Print a nested cross-validation result

Printing a nested_results shows its outer folds as tibble rows, the outer resampling scheme, and how many folds did not complete.

Usage

# S3 method for class 'nested_results'
print(x, ..., n = NULL, width = NULL)

Arguments

x

A nested_results from nested_tune_grid() or a related function.

...

Not used, and must be empty. Give n and width by name.

n

Number of fold rows to show, passed to tibble's printing. NULL, the default, leaves the choice to tibble. Inf shows every fold.

width

Width of the printed rows, passed to tibble's printing.

Value

x, invisibly.

Details

When the completed folds scored different sets of candidates, the print adds each fold's candidate count. A grid given as a size is the usual cause, because each fold builds its own grid under its own seed. Folds that disagree then did not choose from the same candidates.

summary.nested_results() shows the selections and the nested estimate. The nested estimate describes the tuning procedure, not a fitted model.

See also

summary.nested_results(), nested_tune_grid(), collect_metrics()

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))
res
#> 
#> ── Nested cross-validation results ────────────────────────────────────
#> Outer resamples: 2-fold cross-validation
#> # A tibble: 2 × 9
#>   splits          id    .metrics .selected .inner_metrics   .notes  
#>   <list>          <chr> <list>   <list>    <list>           <list>  
#> 1 <split [16/16]> Fold1 <tibble> <tibble>  <tibble [4 × 7]> <tibble>
#> 2 <split [16/16]> Fold2 <tibble> <tibble>  <tibble [4 × 7]> <tibble>
#> # ℹ 3 more variables: .completed <lgl>, .tuning_seed <int>,
#> #   .outer_fit_seed <int>
#> ℹ Use `summary()` for what the run means: which folds failed, what
#>   each one selected, and the estimate across them.