I'm practicing feature selection. I got stuck in getting feature rank and profit factor at the same time.

How can I get feature rank/importance while using advise(NEURAL) with simple linear regression like lm() in each WFO cycle?

Thanks.