step_outliers_univariate creates a specification of a recipe
step that will calculate the outlier score using the user supplied outlier_function on a univariate basis
the combination is of multiple variables is handled by the combination_function
step_outliers_univariate(
recipe,
...,
role = NA,
trained = FALSE,
outlier_score = NULL,
columns = NULL,
name_mutate = ".outliers_univariate",
outlier_function = function(x) {
abs((percent_rank(x) - 0.5)) * 2
},
combination_function = mean,
skip = TRUE,
id = rand_id("outliers_univariate")
)
# S3 method for step_outliers_univariate
tidy(x, ...)A recipe object. The step will be added to the sequence of operations for this recipe.
One or more selector functions to choose which
variables will be transformed. See selections() for
more details. For the tidy method, these are not
currently used.
not defined for this function
A logical to indicate if the quantities for preprocessing have been estimated.
a placeholder for the exit of this function don't change
A character string of variable names that will be populated (eventually) by the terms argument.
the name of the generated column with univariate scores
the function used on a column basis to calculate the outlier score.
the function used on a row basis to combine different outlier function results.
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.
A step_outliers_univariate object.
An updated version of recipe with the new step
added to the sequence of existing steps (if any), with the name on name_mutate and the scores calculated. For the
tidy method, a tibble with columns index (the row indexes of the tibble) and outlier_score (the scores).
All columns in the data are sampled and returned by juice()
and bake().
All columns used in this step must be numeric with no missing data.
When used in modeling, users should strongly consider using the
option skip = TRUE so that this operation is not
conducted outside of the training set.
library(recipes)
library(tidy.outliers)
rec <-
recipe(mpg ~ ., data = mtcars) %>%
step_outliers_univariate(all_numeric_predictors()) %>%
prep(mtcars)
bake(rec, new_data = NULL)
#> # A tibble: 32 × 12
#> cyl disp hp drat wt qsec vs am gear carb mpg .outliers…¹
#> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 6 160 110 3.9 2.62 16.5 0 1 4 4 21 0.381
#> 2 6 160 110 3.9 2.88 17.0 0 1 4 4 21 0.335
#> 3 4 108 93 3.85 2.32 18.6 1 1 4 1 22.8 0.497
#> 4 6 258 110 3.08 3.22 19.4 1 0 3 1 21.4 0.503
#> 5 8 360 175 3.15 3.44 17.0 0 0 3 2 18.7 0.555
#> 6 6 225 105 2.76 3.46 20.2 1 0 3 1 18.1 0.613
#> 7 8 360 245 3.21 3.57 15.8 0 0 3 4 14.3 0.639
#> 8 4 147. 62 3.69 3.19 20 1 0 4 2 24.4 0.490
#> 9 4 141. 95 3.92 3.15 22.9 1 0 4 2 22.8 0.535
#> 10 6 168. 123 3.92 3.44 18.3 1 0 4 4 19.2 0.265
#> # … with 22 more rows, and abbreviated variable name ¹.outliers_univariate
tidy(rec, number = 1)
#> # A tibble: 32 × 3
#> index outlier_score id
#> <int> <dbl> <chr>
#> 1 1 0.381 outliers_univariate_LnzXx
#> 2 2 0.335 outliers_univariate_LnzXx
#> 3 3 0.497 outliers_univariate_LnzXx
#> 4 4 0.503 outliers_univariate_LnzXx
#> 5 5 0.555 outliers_univariate_LnzXx
#> 6 6 0.613 outliers_univariate_LnzXx
#> 7 7 0.639 outliers_univariate_LnzXx
#> 8 8 0.490 outliers_univariate_LnzXx
#> 9 9 0.535 outliers_univariate_LnzXx
#> 10 10 0.265 outliers_univariate_LnzXx
#> # … with 22 more rows