I'd like to use decision tree / forest but I need to use a special objective function that can't be differentiated and hence I can't use XGBOost etc. That leaves a genetic algorithm where I use the fitness function to incorporate my special objective.

Other then building everything from scratch, are there any python packages that will allow me to do this easily?

  • $\begingroup$ autograd maybe, can differentiate anything $\endgroup$
    – Nikos M.
    Dec 26 '20 at 18:07

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.