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.
Asked
Active
Viewed 38 times
