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
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
|
#' Sample rows using dplyr
#'
#' `step_sample` creates a *specification* of a recipe step
#' that will sample rows using [dplyr::sample_n()] or
#' [dplyr::sample_frac()].
#'
#' @template row-ops
#' @inheritParams step_center
#' @param ... Argument ignored; included for consistency with other step
#' specification functions. For the `tidy`
#' method, these are not currently used.
#' @param role Not used by this step since no new variables are
#' created.
#' @param size An integer or fraction. If the value is within (0, 1),
#' [dplyr::sample_frac()] is applied to the data. If an integer
#' value of 1 or greater is used, [dplyr::sample_n()] is applied.
#' The default of `NULL` uses [dplyr::sample_n()] with the size
#' of the training set (or smaller for smaller `new_data`).
#' @param skip A logical. Should the step be skipped when the
#' recipe is baked by [bake.recipe()]? While all operations are baked
#' when [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 `skip = FALSE`.
#' @param replace Sample with or without replacement?
#' @return An updated version of `recipe` with the new step
#' added to the sequence of existing steps (if any). For the
#' `tidy` method, a tibble with columns `size`, `replace`,
#' and `id`.
#' @keywords datagen
#' @concept preprocessing
#' @export
#' @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 [step_filter()] [step_naomit()] [step_slice()]
step_sample <- function(
recipe, ...,
role = NA,
trained = FALSE,
size = NULL,
replace = FALSE,
skip = TRUE,
id = rand_id("sample")
) {
if (length(list(...)) > 0) {
rlang::warn("Selectors are not used for this step.")
}
if (!is_tune(size) & !is_varying(size)) {
if (!is.null(size) & (!is.numeric(size) || size < 0)) {
rlang::abort("`size` should be a positive number or NULL.")
}
}
if (!is_tune(replace) & !is_varying(replace)) {
if (!is.logical(replace)) {
rlang::abort("`replace` should be a single logical.")
}
}
add_step(
recipe,
step_sample_new(
terms = terms,
trained = trained,
role = role,
size = size,
replace = replace,
skip = skip,
id = id
)
)
}
step_sample_new <-
function(terms, role, trained, size, replace, skip, id) {
step(
subclass = "sample",
terms = terms,
role = role,
trained = trained,
size = size,
replace = replace,
skip = skip,
id = id
)
}
#' @export
prep.step_sample <- function(x, training, info = NULL, ...) {
if (is.null(x$size)) {
x$size <- nrow(training)
}
step_sample_new(
terms = x$terms,
trained = TRUE,
role = x$role,
size = x$size,
replace = x$replace,
skip = x$skip,
id = x$id
)
}
#' @export
bake.step_sample <- function(object, new_data, ...) {
if (object$size >= 1) {
n <- min(object$size, nrow(new_data))
new_data <-
dplyr::sample_n(new_data, size = floor(n), replace = object$replace)
} else {
new_data <-
dplyr::sample_frac(new_data, size = object$size, replace = object$replace)
}
new_data
}
print.step_sample <-
function(x, width = max(20, options()$width - 35), ...) {
cat("Row sampling")
if (x$replace)
cat(" with replacement")
if (x$trained) {
cat(" [trained]\n")
} else {
cat("\n")
}
invisible(x)
}
#' @rdname step_sample
#' @param x A `step_sample` object
#' @export
tidy.step_sample <- function(x, ...) {
tibble(
size = x$size,
replace = x$replace,
id = x$inputs
)
}
|
