split results into bins for evaluation
Another fun puzzle for the RM team.
I've run a model that outputs confidence values for an SVM class. The values range from 0 to 1 (as expected for this type of model.)
One very common method of evaluation I've seen in papers is to break the results into "bins" or "groups" by confidence range and then report the accuracy of each range.
|Range||# predicted||# correct||% correct|