XGBoost modeling time
XGBoost(Extreme GradientBoost) model is a special gradient lifting decision tree (GBDT), and XGBoost is an open source framework for gradient lifting created by Dr. Chen Tianqi of Washington University. It just tries to maximize speed and efficiency, so it is called X(Extreme)gradientboost.
Xgboost is essentially a method based on tree structure and ensemble learning, and its basic tree structure is CART (Classified Regression Tree). Similar to the local weighted linear regression algorithm, the tree-based regression algorithm is also a local regression algorithm. By dividing the data set into multiple parts, the model is built separately on each data. But the difference is that the tree-based regression algorithm is a parameter-based learning algorithm. After training the model with training data, once the parameters are determined, there is no need to change them. Classified regression tree is a structure based on decision tree, which can be used to solve both classification problems and regression problems. It is one of the top ten classic algorithms in the field of data mining selected by the international authoritative academic organization IEEE International Conference on Data Mining (ICDM) earlier.