I am using MadLibs linear regression train function to analyze coef
. While doing so I came across baseline referencing for independent variables.
Something similar to relevel()
in R. How to force R to use a specified factor level as reference in a regression?
Does MadLib provide any feature which could be used to achieve this?
Or if Madlib implements/run it internally with linregr_train ?
I couldn't find the parameter where I could specify a baseline for independent variables on Madlib.
By baseline reference, do you mean the reference for dummy encoding of a categorical independent variable? If yes, then MADlib expects that the data has been encoded before passing it to the regression methods. SeeĀ Encoding Categorical Variables for the procedure to dummy encode. The
value_to_drop
parameter provides the functionality for user-defined reference level.