Predicting from a nested run

predict() on a nested run gives an error that points to nested_final_fit(). The outer folds' models produce the estimate, and none of them is the model to deploy.

Usage

# S3 method for class 'nested_results'
predict(object, ...)

# S3 method for class 'nested_results_set'
predict(object, ...)

Arguments

object

A nested_results or a nested_results_set.

...

Not used.

Value

None. The call is an error.

Details

nested_final_fit() fits the model to deploy on all the data. For a workflow set, its id argument picks the workflow.

See also

nested_final_fit(), predict.nested_final_fit()

Examples

data(mtcars)

rec <- recipes::recipe(mpg ~ ., data = mtcars) |>
  recipes::step_pca(recipes::all_predictors(), num_comp = tune::tune())
wf <- workflows::workflow(rec, parsnip::linear_reg())

set.seed(1)
folds <- nested_resamples(
  mtcars,
  outside = rsample::vfold_cv(v = 2),
  inside = rsample::vfold_cv(v = 2)
)
set.seed(2)
res <- nested_tune_grid(wf, folds, grid = data.frame(num_comp = 1:2))
set.seed(3)
final <- nested_final_fit(wf, res)
# Predict from the final fit, not from `res`.
predict(final, new_data = mtcars[1:3, ])
#> # A tibble: 3 × 1
#>   .pred
#>   <dbl>
#> 1  23.1
#> 2  23.1
#> 3  25.2