tune: Tidy Tuning Tools

check_predictions <- function(split, pred, tune_df) {
  assess <- rsample::assessment(split)
  n_te <- nrow(assess)
  n_pm <- nrow(tune_df)
  ind_te <- as.integer(split, data = "assessment")
  expect_true(tibble::is_tibble(pred))
  expect_equal(nrow(pred), n_te * n_pm)
  exp_nms <- c(".pred", ".row", names(tune_df), "mpg", ".config")
  expect_equal(names(pred), exp_nms)
  expect_equal(sort(unique(ind_te)), sort(unique(pred$.row)))
  TRUE
}