Decision Tree implementation (both classification and regression) following the algorithm proposed in the PLANET paper (http://static.googleusercontent.com/external_content/untrusted_dlcp/research.google.com/en/us/pubs/archive/36296.pdf)
Data Generator with numeric and categorical variables
Unit / Regression tests leveraging this data generator
We are looking forward into it.
We have implemented a couple of versions of the decision tree algorithms and have performed testing to see where additional improvements can be made. Based on the earlier implementations, we have finalized the architecture and I am working on the final version of the code at present. I am expecting a PR for this feature in January.
Any update on this?