Hi,

I'm trying to develop a model based on 20.000 records, and the phenomenon I want to predict occurs in about 0.5 to 1.5 % of the cases. When I try to run it through RDS, it starts up without any error messages, and I can see my CPU usage spike to 100%, but after a few seconds it just shuts down, the progress bar disappears and the program becomes inactive. Am I using to many records? Are the occurences I try to predict to rare? I think I'm using the right classifications, and even if I only use numeric values I still get the same problem.

Thanks a lot for any help you can offer!
Greetings,
Koen