The sklearn's documentation of the method roc_auc_score
states that the parameter multi_class
can take the value 'OvR'
(which stands for One-vs-Rest) or 'OvO'
(which stands for One-vs-One). These values are only applicable for multi-class classification problems.
Does anyone know in what particular cases we would use OvR as opposed to OvO? In the general academic literature, is there a preference given to one?