Do you need to bake testing data prior to predicting tidymodels/stacks?

tidyverse
,
1

I am using a tidymodels workflow with a custom recipe to tune multiple models. These models are then fed into tidymodels/stacks for blending and fitting. For example:

  # Create the base stacked model
  stacked_models <- 
    stacks() %>%
    add_candidates(tuned_model) 
  
  # Blend all of the model components
  blended_stack <-
    stacked_models %>%
    blend_predictions(
      control = tune::control_grid(allow_par = TRUE))
  
  # Produce the Final Stacked Model
  fit_stacked_model <-
    blended_stack %>%
    fit_members()

When predicting on the fit_stacked_model object, does the testing data need to be prep() and bake() prior to predicting, or does stacks account for that in the backend?

2

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.