I know that f_classif method is mostly used for scoring numerical features and categorical targets. But I'm wondering if it's possible to score features coded by integers or binary codes against categorical targets - as the data below:

Feature = [0, 1, 0, 2, 1, 2, ...]

Target = [1, 0, 0, 1, 1, 1, ...]

Also, I have another question - if data includes mix of both categorical and numerical features, what's the best method to score their importance against categorical targets?



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.