class-roc_dist.md
In yardstick: Tidy Characterizations of Model Performance

work with class_pred input

Code
  roc_dist_vec(cp_truth, cp_estimate)
Condition
  Error in `roc_dist_vec()`:
  ! `truth` should not a <class_pred> object.

na_rm argument check

Code
  roc_dist_vec(1, 1, na_rm = "yes")
Condition
  Error in `roc_dist_vec()`:
  ! `na_rm` must be `TRUE` or `FALSE`, not the string "yes".

Binary returns `NA` with a warning when results are undefined (#98)

Code
  out <- roc_dist_vec(truth, estimate)
Condition
  Warning:
  While computing binary `sens()`, no true events were detected (i.e. `true_positive + false_negative = 0`).
  Sensitivity is undefined in this case, and `NA` will be returned.
  Note that 1 predicted event(s) actually occurred for the problematic event level, a

Code
  out <- roc_dist_vec(truth, estimate)
Condition
  Warning:
  While computing binary `spec()`, no true negatives were detected (i.e. `true_negative + false_positive = 0`).
  Specificity is undefined in this case, and `NA` will be returned.
  Note that 1 predicted negatives(s) actually occurred for the problematic event level, a