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
166
167
168
169
170
171
172
|
#' Scaling Numeric Data
#'
#' `step_scale` creates a *specification* of a recipe
#' step that will normalize numeric data to have a standard
#' deviation of one.
#'
#' @inheritParams step_center
#' @param sds A named numeric vector of standard deviations. This is `NULL`
#' until computed by [prep()].
#' @param factor A numeric value of either 1 or 2 that scales the
#' numeric inputs by one or two standard deviations. By dividing
#' by two standard deviations, the coefficients attached to
#' continuous predictors can be interpreted the same way as with
#' binary inputs. Defaults to `1`. More in reference below.
#' @param na_rm A logical value indicating whether `NA`
#' values should be removed when computing the standard deviation.
#' @template step-return
#' @family normalization steps
#' @export
#' @details Scaling data means that the standard deviation of a
#' variable is divided out of the data. `step_scale` estimates
#' the variable standard deviations from the data used in the
#' `training` argument of `prep.recipe`.
#' `bake.recipe` then applies the scaling to new data sets
#' using these standard deviations.
#'
#' # Tidying
#'
#' When you [`tidy()`][tidy.recipe()] this step, a tibble with columns
#' `terms` (the selectors or variables selected) and `value` (the
#' standard deviations) is returned.
#'
#' @template case-weights-unsupervised
#'
#' @references Gelman, A. (2007) "Scaling regression inputs by
#' dividing by two standard deviations." Unpublished. Source:
#' \url{http://www.stat.columbia.edu/~gelman/research/unpublished/standardizing.pdf}.
#' @examplesIf rlang::is_installed("modeldata")
#' data(biomass, package = "modeldata")
#'
#' biomass_tr <- biomass[biomass$dataset == "Training", ]
#' biomass_te <- biomass[biomass$dataset == "Testing", ]
#'
#' rec <- recipe(
#' HHV ~ carbon + hydrogen + oxygen + nitrogen + sulfur,
#' data = biomass_tr
#' )
#'
#' scaled_trans <- rec %>%
#' step_scale(carbon, hydrogen)
#'
#' scaled_obj <- prep(scaled_trans, training = biomass_tr)
#'
#' transformed_te <- bake(scaled_obj, biomass_te)
#'
#' biomass_te[1:10, names(transformed_te)]
#' transformed_te
#' tidy(scaled_trans, number = 1)
#' tidy(scaled_obj, number = 1)
step_scale <-
function(recipe,
...,
role = NA,
trained = FALSE,
sds = NULL,
factor = 1,
na_rm = TRUE,
skip = FALSE,
id = rand_id("scale")) {
add_step(
recipe,
step_scale_new(
terms = enquos(...),
role = role,
trained = trained,
sds = sds,
factor = factor,
na_rm = na_rm,
skip = skip,
id = id,
case_weights = NULL
)
)
}
step_scale_new <-
function(terms, role, trained, sds, factor, na_rm, skip, id, case_weights) {
step(
subclass = "scale",
terms = terms,
role = role,
trained = trained,
sds = sds,
factor = factor,
na_rm = na_rm,
skip = skip,
id = id,
case_weights = case_weights
)
}
#' @export
prep.step_scale <- function(x, training, info = NULL, ...) {
col_names <- recipes_eval_select(x$terms, training, info)
check_type(training[, col_names], types = c("double", "integer"))
wts <- get_case_weights(info, training)
were_weights_used <- are_weights_used(wts, unsupervised = TRUE)
if (isFALSE(were_weights_used)) {
wts <- NULL
}
if (x$factor != 1 & x$factor != 2) {
rlang::warn("Scaling `factor` should take either a value of 1 or 2")
}
vars <- variances(training[, col_names], wts, na_rm = x$na_rm)
sds <- sqrt(vars)
sds <- sd_check(sds)
sds <- sds * x$factor
step_scale_new(
terms = x$terms,
role = x$role,
trained = TRUE,
sds = sds,
factor = x$factor,
na_rm = x$na_rm,
skip = x$skip,
id = x$id,
case_weights = were_weights_used
)
}
#' @export
bake.step_scale <- function(object, new_data, ...) {
check_new_data(names(object$sds), object, new_data)
for (column in names(object$sds)) {
sd <- object$sds[column]
new_data[[column]] <- new_data[[column]] / sd
}
new_data
}
print.step_scale <-
function(x, width = max(20, options()$width - 30), ...) {
title <- "Scaling for "
print_step(names(x$sds), x$terms, x$trained, title, width,
case_weights = x$case_weights)
invisible(x)
}
#' @rdname tidy.recipe
#' @export
tidy.step_scale <- function(x, ...) {
if (is_trained(x)) {
res <- tibble(
terms = names(x$sds),
value = unname(x$sds)
)
} else {
term_names <- sel2char(x$terms)
res <- tibble(
terms = term_names,
value = na_dbl
)
}
res$id <- x$id
res
}
|
