Machine Learning with BigQuery ML: Create, execute, and improve machine learning models in BigQuery using standard SQL queries

✍ Scribed by Alessandro Marrandino

• Publisher: Packt Publishing
• Year: 2021
• Tongue: English
• Leaves: 344
• Category: Library

✦ Synopsis

Manage different business scenarios with the right machine learning technique using Google's highly scalable BigQuery ML

Key Features

• Gain a clear understanding of AI and machine learning services on GCP, learn when to use these, and find out how to integrate them with BigQuery ML
• Leverage SQL syntax to train, evaluate, test, and use ML models
• Discover how BigQuery works and understand the capabilities of BigQuery ML using examples

Book Description

BigQuery ML enables you to easily build machine learning (ML) models with SQL without much coding. This book will help you to accelerate the development and deployment of ML models with BigQuery ML.

The book starts with a quick overview of Google Cloud and BigQuery architecture. You'll then learn how to configure a Google Cloud project, understand the architectural components and capabilities of BigQuery, and find out how to build ML models with BigQuery ML. The book teaches you how to use ML using SQL on BigQuery. You'll analyze the key phases of a ML model's lifecycle and get to grips with the SQL statements used to train, evaluate, test, and use a model. As you advance, you'll build a series of use cases by applying different ML techniques such as linear regression, binary and multiclass logistic regression, k-means, ARIMA time series, deep neural networks, and XGBoost using practical use cases. Moving on, you'll cover matrix factorization and deep neural networks using BigQuery ML's capabilities. Finally, you'll explore the integration of BigQuery ML with other Google Cloud Platform components such as AI Platform Notebooks and TensorFlow along with discovering best practices and tips and tricks for hyperparameter tuning and performance enhancement.

By the end of this BigQuery book, you'll be able to build and evaluate your own ML models with BigQuery ML.

What you will learn

• Discover how to prepare datasets to build an effective ML model
• Forecast business KPIs by leveraging various ML models and BigQuery ML
• Build and train a recommendation engine to suggest the best products for your customers using BigQuery ML
• Develop, train, and share a BigQuery ML model from previous parts with AI Platform Notebooks
• Find out how to invoke a trained TensorFlow model directly from BigQuery
• Get to grips with BigQuery ML best practices to maximize your ML performance

Who this book is for

This book is for data scientists, data analysts, data engineers, and anyone looking to get started with Google's BigQuery ML. You'll also find this book useful if you want to accelerate the development of ML models or if you are a business user who wants to apply ML in an easy way using SQL. Basic knowledge of BigQuery and SQL is required.

Table of Contents

1. Introduction to Google Cloud and BigQuery
2. Setting Up Your GCP and BigQuery Environment
3. Introducing BigQuery Syntax
4. Predicting Numerical Values with Linear Regression
5. Predicting Boolean Values Using Binary Logistic Regression
6. Classifying Trees with Multiclass Logistic Regression
7. Clustering Using the K-Means Algorithm
8. Forecasting Using Time Series
9. Suggesting the Right Product by Using Matrix Factorization
10. Predicting Boolean Values Using XGBoost
11. Implementing Deep Neural Networks
12. Using BigQuery ML with AI Notebooks
13. Running TensorFlow Models with BigQuery ML
14. BigQuery ML Tips and Best Practices

✦ Table of Contents

Cover
Title Page
Copyright and Credits
Contributors
Table of Contents
Preface
Section 1: Introduction and Environment Setup
Chapter 1: Introduction to Google Cloud and BigQuery
Introducing Google Cloud Platform
Interacting with GCP
Discovering GCP's key differentiators
Exploring AI and ML services on GCP
Core platform services
Building blocks
Solutions
Introducing BigQuery
BigQuery architecture
BigQuery's advantages over traditional data warehouses
Interacting with BigQuery
BigQuery data structures
Discovering BigQuery ML
BigQuery ML benefits
BigQuery ML algorithms
Understanding BigQuery pricing
BigQuery pricing
BigQuery ML pricing
Free operations and free tiers
Pricing calculator
Summary
Further resources
Chapter 2: Setting Up Your GCP and BigQuery Environment
Technical requirements
Creating your GCP account and project
Registering a GCP account
Exploring Google Cloud Console
Creating a GCP project
Activating BigQuery
Discovering the BigQuery web UI
Exploring the BigQuery public datasets
Searching for a public dataset
Analyzing a table
Summary
Further reading
Chapter 3: Introducing BigQuery Syntax
Technical requirements
Creating a BigQuery dataset
Discovering BigQuery SQL
CRUD operations
Diving into BigQuery ML
Summary
Further resources
Section 2: Deep Learning Networks
Chapter 4: Predicting Numerical Values with Linear Regression
Technical requirements
Introducing the business scenario
Discovering linear regression
Exploring and understanding the dataset
Understanding the data
Checking the data's quality
Segmenting the dataset
Training the linear regression model
Evaluating the linear regression model
Utilizing the linear regression model
Drawing business conclusions
Summary
Further reading
Chapter 5: Predicting Boolean Values Using Binary Logistic Regression
Technical requirements
Introducing the business scenario
Discovering binary logistic regression
Exploring and understanding the dataset
Understanding the data
Segmenting the dataset
Training the binary logistic regression model
Evaluating the binary logistic regression model
Using the binary logistic regression model
Drawing business conclusions
Summary
Further resources
Chapter 6: Classifying Trees with Multiclass Logistic Regression
Technical requirements
Introducing the business scenario
Discovering multiclass logistic regression
Exploring and understanding the dataset
Understanding the data
Checking the data quality
Segmenting the dataset
Training the multiclass logistic regression model
Evaluating the multiclass logistic regression model
Using the multiclass logistic regression model
Drawing business conclusions
Summary
Further resources
Section 3: Advanced Models with BigQuery ML
Chapter 7: Clustering Using the K-Means Algorithm
Technical requirements
Introducing the business scenario
Discovering K-Means clustering
Exploring and understanding the dataset
Understanding the data
Checking the data quality
Creating the training datasets
Training the K-Means clustering model
Evaluating the K-Means clustering model
Using the K-Means clustering model
Drawing business conclusions
Summary
Further resources
Chapter 8: Forecasting Using Time Series
Technical requirements
Introducing the business scenario
Discovering time series forecasting
Exploring and understanding the dataset
Understanding the data
Checking the data quality
Creating the training dataset
Training the time series forecasting model
Evaluating the time series forecasting model
Using the time series forecasting model
Presenting the forecast
Summary
Further resources
Chapter 9: Suggesting the Right Product by Using Matrix Factorization
Technical requirements
Introducing the business scenario
Discovering matrix factorization
Configuring BigQuery Flex Slots
Exploring and preparing the dataset
Understanding the data
Creating the training dataset
Training the matrix factorization model
Evaluating the matrix factorization model
Using the matrix factorization model
Drawing business conclusions
Summary
Further resources
Chapter 10: Predicting Boolean Values Using XGBoost
Technical requirements
Introducing the business scenario
Discovering the XGBoost Boosted Tree classification model
Exploring and understanding the dataset
Checking the data quality
Segmenting the dataset
Training the XGBoost classification model
Evaluating the XGBoost classification model
Using the XGBoost classification model
Drawing business conclusions
Summary
Further resources
Chapter 11: Implementing Deep Neural Networks
Technical requirements
Introducing the business scenario
Discovering DNNs
DNNs in BigQuery ML
Preparing the dataset
Training the DNN models
Evaluating the DNN models
Using the DNN models
Drawing business conclusions
Deep neural networks versus linear models
Summary
Further resources
Section 4: Further Extending Your ML Capabilities with GCP
Chapter 12: Using BigQuery ML with AI Notebooks
Technical requirements
Discovering AI Platform Notebooks
AI Platform Notebooks pricing
Configuring the first notebook
Implementing BigQuery ML models within notebooks
Compiling the AI notebook
Running the code in the AI notebook
Summary
Further resources
Chapter 13: Running TensorFlow Models with BigQuery ML
Technical requirements
Introducing TensorFlow
Discovering the relationship between BigQuery ML and TensorFlow
Understanding commonalities and differences
Collaborating with BigQuery ML and TensorFlow
Converting BigQuery ML models into TensorFlow
Training the BigQuery ML to export it
Exporting the BigQuery ML model
Running TensorFlow models with BigQuery ML
Summary
Further resources
Chapter 14: BigQuery ML Tips and Best Practices
Choosing the right BigQuery ML algorithm
Preparing the datasets
Working with high-quality data
Segmenting the datasets
Understanding feature engineering
Tuning hyperparameters
Using BigQuery ML for online predictions
Summary
Further resources
Other Books You May Enjoy
Index