I have a Regression problem solved with both Deep Feedforward NN (in keras) and Random Forest (sklearn). The results are pretty good. Now I would like to know if I can change the importance of only one feature (among the 41) so that the algorithms would learn better the relation between such feature and the output proving a way to be able to generalize even in case such feature change it parameter ranges.
is it possible in keras and sklearn? and also could it make sense?
question from:https://stackoverflow.com/questions/65934767/change-the-importance-of-a-feature-in-the-dataset-in-nn-and-rf