How come there is a difference in the Model and the Prediction in a decision tree made by automodel?

User132127User132127 Member Posts: 1 Contributor I
Hi everyone,
I am training a model for classic churn prediction. I would like to get a decision tree made by the a feature.
What I don't understand is, that it creates a model/tree, but the predictions do not match this tree.
As you can see we have a customer (number 12) which was predicted as 'Kündiger' he has a LZ Abo bis Kündigung M of  25, 0 Reklatage_180T. So he should fall right in to the most common path and be classified as 'Kleiber'. Also looked if Model or Production Model differ, but they don'T. So can someone please explain to me how to read the tree or the prediction correctly?
Sign In or Register to comment.