probably: Tools for Post-Processing Class Probability Estimates

Models can be improved by post-processing class probabilities, by: recalibration, conversion to hard probabilities, assessment of equivocal zones, and other activities. 'probably' contains tools for conducting these operations.

Version:	0.1.0
Depends:	R (≥ 3.4.0)
Imports:	dplyr (≥ 1.0.9), generics (≥ 0.1.3), magrittr (≥ 2.0.3), rlang (≥ 1.0.4), tidyselect (≥ 1.1.2), vctrs (≥ 0.4.1), yardstick (≥ 1.0.0)
Suggests:	covr, ggplot2, knitr, modeldata, parsnip, rmarkdown, rsample, testthat (≥ 3.0.0)
Published:	2022-08-29
Author:	Max Kuhn [aut, cre], Davis Vaughan [aut], RStudio [cph, fnd]
Maintainer:	Max Kuhn <max at rstudio.com>
BugReports:	https://github.com/tidymodels/probably/issues
License:	MIT + file LICENSE
URL:	https://github.com/tidymodels/probably/, https://probably.tidymodels.org
NeedsCompilation:	no
Materials:	README NEWS
CRAN checks:	probably results

Documentation:

Reference manual:	probably.pdf
Vignettes:	Equivocal zones Where does probably fit in?

Downloads:

Package source:	probably_0.1.0.tar.gz
Windows binaries:	r-devel: probably_0.1.0.zip, r-release: probably_0.1.0.zip, r-oldrel: probably_0.1.0.zip
macOS binaries:	r-release (arm64): probably_0.1.0.tgz, r-oldrel (arm64): probably_0.1.0.tgz, r-release (x86_64): probably_0.1.0.tgz, r-oldrel (x86_64): probably_0.1.0.tgz
Old sources:	probably archive

Reverse dependencies:

Reverse suggests:

yardstick

Linking:

Please use the canonical form https://CRAN.R-project.org/package=probably to link to this page.