My goal with this PR is to simplify some of the existing code to make it easier for me to make improvements. I tried to use scikit-learn's structure to guide changes (so that it might be more easily merged down the road). I noticed that the boston dataset's target is a numeric variable, so I searched for a good ordinal categorial dataset to replace boston. I found one called the Copenhagen Housing Survey in the MASS R package. The MASS package (library) is a very stable and old package (2002) created by three R core developers and maintained by a current member. R ships with built-in datasets. Ideally, I could add another example(s) from that set (I have yet to find a good ordinal dataset there). I'd also like to add an example from UCI if I can find one.
I'm an expert R programmer, but somewhat new to Python, critiques and guidance welcome. :)
My goal with this PR is to simplify some of the existing code to make it easier for me to make improvements. I tried to use scikit-learn's structure to guide changes (so that it might be more easily merged down the road). I noticed that the
boston
dataset'starget
is a numeric variable, so I searched for a good ordinal categorial dataset to replaceboston
. I found one called theCopenhagen Housing Survey
in theMASS
R package. TheMASS
package (library) is a very stable and old package (2002) created by three R core developers and maintained by a current member. R ships with built-in datasets. Ideally, I could add another example(s) from that set (I have yet to find a good ordinal dataset there). I'd also like to add an example from UCI if I can find one.I'm an expert R programmer, but somewhat new to Python, critiques and guidance welcome. :)