XGBoost for Regression Predictive Modeling and Time Series Analysis: Build intuitive understanding, develop, build, evaluate and deploy model
1st Edition, Kindle Edition
by Partha Pritam Deka (Author), Joyce Weiner (Author) Format: Kindle Edition
Build an understanding of XGBoost and gain hands-on experience with the XGBoost Python API through multiple practical use cases for classification, Regression and Time series analysis including model testing and deployment.
Key Features
Book Description
XGBoost is a popular open-source library that provides an efficient, effective, scalable and high-performance implementation of the gradient boosting algorithm. You will be able to build an intuitive and practical understanding of the XGBoost algorithm through our demystifying the complex math underneath and explanation of XGBoost’s benefits over other decision tree ensemble models, including when to use XGBoost or other prediction algorithms. This book provides a hands-on approach to implementation of the XGBoost Python API as well as the scikit-learn API that will help one to be up-and-running and productive in no time. Complete with step-by-step explanations of essential concepts, as well as practical examples, this book begins with a brief introduction to machine learning concepts, then dives into the fundamentals of XGBoost and its benefits before exploring practical applications. You will get hands-on experience using XGBoost through practical use cases on classification, regression, and time-series data. By the end of this book, you will have an understanding of the XGBoost algorithm, have installed the XGBoost API, downloaded and prepared a practical dataset, trained the XGBoost model, make predictions, and evaluated and deployed models using the Python and scikit-learn API.
What you will learn
Who this book is for
This book is for data scientists, machine learning developers, and anyone with basic coding knowledge and familiarity with Python, GitHub and other Dev Ops tools, looking to build effective predictive models using XGBoost. We address the top three common problems when building predictive models: problems with available data such as missing data and non-normal data, the desire to combine numeric and text (categorical) data, how to get value out from non-numeric data to improve predictions, and how to deploy and sustain a model, how to measure and improve model fitting.
Table of Contents
Introduction: Machine Learning Overview, Classification, and Regression
XGBoost Quick Start Guide with Iris data Case Study
Demystifying the XGBoost Paper
Adding on to the Quick Start: Switching out the dataset with Housing data Case Study
Classification and Regression Trees, Ensembles, and Deep Learning Models – what’s best for your data?
Data cleaning, Imbalanced Data, and Other Data Problems
Data cleaning, Imbalanced Data, and Other Data Problems Feature Engineering & Feature Selection
Encoding Techniques for Categorical Features
How to Use XGBoost for Time Series Forecasting
Model interpretability, Explainability, and Feature Importance
Metrics for Model evaluations and Comparisons