Automated Machine Learning on AWS: Fast-track the development of your production-ready machine learning applications the AWS way

Cover
Title Page
Copyright and Credits
Foreword
Contributors
Table of Contents
Preface
Section 1: Fundamentals of the Automated Machine Learning Process and AutoML on AWS
Chapter 1: Getting Started with Automated Machine Learning on AWS
	Technical requirements
	Overview of the ML process
	Complexities in the ML process
	An example of the end-to-end ML process
		Introducing ACME Fishing Logistics
		The case for ML
		Getting insights from the data
		Building the right model
		Training the model
		Evaluating the trained model
		Exploring possible next steps
		Tuning our model
		Deploying the optimized model into production
		Streamlining the ML process with AutoML
	How AWS makes automating the ML development and deployment process easier
	Summary
Chapter 2: Automating Machine Learning Model Development Using SageMaker Autopilot
	Technical requirements
	Introducing the AWS AI and ML landscape
	Overview of SageMaker Autopilot
	Overcoming automation challenges with SageMaker Autopilot
		Getting started with SageMaker Studio
		Preparing the experiment data
		Starting the Autopilot experiment
		Running the Autopilot experiment
		Post-experimentation tasks
	Using the SageMaker SDK to automate the ML experiment
		Codifying the Autopilot experiment
		Analyzing the Autopilot experiment with code
		Deploying the best candidate
		Cleaning up
	Summary
Chapter 3: Automating Complicated Model Development with AutoGluon
	Technical requirements
	Introducing the AutoGluon library
	Using AutoGluon for tabular data
		Prerequisites
		Creating the AutoML experiment with AutoGluon
		Evaluating the experiment results
	Using AutoGluon for image data
		Prerequisites
		Creating an image prediction experiment
		Evaluating the experiment results
	Summary
Section 2: Automating the Machine Learning Process with Continuous Integration and Continuous Delivery (CI/CD)
Chapter 4: Continuous Integration and Continuous Delivery (CI/CD) for Machine Learning
	Technical requirements
	Introducing the CI/CD methodology
		Introducing the CI part of CI/CD
		Introducing the CD part of CI/CD
		Closing the loop
	Automating ML with CI/CD
		Taking a deployment-centric approach
		Creating an MLOps methodology
	Creating a CI/CD pipeline on AWS
		Using the AWS CI/CD toolchain
		Working with additional AWS developer tools
		Creating a cloud-native CI/CD pipeline for a production ML model
		Preparing the development environment
		Creating the pipeline artifact repository
		Developing the application artifacts
	Summary
Chapter 5: Continuous Deployment of a Production ML Model
	Technical requirements
	Deploying the CI/CD pipeline
		Codifying the pipeline construct
		Creating the CDK application
		Deploying the pipeline application
	Building the ML model artifacts
		Reviewing the modeling file
		Reviewing the application file
		Reviewing the model serving files
		Reviewing the container build file
		Committing the ML artifacts
	Executing the automated ML model deployment
		Cleanup
	Summary
Section 3: Optimizing a Source Code-Centric Approach to Automated Machine Learning
Chapter 6: Automating the Machine Learning Process Using AWS Step Functions
	Technical requirements
	Introducing AWS Step Functions
		Creating a state machine
		Addressing state machine complexity
	Using the Step Functions Data Science SDK for CI/CD
	Building the CI/CD pipeline resources
		Updating the development environment
		Creating the pipeline artifact repository
		Building the pipeline application artifacts
		Deploying the CI/CD pipeline
	Summary
Chapter 7: Building the ML Workflow Using AWS Step Functions
	Technical requirements
	Building the state machine workflow
		Setting up the service permissions
		Creating an ML workflow
	Performing the integration test
	Monitoring the pipeline's progress
	Summary
Section 4: Optimizing a Data-Centric Approach to Automated Machine Learning
Chapter 8: Automating the Machine Learning Process Using Apache Airflow
	Technical requirements
	Introducing Apache Airflow
	Introducing Amazon MWAA
	Using Airflow to process the abalone dataset
	Configuring the MWAA prerequisites
	Configuring the MWAA environment
	Summary
Chapter 9: Building the ML Workflow Using Amazon Managed Workflows for Apache Airflow
	Technical requirements
	Developing the data-centric workflow
		Building and unit testing the data ETL artifacts
		Building the Airflow DAG
	Creating synthetic Abalone survey data
	Executing the data-centric workflow
		Cleanup
	Summary
Section 5: Automating the End-to-End Production Application on AWS
Chapter 10: An Introduction to the Machine Learning Software Development Life Cycle (MLSDLC)
	Technical requirements
	Introducing the MLSDLC
	Building the application platform
		Examining the role of the application owner
		Examining the role of the platform engineers
		Examining the role of the frontend developers
	Examining ML and data engineering roles
		Creating a SageMaker Feature Store
		Creating ML artifacts
		Creating continuous training artifacts
	Understanding the security lens
		Securing the data
		Securing the code
		Securing the website
	Summary
Chapter 11: Continuous Integration, Deployment, and Training for the MLSDLC
	Technical requirements
	Codifying the continuous integration stage
		Building the integration artifacts
		Building the test artifacts
		Building the production artifacts
		Automating the continuous integration process
	Managing the continuous deployment stage
		Reviewing the build phase
		Reviewing the test phase
		Reviewing the deploy and maintain phases
		Reviewing the application user experience
	Managing continuous training
		Creating new Abalone survey data
		Reviewing the continuous training process
		Cleanup
	Summary
	Further reading
Index
Other Books You May Enjoy