tests/testthat/_snaps/YeoJohnson.md

In tidymodels/recipes: Preprocessing and Feature Engineering Steps for Modeling

missing data

Code
  prep(rec_false, training = ex_dat, verbose = FALSE)
Condition
  Error in `step_YeoJohnson()`:
  Caused by error in `prep()`:
  x Missing values are not allowed for the YJ transformation.
  i See `na_rm` option.

bake method errors when needed non-standard role columns are missing

Code
  bake(rec_trained, new_data = ex_dat[, 1:2])
Condition
  Error in `step_YeoJohnson()`:
  ! The following required column is missing from `new_data`: x4.

empty printing

Code
  rec
Message

  -- Recipe ----------------------------------------------------------------------

  -- Inputs 
  Number of variables by role
  outcome:    1
  predictor: 10

  -- Operations 
  * Yeo-Johnson transformation on: <none>

Code
  rec
Message

  -- Recipe ----------------------------------------------------------------------

  -- Inputs 
  Number of variables by role
  outcome:    1
  predictor: 10

  -- Training information 
  Training data contained 32 data points and no incomplete rows.

  -- Operations 
  * Yeo-Johnson transformation on: <none> | Trained

printing

Code
  print(rec)
Message

  -- Recipe ----------------------------------------------------------------------

  -- Inputs 
  Number of variables by role
  predictor: 4

  -- Operations 
  * Yeo-Johnson transformation on: x1, x2, x3, x4

Code
  prep(rec)
Message

  -- Recipe ----------------------------------------------------------------------

  -- Inputs 
  Number of variables by role
  predictor: 4

  -- Training information 
  Training data contained 20 data points and no incomplete rows.

  -- Operations 
  * Yeo-Johnson transformation on: x1, x2, x4 | Trained

bad args

Code
  recipe(~., data = ex_dat) %>% step_YeoJohnson(x1, x2, x3, x4, na_rm = "yes") %>%
    prep()
Condition
  Error in `step_YeoJohnson()`:
  Caused by error in `prep()`:
  ! `na_rm` must be `TRUE` or `FALSE`, not the string "yes".

Code
  recipe(~., data = ex_dat) %>% step_YeoJohnson(x1, x2, x3, x4, num_unique = "yes") %>%
    prep()
Condition
  Error in `step_YeoJohnson()`:
  Caused by error in `prep()`:
  ! `x$num_unique` must be a whole number, not the string "yes".

Code
  recipe(~., data = ex_dat) %>% step_YeoJohnson(x1, x2, x3, x4, limits = NA_real_) %>%
    prep()
Condition
  Error in `step_YeoJohnson()`:
  Caused by error in `prep()`:
  ! `limits` should be a numeric vector with two values, not a double vector

tidymodels/recipes documentation built on Jan. 25, 2025, 7:30 a.m.