[Author: Changjin Liu] This project is for the NYC Taxi Travel Duration Prediction Kaggle Competition, integrating several regression machine learning methods such as Gradient Boosting Regressor, XGBoost, LightGBM et al, and the result of 287.6 RMSE gets the 5th / 72 place.