tests/testthat/_snaps/class.md

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

bake_check_class helper function gives expected output

Code
  bake_check_class_core(x1, "character", "x1")
Condition
  Error in `bake_check_class_core()`:
  ! `x1` should have the class <character> but has the class <numeric>.

Code
  bake_check_class_core(x2, c("POSIXct", "Julian"), "x2")
Condition
  Error in `bake_check_class_core()`:
  ! `x2` should have the class <POSIXct/Julian> but has the classes <POSIXct/POSIXt>.

Code
  bake_check_class_core(x2, "POSIXct", "x2")
Condition
  Error in `bake_check_class_core()`:
  x `x2` has class <POSIXct/POSIXt> but only the following are asked: <POSIXct>.
  i This error is shown because `allow_additional` is set to "FALSE".

check_class works when class is learned

Code
  bake(rec1, x_newdata)
Condition
  Error:
  ! `x1` should have the class <numeric> but has the class <character>.

Code
  bake(rec1, x_newdata_2)
Condition
  Error:
  x `x2` has class <POSIXct/POSIXt/Julian> but only the following are asked: <POSIXct/POSIXt>.
  i This error is shown because `allow_additional` is set to "FALSE".

check_class works when class is provided

Code
  bake(rec2, x_newdata)
Condition
  Error:
  ! `x1` should have the class <numeric> but has the class <character>.

Code
  bake(rec3, x_newdata_2)
Condition
  Error:
  x `x2` has class <POSIXct/POSIXt/Julian> but only the following are asked: <POSIXct/POSIXt>.
  i This error is shown because `allow_additional` is set to "FALSE".

characters are handled correctly

Code
  bake(rec6_NULL, sacr_fac[11:20, ])
Condition
  Error:
  ! `city` should have the class <factor> but has the class <character>.

Code
  bake(rec6_man, sacr_fac[11:20, ])
Condition
  Error:
  ! `type` should have the class <factor> but has the class <character>.

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

Code
  bake(rec_trained, new_data = x[, -1])
Condition
  Error in `check_class()`:
  ! The following required column is missing from `new_data`: x1.

empty printing

Code
  rec
Message

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

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

  -- Operations 
  * Checking the class(es) for: <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 
  * Checking the class(es) for: <none> | Trained

printing

Code
  print(rec7)
Message

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

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

  -- Operations 
  * Checking the class(es) for: all_predictors()

Code
  prep(rec7)
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 
  * Checking the class(es) for: cyl, disp, hp, drat, wt, qsec, ... | Trained

bad args

Code
  recipe(mpg ~ ., mtcars) %>% check_class(all_predictors(), class_nm = 1)
Condition
  Error:
  ! `class_nm` must be a character vector or `NULL`, not the number 1.

Code
  recipe(mpg ~ ., mtcars) %>% check_class(all_predictors(), allow_additional = "yes")
Condition
  Error:
  ! `allow_additional` must be `TRUE` or `FALSE`, not the string "yes".

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