Classification and Regression Trees, Ensembles, and Deep Learning Models – What’s Best for Your Data?
This chapter covers when to use ensemble models or simple classification and regression tree models. Through experimenting with the housing dataset that we used in Chapter 4, you will see the differences between models in terms of model accuracy, as measured by R2 and RMSE, how long it takes for the models to be trained, and how long it takes for the models to make predictions. To do so, you will test and compare XGBoost, scikit-learn gradient boosting, and random forest models. You will learn when to use decision-tree-based machine learning versus deep learning. Lastly, you will learn how to set the various hyperparameters (parameters that control how the model learns) for XGBoost.
In this chapter, we will cover the following main topics:
- When to use ensemble models versus single Classification and Regression Tree (CART) models
- Comparing models with the...