New Python ML package with 3x faster XGBoost


#1

Hi all,
I am working on a new ML package in pure Python:

It contains at the moment a high performance implementation of XGBoost logistic algorithm and innovative treatment of categorical data.

Any feedback and contributions are welcome!

See demo notebooks:
https://www.kaggle.com/neostat/cortado-quick-start
https://www.kaggle.com/neostat/cortado-vs-xgboost

Cheers
Adam