I want to set up a process for data drift and I am trying to see which metric to pick. Using KS test, it flags half of my features to be drifted although when I look at the distribution between the two datasets, I dont see much of variation. PSI only flags one feature and thats the only one that is obviously visible to be drifted between the two sets. I wonder why this behavior is happening? Is it something related to my data that KS is behaving very aggressively? I tried very small alphas and still same situation happens.

  • $\begingroup$ What test do you use with PSI? $\endgroup$
    – Ben Reiniger
    Apr 13, 2023 at 1:14


Your Answer

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

Browse other questions tagged or ask your own question.