workflowsets: Create a Collection of 'tidymodels' Workflows

A workflow is a combination of a model and preprocessors (e.g, a formula, recipe, etc.) (Kuhn and Silge (2021) <>). In order to try different combinations of these, an object can be created that contains many workflows. There are functions to create workflows en masse as well as training them and visualizing the results.

Version: 0.2.1
Depends: R (≥ 3.4)
Imports: cli, dplyr (≥ 1.0.0), generics, ggplot2, hardhat (≥ 0.1.6), lifecycle (≥ 1.0.0), prettyunits, purrr, rlang, rsample (≥ 0.0.9), stats, tibble (≥ 3.1.0), tidyr, tune (≥ 0.1.3), vctrs, withr, workflows (≥ 0.2.4)
Suggests: covr, dials, kknn, knitr, modeldata, parsnip (≥ 0.2.0), recipes (≥ 0.1.15), rmarkdown, spelling, testthat (≥ 3.0.0), yardstick
Published: 2022-03-15
Author: Max Kuhn ORCID iD [aut, cre], RStudio [cph, fnd]
Maintainer: Max Kuhn <max at>
License: MIT + file LICENSE
NeedsCompilation: no
Language: en-US
Materials: README NEWS
CRAN checks: workflowsets results


Reference manual: workflowsets.pdf
Vignettes: Evaluating different predictor sets


Package source: workflowsets_0.2.1.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release (arm64): workflowsets_0.2.1.tgz, r-release (x86_64): workflowsets_0.2.1.tgz, r-oldrel: workflowsets_0.2.1.tgz
Old sources: workflowsets archive

Reverse dependencies:

Reverse imports: healthyR.ts, stacks, tidymodels, tidyposterior
Reverse suggests: modeltime, sknifedatar


Please use the canonical form to link to this page.