1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
|
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/sample.R
\name{step_sample}
\alias{step_sample}
\alias{tidy.step_sample}
\title{Sample rows using dplyr}
\usage{
step_sample(
recipe,
...,
role = NA,
trained = FALSE,
size = NULL,
replace = FALSE,
skip = TRUE,
id = rand_id("sample")
)
\method{tidy}{step_sample}(x, ...)
}
\arguments{
\item{recipe}{A recipe object. The step will be added to the
sequence of operations for this recipe.}
\item{...}{Argument ignored; included for consistency with other step
specification functions. For the \code{tidy}
method, these are not currently used.}
\item{role}{Not used by this step since no new variables are
created.}
\item{trained}{A logical to indicate if the quantities for
preprocessing have been estimated.}
\item{size}{An integer or fraction. If the value is within (0, 1),
\code{\link[dplyr:sample_n]{dplyr::sample_frac()}} is applied to the data. If an integer
value of 1 or greater is used, \code{\link[dplyr:sample_n]{dplyr::sample_n()}} is applied.
The default of \code{NULL} uses \code{\link[dplyr:sample_n]{dplyr::sample_n()}} with the size
of the training set (or smaller for smaller \code{new_data}).}
\item{replace}{Sample with or without replacement?}
\item{skip}{A logical. Should the step be skipped when the
recipe is baked by \code{\link[=bake.recipe]{bake.recipe()}}? While all operations are baked
when \code{\link[=prep.recipe]{prep.recipe()}} is run, some operations may not be able to be
conducted on new data (e.g. processing the outcome variable(s)).
Care should be taken when using \code{skip = FALSE}.}
\item{id}{A character string that is unique to this step to identify it.}
\item{x}{A \code{step_sample} object}
}
\value{
An updated version of \code{recipe} with the new step
added to the sequence of existing steps (if any). For the
\code{tidy} method, a tibble with columns \code{size}, \code{replace},
and \code{id}.
}
\description{
\code{step_sample} creates a \emph{specification} of a recipe step
that will sample rows using \code{\link[dplyr:sample_n]{dplyr::sample_n()}} or
\code{\link[dplyr:sample_n]{dplyr::sample_frac()}}.
}
\section{Row Filtering}{
This step can entirely remove observations (rows of data), which can have
unintended and/or problematic consequences when applying the step to new
data later via \code{\link[=bake.recipe]{bake.recipe()}}. Consider whether \code{skip = TRUE} or
\code{skip = FALSE} is more appropriate in any given use case. In most instances
that affect the rows of the data being predicted, this step probably should
not be applied at all; instead, execute operations like this outside and
before starting a preprocessing \code{\link[=recipe]{recipe()}}.
}
\examples{
# Uses `sample_n`
recipe( ~ ., data = mtcars) \%>\%
step_sample(size = 1) \%>\%
prep(training = mtcars) \%>\%
bake(new_data = NULL) \%>\%
nrow()
# Uses `sample_frac`
recipe( ~ ., data = mtcars) \%>\%
step_sample(size = 0.9999) \%>\%
prep(training = mtcars) \%>\%
bake(new_data = NULL) \%>\%
nrow()
# Uses `sample_n` and returns _at maximum_ 20 samples.
smaller_cars <-
recipe( ~ ., data = mtcars) \%>\%
step_sample() \%>\%
prep(training = mtcars \%>\% slice(1:20))
bake(smaller_cars, new_data = NULL) \%>\% nrow()
bake(smaller_cars, new_data = mtcars \%>\% slice(21:32)) \%>\% nrow()
}
\seealso{
\code{\link[=step_filter]{step_filter()}} \code{\link[=step_naomit]{step_naomit()}} \code{\link[=step_slice]{step_slice()}}
}
\concept{preprocessing}
\keyword{datagen}
|
