Predictor Bin Snapshot
We noticed that a predictor performance is different for every bin and new entry is written into snapshot table for every bin.
But when we see the ADM model monitoring, how does pega defines the active/inactive predictors?
Example: If a symbolic predictor Gender has two bins. It was written twice into snapshot table for every bin. The performance of the predictor is different for every bin. If the performance of the predictor bin is beyond AUC value(ADM advanced settings), it will be an active predictor. Now, lets say in one bin its performance is high but in another bin its performance is low. What will be the overall status of the Predictor? Is it active or inactive?
Let me know if my understanding is right. I am trying to understand how predictor bins, predictor performance and active/inactive status are related to each other.