What is the function to measure the quality of a split in this xgboost implementation?
I didn’t find this info in the documentation.
In SKlearn, I know that the hyperparameter ‘criterion’ may be set to use a specific measure for the splitting.
XGBoost uses gradient boosting, where the gradient (partial derivative) of the loss function is used as a measure of quality of splits. See my master’s thesis at https://drive.google.com/file/d/0B0c0MbnP6Nn-eUNRRkVOOGpkbFk/view?usp=sharing&resourcekey=0-nVw3WhovKW5FPvPM5GPHfg.