book

Machine Learning Engineering on AWS

Name: Machine Learning Engineering on AWS
Author: Joshua Arvin Lat
ISBN: 9781803247595

by Joshua Arvin Lat

October 2022

Intermediate to advanced

530 pages

11h 57m

English

Packt Publishing

Read now

Unlock full access

Machine Learning Engineering on AWS
Copyright © 2022 Packt PublishingContributorsAbout the authorAbout the reviewers
Preface
Who this book is forWhat this book coversTo get the most out of this bookDownload the example code filesDownload the color imagesConventions usedGet in touchShare Your Thoughts
Part 1: Getting Started with Machine Learning Engineering on AWS
Chapter 1: Introduction to ML Engineering on AWS
Technical requirementsWhat is expected from ML engineers?How ML engineers can get the most out of AWSEssential prerequisitesCreating the Cloud9 environmentIncreasing Cloud9’s storageInstalling the Python prerequisitesPreparing the datasetGenerating a synthetic dataset using a deep learning modelExploratory data analysisTrain-test splitUploading the dataset to Amazon S3AutoML with AutoGluonSetting up and installing AutoGluonPerforming your first AutoGluon AutoML experimentGetting started with SageMaker and SageMaker StudioOnboarding with SageMaker StudioAdding a user to an existing SageMaker DomainNo-code machine learning with SageMaker CanvasAutoML with SageMaker AutopilotSummaryFurther reading
Chapter 2: Deep Learning AMIs
Technical requirementsGetting started with Deep Learning AMIsLaunching an EC2 instance using a Deep Learning AMILocating the framework-specific DLAMIChoosing the instance typeEnsuring a default secure configurationLaunching the instance and connecting to it using EC2 Instance ConnectDownloading the sample datasetTraining an ML modelLoading and evaluating the modelCleaning upUnderstanding how AWS pricing works for EC2 instancesUsing multiple smaller instances to reduce the overall cost of running ML workloadsUsing spot instances to reduce the cost of running training jobsSummaryFurther reading
Chapter 3: Deep Learning Containers
Technical requirementsGetting started with AWS Deep Learning ContainersEssential prerequisitesPreparing the Cloud9 environmentDownloading the sample datasetUsing AWS Deep Learning Containers to train an ML modelServerless ML deployment with Lambda’s container image supportBuilding the custom container imageTesting the container imagePushing the container image to Amazon ECRRunning ML predictions on AWS LambdaCompleting and testing the serverless API setupSummaryFurther reading
Part 2:Solving Data Engineering and Analysis Requirements
Chapter 4: Serverless Data Management on AWS
Technical requirementsGetting started with serverless data managementPreparing the essential prerequisitesOpening a text editor on your local machineCreating an IAM userCreating a new VPCUploading the dataset to S3Running analytics at scale with Amazon Redshift ServerlessSetting up a Redshift Serverless endpointOpening Redshift query editor v2Creating a tableLoading data from S3Querying the databaseUnloading data to S3Setting up Lake FormationCreating a databaseCreating a table using an AWS Glue CrawlerUsing Amazon Athena to query data in Amazon S3Setting up the query result locationRunning SQL queries using AthenaSummaryFurther reading
Chapter 5: Pragmatic Data Processing and Analysis
Technical requirementsGetting started with data processing and analysisPreparing the essential prerequisitesDownloading the Parquet file Preparing the S3 bucketAutomating data preparation and analysis with AWS Glue DataBrewCreating a new datasetCreating and running a profile jobCreating a project and configuring a recipeCreating and running a recipe jobVerifying the resultsPreparing ML data with Amazon SageMaker Data WranglerAccessing Data WranglerImporting dataTransforming the dataAnalyzing the dataExporting the data flowTurning off the resourcesVerifying the resultsSummaryFurther reading
Part 3: Diving Deeper with Relevant Model Training and Deployment Solutions

Chapter 6: SageMaker Training and Debugging Solutions
Technical requirementsGetting started with the SageMaker Python SDKPreparing the essential prerequisitesCreating a service limit increase requestTraining an image classification model with the SageMaker Python SDKCreating a new Notebook in SageMaker StudioDownloading the training, validation, and test datasetsUploading the data to S3Using the SageMaker Python SDK to train an ML modelUsing the %store magic to store dataUsing the SageMaker Python SDK to deploy an ML modelUsing the Debugger Insights DashboardUtilizing Managed Spot Training and CheckpointsCleaning upSummaryFurther reading
Chapter 7: SageMaker Deployment Solutions
Technical requirementsGetting started with model deployments in SageMakerPreparing the pre-trained model artifacts Preparing the SageMaker script mode prerequisitesPreparing the inference.py filePreparing the requirements.txt filePreparing the setup.py fileDeploying a pre-trained model to a real-time inference endpointDeploying a pre-trained model to a serverless inference endpointDeploying a pre-trained model to an asynchronous inference endpointCreating the input JSON fileAdding an artificial delay to the inference scriptDeploying and testing an asynchronous inference endpointCleaning upDeployment strategies and best practices SummaryFurther reading
Part 4:Securing, Monitoring, and Managing Machine Learning Systems and Environments
Chapter 8: Model Monitoring and Management Solutions
Technical prerequisitesRegistering models to SageMaker Model RegistryCreating a new notebook in SageMaker StudioRegistering models to SageMaker Model Registry using the boto3 libraryDeploying models from SageMaker Model RegistryEnabling data capture and simulating predictionsScheduled monitoring with SageMaker Model MonitorAnalyzing the captured dataDeleting an endpoint with a monitoring scheduleCleaning upSummaryFurther reading
Chapter 9: Security, Governance, and Compliance Strategies
Managing the security and compliance of ML environmentsAuthentication and authorizationNetwork securityEncryption at rest and in transitManaging compliance reportsVulnerability managementPreserving data privacy and model privacyFederated LearningDifferential PrivacyPrivacy-preserving machine learningOther solutions and optionsEstablishing ML governanceLineage Tracking and reproducibilityModel inventoryModel validationML explainabilityBias detectionModel monitoringTraceability, observability, and auditingData quality analysis and reportingData integrity managementSummaryFurther reading
Part 5:Designing and Building End-to-end MLOps Pipelines
Chapter 10: Machine Learning Pipelines with Kubeflow on Amazon EKS
Technical requirementsDiving deeper into Kubeflow, Kubernetes, and EKSPreparing the essential prerequisitesPreparing the IAM role for the EC2 instance of the Cloud9 environmentAttaching the IAM role to the EC2 instance of the Cloud9 environmentUpdating the Cloud9 environment with the essential prerequisitesSetting up Kubeflow on Amazon EKSRunning our first Kubeflow pipelineUsing the Kubeflow Pipelines SDK to build ML workflowsCleaning upRecommended strategies and best practicesSummaryFurther reading
Chapter 11: Machine Learning Pipelines with SageMaker Pipelines
Technical requirementsDiving deeper into SageMaker PipelinesPreparing the essential prerequisitesRunning our first pipeline with SageMaker PipelinesDefining and preparing our first ML pipelineRunning our first ML pipelineCreating Lambda functions for deploymentPreparing the Lambda function for deploying a model to a new endpointPreparing the Lambda function for checking whether an endpoint existsPreparing the Lambda function for deploying a model to an existing endpointTesting our ML inference endpointCompleting the end-to-end ML pipelineDefining and preparing the complete ML pipelineRunning the complete ML pipelineCleaning upRecommended strategies and best practicesSummaryFurther reading
Index
Why subscribe?
Other Books You May EnjoyPackt is searching for authors like youShare Your Thoughts

Overview

"Machine Learning Engineering on AWS" is your comprehensive guide to mastering machine learning operations on Amazon Web Services (AWS). In this book, you'll learn how to use key AWS services like SageMaker and EKS to build, scale, and deploy robust ML pipelines. With practical examples, you'll gain the skills needed to tackle real-world ML engineering tasks with confidence.

What this Book will help me do

Build and optimize ML models with TensorFlow and PyTorch on AWS.
Set up scalable, secure ML pipelines using services like SageMaker and EKS.
Leverage AWS Glue DataBrew and other tools for efficient data handling.
Understand cost optimization techniques for running ML systems on the cloud.
Ensure compliance and privacy in ML projects by following best practices.

Author(s)

Joshua Arvin Lat is a seasoned AWS expert and a passionate advocate for cloud-based ML solutions. With extensive experience in data engineering and machine learning, Joshua excels at simplifying complex topics and providing actionable insights. His writing focuses on practical applications to help readers gain a competitive advantage in the field of ML engineering.

Who is it for?

This book is ideal for machine learning engineers, data scientists, and cloud engineers who want to deploy ML systems efficiently on AWS. A prior understanding of AWS, ML principles, and Python will be advantageous. If you're looking to implement scalable and secure ML pipelines in a production environment, this book is for you.

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

Read now

Unlock full access

More than 5,000 organizations count on O’Reilly

O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.

Julian F.

Head of Cybersecurity

I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.

Addison B.

Field Engineer

I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.

Amir M.

Data Platform Tech Lead

I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.

Mark W.

Embedded Software Engineer

Publisher Resources

ISBN: 9781803247595

Cloud Computing

Data Engineering

Data Science

AI & ML

Programming Languages

Software Architecture

IT/Ops

Security

Design

Business

Soft Skills