Boosting is a way to get a Strong Learner by combining a group of Weak Learners and trying to improve each error. When using Weak Learner as a Boosting Base Learner, in addition to being able to quickly train many models to combine, the low complexity of Weak Learner also makes the final combined Strong Learner have good resistance to Overfitting. ORAI uses GradientBoostingRegressor in scikit-learn as a tool to implement GradientBoosting-Regression. Users only need to prepare data and complete the data analysis through the ORAI process. This solution uses Boston house price dataset as training data.