What this book covers

Chapter 1, Getting Started with Amazon SageMaker, provides an overview of Amazon SageMaker, what its capabilities are, and how it helps solve many pain points faced by ML projects today.

Chapter 2, Handling Data Preparation Techniques, discusses data preparation options. Although this it isn't the core subject of the book, data preparation is a key topic in ML, and it should be covered at a high level.

Chapter 3, AutoML with Amazon SageMaker AutoPilot, shows you how to build, train, and optimize ML models automatically with Amazon SageMaker AutoPilot.

Chapter 4, Training Machine Learning Models, shows you how to build and train models using the collection of statistical ML algorithms built into Amazon SageMaker.

Chapter 5, Training Computer Vision Models, shows you how to build and train models using the collection of computer vision algorithms built into Amazon SageMaker.

Chapter 6, Training Natural Language Processing Models, shows you how to build and train models using the collection of natural language processing algorithms built into Amazon SageMaker.

Chapter 7, Extending Machine Learning Services Using Built-In Frameworks, shows you how to build and train ML models using the collection of built-in open source frameworks in Amazon SageMaker.

Chapter 8, Using Your Algorithms and Code, shows you how to build and train ML models using your own code on Amazon SageMaker, for example, R or custom Python.

Chapter 9, Scaling Your Training Jobs, shows you how to distribute training jobs to many managed instances, using either built-in algorithms or built-in frameworks.

Chapter 10, Advanced Training Techniques, shows you how to leverage advanced training in Amazon SageMaker.

Chapter 11, Deploying Machine Learning Models, shows you how to deploy ML models in a variety of configurations.

Chapter 12, Automating Deployment Tasks, shows you how to automate the deployment of ML models on Amazon SageMaker.

Chapter 13, Optimizing Cost and Performance, shows you how to optimize model deployments, both from an infrastructure perspective and from a cost perspective.