I know this is a basic problem, but still could not find answer, I feel like most books/tutorials avoid talking about scaling the output feature instead they just mention scaling input features.

So my question is: in Machine/Deep Learning regression problems should the output (target feature) be scaled(for example by using MinMax or Standard scaler) or not?

And if scaling should be applied only to some ML/DL architectures, than to which ones?

  • 2
    $\begingroup$ In the regression problem, you should scale in the output. All DL architectures need scaling, but ML architectures like decision tree, random forest do not need scaling, but if you do not know which one needs scaling or not, it's best to just apply scaling. $\endgroup$ – Swakshar Deb May 12 '20 at 8:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.