This question already has an answer here:

Is it appropriate to conduct feature selection (e.g., Recursive Feature Selection) on a data set IN ADVANCE of model fitting to scale down features for more expedient machine learning model fitting?

I have some resources that suggest that. When I've tried it it has hurt overall model performance.

Insight please


marked as duplicate by Sycorax, Peter Flom Apr 1 at 11:32

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.