Deriving the response function for chosen model

[PFerrara3q]

i am interested in the possibility of deriving the final response function of a chosen model. Is there a way to obtain this model and the associated hyperparamters to be used in one-off scenarios where i can further optimize with constraints and/or just test different variables in the underlying best model?

Any thoughts or suggestions would be greatly appreciated. Cheers