NEW R package: The XGBoost Explainer – Applied Data Science – Medium

In this post I'm going to try to do three things: Code and links to the package are included at the bottom of this post. A decision tree is fully interpretable. The coefficients or branches of the model tells you the'why' of each prediction. For example, take the following decision tree, that predicts the likelihood of an employee leaving the company. Predictions made using this tree are entirely transparent -- i.e. you can say exactly how each feature has influenced the prediction.