Re-running the program and overfitting
Does re-running the program until I find the best metrics lead to my algorithm overfitting?
Does re-running the program until I find the best metrics lead to my algorithm overfitting?
Hi @oksifoxy
Re-running an algorithm does not lead to over-fitting. It's okay to re-run and select the best model. But when you try to compare two algorithms you must average over large enough number of re-runs.
Cheers!
Just drop us an email to ... Show more