Prevent LOO through vfold_cv()

[hfrick]

We purposefully don't (fully) support LOO cross-validation in tidymodels: we have loo_cv() in rsample but it's not supported in tune.

However, you can create LOO cross-validation by doing:

vfold_cv(df,  v = nrow(df))

which then does not get caught in tune. Let's make vfold_cv() error for this case, with a reference to loo_cv().

library(tidymodels)

loo_folds <- loo_cv(mtcars)
v_folds <- vfold_cv(mtcars, v = nrow(mtcars))

wflow <- workflow(mpg ~ ., linear_reg())

fit_resamples(wflow, resamples = loo_folds)
#> Error in `check_rset()` at tune/R/tune_grid.R:334:2:
#> ! Leave-one-out cross-validation is not currently supported with tune.

fit_resamples(wflow, resamples = v_folds)
#> → A | warning: A correlation computation is required, but the inputs are size zero or one and the standard deviation cannot be computed. `NA` will be returned.
#> There were issues with some computations   A: x1
#> There were issues with some computations   A: x32
#> 
#> # Resampling results
#> # 32-fold cross-validation 
#> # A tibble: 32 × 4
#>    splits         id     .metrics         .notes          
#>    <list>         <chr>  <list>           <list>          
#>  1 <split [31/1]> Fold01 <tibble [2 × 4]> <tibble [1 × 3]>
#>  2 <split [31/1]> Fold02 <tibble [2 × 4]> <tibble [1 × 3]>
#>  3 <split [31/1]> Fold03 <tibble [2 × 4]> <tibble [1 × 3]>
#>  4 <split [31/1]> Fold04 <tibble [2 × 4]> <tibble [1 × 3]>
#>  5 <split [31/1]> Fold05 <tibble [2 × 4]> <tibble [1 × 3]>
#>  6 <split [31/1]> Fold06 <tibble [2 × 4]> <tibble [1 × 3]>
#>  7 <split [31/1]> Fold07 <tibble [2 × 4]> <tibble [1 × 3]>
#>  8 <split [31/1]> Fold08 <tibble [2 × 4]> <tibble [1 × 3]>
#>  9 <split [31/1]> Fold09 <tibble [2 × 4]> <tibble [1 × 3]>
#> 10 <split [31/1]> Fold10 <tibble [2 × 4]> <tibble [1 × 3]>
#> # ℹ 22 more rows
#> 
#> There were issues with some computations:
#> 
#>   - Warning(s) x32: A correlation computation is required, but the inputs are size ze...
#> 
#> Run `show_notes(.Last.tune.result)` for more information.

^{Created on 2023-07-12 with reprex v2.0.2}

[github-actions]

This issue has been automatically locked. If you believe you have found a related problem, please file a new issue (with a reprex: https://reprex.tidyverse.org) and link to this issue.