Interfaces for choosing important predictors in supervised
regression, classification, and censored regression models. Permuted
importance scores (Biecek and Burzykowski (2021)
<doi:10.1201/9780429027192>) can be computed for 'tidymodels' model
fits.
| Version: |
0.2.1 |
| Depends: |
R (≥ 4.1.0), recipes (≥ 1.1.0) |
| Imports: |
cli, desirability2 (≥ 0.2.0), dplyr, filtro (≥ 0.2.0), generics, ggplot2, hardhat (≥ 1.4.1), purrr, rlang (≥ 1.1.0), S7, tibble, tidyr, tune, vctrs, withr, workflows |
| Suggests: |
censored, future, future.apply, mirai, modeldata, parsnip, ranger, spelling, survival, testthat (≥ 3.0.0), yardstick |
| Published: |
2025-10-06 |
| DOI: |
10.32614/CRAN.package.important |
| Author: |
Max Kuhn  [aut,
cre],
Posit Software, PBC  [cph,
fnd] |
| Maintainer: |
Max Kuhn <max at posit.co> |
| BugReports: |
https://github.com/tidymodels/important/issues |
| License: |
MIT + file LICENSE |
| URL: |
https://important.tidymodels.org/,
https://github.com/tidymodels/important |
| NeedsCompilation: |
no |
| Language: |
en-US |
| Materials: |
README, NEWS |
| CRAN checks: |
important results |