What is the difference between M5 and Cubist packages?

Are the packages M5 and Cubist different? I trained my models with both using the caret::train function
and got significantly different results. I thought the two packages generate M5 model trees.

They are different models.

M5 is a model tree that approximates Quinlan's model tree method. It can be configured to use trees or convert the trees to rule-based models.

Cubist is not a model tree. It makes one then converts it to rules. It also has boosting and a neighbor-based correction and uses Quinlan's code.

Thank you so much for your reply, Max.
I will PM you.

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.