1% Generated by roxygen2: do not edit by hand
2% Please edit documentation in R/selections.R
3\name{has_role}
4\alias{has_role}
5\alias{all_predictors}
6\alias{all_numeric_predictors}
7\alias{all_nominal_predictors}
8\alias{all_outcomes}
9\alias{has_type}
10\alias{all_numeric}
11\alias{all_nominal}
12\alias{current_info}
13\title{Role Selection}
14\usage{
15has_role(match = "predictor")
16
17all_predictors()
18
19all_numeric_predictors()
20
21all_nominal_predictors()
22
23all_outcomes()
24
25has_type(match = "numeric")
26
27all_numeric()
28
29all_nominal()
30
31current_info()
32}
33\arguments{
34\item{match}{A single character string for the query. Exact
35matching is used (i.e. regular expressions won't work).}
36}
37\value{
38Selector functions return an integer vector.
39
40\code{current_info()} returns an environment with objects \code{vars} and \code{data}.
41}
42\description{
43\code{has_role()}, \code{all_predictors()}, and \code{all_outcomes()} can be used to
44select variables in a formula that have certain roles.
45
46Similarly, \code{has_type()}, \code{all_numeric()}, and \code{all_nominal()} are used to
47select columns based on their data type. Nominal variables include both
48character and factor.
49
50\strong{In most cases}, the selectors \code{all_numeric_predictors()} and
51\code{all_nominal_predictors()}, which select on role and type, will be the right
52approach for users.
53
54See \link{selections} for more details.
55
56\code{current_info()} is an internal function.
57
58All of these functions have have limited utility outside of column selection
59in step functions.
60}
61\examples{
62library(modeldata)
63data(biomass)
64
65rec <- recipe(biomass) \%>\%
66  update_role(
67    carbon, hydrogen, oxygen, nitrogen, sulfur,
68    new_role = "predictor"
69  ) \%>\%
70  update_role(HHV, new_role = "outcome") \%>\%
71  update_role(sample, new_role = "id variable") \%>\%
72  update_role(dataset, new_role = "splitting indicator")
73
74recipe_info <- summary(rec)
75recipe_info
76
77# Centering on all predictors except carbon
78rec \%>\%
79  step_center(all_predictors(), -carbon) \%>\%
80  prep(training = biomass) \%>\%
81  bake(new_data = NULL)
82
83}
84