Summary

This chapter built upon the work you did in Chapter 2. You changed out the dataset, switching the target from species of irises, which is categorical data, to housing value data, which is a continuous numeric variable. You built a regression model to predict house value given the various features (X parameters). As you did in Chapter 2, you built graphs to look at the data before building a model. You prepared the data to create test and training datasets and trained a regression model with XGBoost. Because it’s a regression model, this time, you used R2, RMSE, and residuals to evaluate the model.

So far, you’ve used XGBoost to build a classification model and, in this chapter, a regression model. The regression tree model created is a gradient-boosted tree model because you used the default settings in XGBoost. XGBoost is an ensemble tree method, and with it, you can build both gradient-boosted tree models and random forest models. The next chapter, Chapter...