step_integer() creates a specification of a recipe step that will convert
new data into a set of integers based on the original data values.
step_integer(
recipe,
...,
role = "predictor",
trained = FALSE,
strict = TRUE,
zero_based = FALSE,
key = NULL,
skip = FALSE,
id = rand_id("integer")
)A recipe object. The step will be added to the sequence of operations for this recipe.
One or more selector functions to choose variables
for this step. See selections() for more details.
For model terms created by this step, what analysis role should they be assigned? By default, the new columns created by this step from the original variables will be used as predictors in a model.
A logical to indicate if the quantities for preprocessing have been estimated.
A logical for whether the values should be returned as integers (as opposed to double).
A logical for whether the integers should start at zero and new values be appended as the largest integer.
A list that contains the information needed to
create integer variables for each variable contained in
terms. This is NULL until the step is trained by
prep().
A logical. Should the step be skipped when the
recipe is baked by bake()? While all operations are baked
when prep() 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 = TRUE as it may affect
the computations for subsequent operations.
A character string that is unique to this step to identify it.
An updated version of recipe with the new step added to the
sequence of any existing operations.
step_integer will determine the unique values of
each variable from the training set (excluding missing values),
order them, and then assign integers to each value. When baked,
each data point is translated to its corresponding integer or a
value of zero for yet unseen data (although see the zero_based
argument above). Missing values propagate.
Factor inputs are ordered by their levels. All others are
ordered by sort.
Despite the name, the new values are returned as numeric unless
strict = TRUE, which will coerce the results to integers.
When you tidy() this step, a tibble is returned with
columns terms, value , and id:
character, the selectors or variables selected
list, a list column with the conversion key
character, id of this step
The underlying operation does not allow for case weights.
Other dummy variable and encoding steps:
step_bin2factor(),
step_count(),
step_date(),
step_dummy(),
step_dummy_extract(),
step_dummy_multi_choice(),
step_factor2string(),
step_holiday(),
step_indicate_na(),
step_novel(),
step_num2factor(),
step_ordinalscore(),
step_other(),
step_regex(),
step_relevel(),
step_string2factor(),
step_time(),
step_unknown(),
step_unorder()
data(Sacramento, package = "modeldata")
sacr_tr <- Sacramento[1:100, ]
sacr_tr$sqft[1] <- NA
sacr_te <- Sacramento[101:105, ]
sacr_te$sqft[1] <- NA
sacr_te$city[1] <- "whoville"
#> Warning: invalid factor level, NA generated
sacr_te$city[2] <- NA
rec <- recipe(type ~ ., data = sacr_tr) %>%
step_integer(all_predictors()) %>%
prep(training = sacr_tr)
bake(rec, sacr_te, all_predictors())
#> # A tibble: 5 × 8
#> city zip beds baths sqft price latitude longitude
#> <int> <int> <int> <int> <int> <int> <int> <int>
#> 1 NA 35 4 2 NA 0 0 0
#> 2 NA 62 3 2 0 0 0 0
#> 3 28 34 3 2 56 0 0 0
#> 4 34 51 3 1 0 0 0 0
#> 5 34 58 4 3 0 0 0 0
tidy(rec, number = 1)
#> # A tibble: 8 × 3
#> terms value id
#> <chr> <list> <chr>
#> 1 city <tibble [37 × 2]> integer_3IckW
#> 2 zip <tibble [68 × 2]> integer_3IckW
#> 3 beds <tibble [5 × 2]> integer_3IckW
#> 4 baths <tibble [4 × 2]> integer_3IckW
#> 5 sqft <tibble [94 × 2]> integer_3IckW
#> 6 price <tibble [95 × 2]> integer_3IckW
#> 7 latitude <tibble [99 × 2]> integer_3IckW
#> 8 longitude <tibble [99 × 2]> integer_3IckW