I'd like to use a custom function (a linear combination of attributes in my case) as a model in RapidMiner to compare with other models. To be specific, I want to use the "default" model in cross-validation to compare with other models with the same settings. How can I accomplish that?
I'd like a model learner operator that discards all inputs and output a user specified function as a model. Is it possible?
Thanks in advance.
Yes, absolutely possible. Rapidminer allows you to create your own extension, and operator.
https://docs.rapidminer.com/developers/creating-your-own-extension/