What percentage of shadow features does each real feature outperform?

scikit-learn-contrib / boruta_py

Python implementations of the Boruta all-relevant feature selection method.

BSD 3-Clause "New" or "Revised" License

1.46k stars 252 forks source link

What percentage of shadow features does each real feature outperform? #112

Open rwjmiller opened 1 year ago

rwjmiller commented 1 year ago

It would be useful to be able to capture this to get a relative sense of feature importance. If I understand correctly, it would also enable one to see the impact of setting different values for perc without having to actually re-run the whole feature selection process for each value of perc.