Book Image

Hands-On Gradient Boosting with XGBoost and scikit-learn

By : Corey Wade
Book Image

Hands-On Gradient Boosting with XGBoost and scikit-learn

By: Corey Wade

Overview of this book

XGBoost is an industry-proven, open-source software library that provides a gradient boosting framework for scaling billions of data points quickly and efficiently. The book introduces machine learning and XGBoost in scikit-learn before building up to the theory behind gradient boosting. You’ll cover decision trees and analyze bagging in the machine learning context, learning hyperparameters that extend to XGBoost along the way. You’ll build gradient boosting models from scratch and extend gradient boosting to big data while recognizing speed limitations using timers. Details in XGBoost are explored with a focus on speed enhancements and deriving parameters mathematically. With the help of detailed case studies, you’ll practice building and fine-tuning XGBoost classifiers and regressors using scikit-learn and the original Python API. You'll leverage XGBoost hyperparameters to improve scores, correct missing values, scale imbalanced datasets, and fine-tune alternative base learners. Finally, you'll apply advanced XGBoost techniques like building non-correlated ensembles, stacking models, and preparing models for industry deployment using sparse matrices, customized transformers, and pipelines. By the end of the book, you’ll be able to build high-performing machine learning models using XGBoost with minimal errors and maximum speed.
Table of Contents (15 chapters)
Section 1: Bagging and Boosting
Section 2: XGBoost
Section 3: Advanced XGBoost

Random forest hyperparameters

The range of random forest hyperparameters is large, unless one already has a working knowledge of decision tree hyperparameters, as covered in Chapter 2, Decision Trees in Depth.

In this section, we will go over additional random forest hyperparameters before grouping the hyperparameters that you have already seen. Many of these hyperparameters will be used by XGBoost.


Our first hyperparameter, and perhaps the most intriguing, is oob_score.

Random forests select decision trees via bagging, meaning that samples are selected with replacement. After all of the samples have been chosen, some samples should remain that have not been chosen.

It's possible to hold back these samples as the test set. After the model is fit on one tree, the model can immediately be scored against this test set. When the hyperparameter is set to oob_score=True, this is exactly what happens.

In other words, oob_score provides a shortcut to get a...