3
$\begingroup$

Reading documentation of Naive Bayes from sklearn, I read the following:

"On the flip side, although naive Bayes is known as a decent classifier, it is known to be a bad estimator, so the probability outputs from predict_proba are not to be taken too seriously."

What that does mean?

$\endgroup$

1 Answer 1

3
$\begingroup$

It sounds like it means that naive Bayes usually gets right whether the probability is > 0.5 or < 0.5 (or whatever cutoff you are using for classification) but often gets it far from correct within those ranges.

That seems odd to me, but I know little about naive Bayes.

$\endgroup$
0

Your Answer

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

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