decision tree viewer


I would like to know why the number of rows are not shown in the decision trees that compose the ensemble of the Radom Forest or Tree ensemble models. By training a simple Decision tree model, the decision tree view shows the number of rows by category in the image. However, in the tree view of the Tree ensemble learner or Random Forest learner the image of their individual trees shows 1 or 0 for the number of rows by category in all nodes of the tree. I am attaching an example from the adult.cvs dataset to compare the single decision tree provided by the Decision Tree Learner node and one of the trees provided by the Random Forest Learner node.

Thank you.

It is by design in order to handle memory issues. If you check the second option you’ll get what you want:


This topic was automatically closed 182 days after the last reply. New replies are no longer allowed.