Custom preprocessing in caret

Question

I use caret for my Machine Learning pipelines. When I have to perform some preprocessing that is not avaliable in the preProcess function, is it possible to create a custom function that can be used instead of the default methods?

Answer 1

Not yet, but I'm 80% done with an interface to caret that uses recipes as the model specification tool. That would allow you to create custom preprocessing methods.

If you want to try it, use the recipe_integration branch of caret on github.