What this book covers
Chapter 1, Introduction to ML Engineering on AWS, focuses on helping you get set up, understand the key concepts, and get your feet wet quickly with several simplified AutoML examples.
Chapter 2, Deep Learning AMIs, introduces AWS Deep Learning AMIs and how they are used to help ML practitioners perform ML experiments faster inside EC2 instances. Here, we will also dive a bit deeper into how AWS pricing works for EC2 instances so that you will have a better idea of how to optimize and reduce the overall costs of running ML workloads in the cloud.
Chapter 3, Deep Learning Containers, introduces AWS Deep Learning Containers and how they are used to help ML practitioners perform ML experiments faster using containers. Here, we will also deploy a trained deep learning model inside an AWS Lambda function using Lambda’s container image support.
Chapter 4, Serverless Data Management on AWS, presents several serverless solutions, such as Amazon Redshift Serverless and AWS Lake Formation, for managing and querying data on AWS.
Chapter 5, Pragmatic Data Processing and Analysis, focuses on the different services available when working on data processing and analysis requirements, such as AWS Glue DataBrew and Amazon SageMaker Data Wrangler.
Chapter 6, SageMaker Training and Debugging Solutions, presents the different solutions and capabilities available when training an ML model using Amazon SageMaker. Here, we dive a bit deeper into the different options and strategies when training and tuning ML models in SageMaker.
Chapter 7, SageMaker Deployment Solutions, focuses on the relevant deployment solutions and strategies when performing ML inference on the AWS platform.
Chapter 8, Model Monitoring and Management Solutions, presents the different monitoring and management solutions available on AWS.
Chapter 9, Security, Governance, and Compliance Strategies, focuses on the relevant security, governance, and compliance strategies needed to secure production environments. Here, we will also dive a bit deeper into the different techniques to ensure data privacy and model privacy.
Chapter 10, Machine Learning Pipelines with Kubeflow on Amazon EKS, focuses on using Kubeflow Pipelines, Kubernetes, and Amazon EKS to deploy an automated end-to-end MLOps pipeline on AWS.
Chapter 11, Machine Learning Pipelines with SageMaker Pipelines, focuses on using SageMaker Pipelines to design and build automated end-to-end MLOps pipelines. Here, we will apply, combine, and connect the different strategies and techniques we learned in the previous chapters of the book.