"unused argument" error when calling `select` after applying the `bake` function

Question

I have a question regarding the bake function. My dataset consists of 2 variables: the dependent CHURN_FLAG and the independent amount of resources MT_RESOURCES.

When I perform the bake function, I get this error:

"Error in select(., -CHURN_FLAG) : unused argument (-CHURN_FLAG)"

This is the code I use:

MT_RESOURCES <- c(10, 20, 34, 5, 20, 9, 10, 60, 5, 1)
CHURN_FLAG <- c("YES","YES","YES","NO","YES","NO","YES","YES","NO","NO")
ddata <- data.frame(MT_RESOURCES, CHURN_FLAG)
glimpse(ddata)

Output:

Rows: 10
Columns: 2
$ MT_RESOURCES <dbl> 10, 20, 34, 49, 20, 44, 10, 60, 44, 1
$ CHURN_FLAG   <chr> "YES", "NO", "YES", "YES", "NO", "NO", "NO", "YES", "NO", "NO"

Code:

rec_obj <- recipe(CHURN_FLAG ~ ., data = ddata) %>%
  step_center(all_predictors(), -all_outcomes()) %>%
  step_scale(all_predictors(), -all_outcomes()) %>%
  prep(data = ddata)

Output:

      role #variables
   outcome          1
 predictor          1

Training data contained 10 data points and no missing data.

Operations:

• Centering for MT_RESOURCES [trained]
• Scaling for MT_RESOURCES [trained]

Code:

x_ddata <- bake(rec_obj, newdata = ddata) %>% select(-CHURN_FLAG)

Output:

Error in select(., -CHURN_FLAG) : unused argument (-CHURN_FLAG)

Answer 1

You may have another package like MASS masking the Select function. Try specify the package the function comes from.

dplyr::select(-CHURN_FLAG) 

Answer 2

You need new_data with the underscore.