reverse dependencies:

Reverse suggests: bundle, DALEXtra, ensModelVis, vetiver


Please use the canonical form to link to this page.