I like to use ClassifierMeasurements. It's nice and doesn't make me think too hard. One of the features I like a lot is the Examples which lets me easily inspect misclassified examples. Another feature I really like is the IndeterminateThreshold option. However, when using the IndeterminateThreshold option I can't seem to figure out how to extract "IndeterminateExamples". For example I can get three 9's accuracy with IndeterminateThreshold -> 0.89774 and there are 43 samples in each class that are Indeterminate. If I wanted to know which sample was misclassified as 0 I could enter something like cm["Examples" -> {1, 0}].

So then, is there a way to ask a ClassifierMeasurement object for "IndeterminateExamples"?

enter image description here

P.S. I realize I can calculate this directly but the question is if it is possible to ask ClassifierMeasurements for this info. It is clear from the confusion matrix plot that ClassifierMeasurements figured out which samples where Indeterminate so how do I access that information?


1 Answer 1


While Indeterminate is not a class per se, looks like it can be used with the Examples option just like a class label. I was able to get the indeterminate samples using

cm["Examples" -> {1, Indeterminate}]


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.