Skip to content

Produce a tibble of identification variables so that single splits can be linked to a particular resample.

Usage

# S3 method for class 'rsplit'
labels(object, ...)

Arguments

object

An rsplit object

...

Not currently used.

Value

A tibble.

See also

add_resample_id

Examples

cv_splits <- vfold_cv(mtcars)
labels(cv_splits$splits[[1]])
#> # A tibble: 1 × 1
#>   id    
#>   <chr> 
#> 1 Fold01