novel.md
In recipes: Preprocessing and Feature Engineering Steps for Modeling

bad args

Code
  recipe(~., data = iris) %>% step_novel(all_predictors()) %>% prep(iris)
Condition
  Error in `step_novel()`:
  Caused by error in `prep()`:
  ! All columns selected for the step should be string, factor, or ordered.

Code
  recipe(~., data = tr_bad) %>% step_novel(all_predictors()) %>% prep(tr_bad)
Condition
  Error in `step_novel()`:
  Caused by error in `prep()`:
  ! Columns already contain the new level: x

empty printing

Code
  rec
Message

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

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

  -- Operations 
  * Novel factor level assignment 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 
  * Novel factor level assignment for: <none> | Trained

printing

Code
  print(rec)
Message

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

  -- Inputs 
  Number of variables by role
  predictor: 5

  -- Operations 
  * Novel factor level assignment for: all_predictors()

Code
  prep(rec)
Message

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

  -- Inputs 
  Number of variables by role
  predictor: 5

  -- Training information 
  Training data contained 200 data points and no incomplete rows.

  -- Operations 
  * Novel factor level assignment for: v, w, x, y, z | Trained