Book Image

Machine Learning Engineering on AWS

By : Joshua Arvin Lat
Book Image

Machine Learning Engineering on AWS

By: Joshua Arvin Lat

Overview of this book

There is a growing need for professionals with experience in working on machine learning (ML) engineering requirements as well as those with knowledge of automating complex MLOps pipelines in the cloud. This book explores a variety of AWS services, such as Amazon Elastic Kubernetes Service, AWS Glue, AWS Lambda, Amazon Redshift, and AWS Lake Formation, which ML practitioners can leverage to meet various data engineering and ML engineering requirements in production. This machine learning book covers the essential concepts as well as step-by-step instructions that are designed to help you get a solid understanding of how to manage and secure ML workloads in the cloud. As you progress through the chapters, you’ll discover how to use several container and serverless solutions when training and deploying TensorFlow and PyTorch deep learning models on AWS. You’ll also delve into proven cost optimization techniques as well as data privacy and model privacy preservation strategies in detail as you explore best practices when using each AWS. By the end of this AWS book, you'll be able to build, scale, and secure your own ML systems and pipelines, which will give you the experience and confidence needed to architect custom solutions using a variety of AWS services for ML engineering requirements.

Related Content you might be interested in

Current Title:

Small book image
Machine Learning Engineering on AWS
Joshua Arvin Lat
Oct, 2022 | 530 pages
Small book image
Machine Learning with Amazon SageMaker Cookbook
Joshua Arvin Lat
Oct, 2021 | 762 pages
Small book image
Building and Automating Penetration Testing Labs in the Cloud
Joshua Arvin Lat
Oct, 2023 | 562 pages
Small book image
Getting Started with Amazon SageMaker Studio
Michael Hsieh
Mar, 2022 | 326 pages
Table of Contents (19 chapters)
Preface
Preface
Who this book is for
What this book covers
To get the most out of this book
Download the example code files
Download the color images
Conventions used
Get in touch
Share Your Thoughts
1
Part 1: Getting Started with Machine Learning Engineering on AWS
Part 1: Getting Started with Machine Learning Engineering on AWS
Free Chapter
2
Chapter 1: Introduction to ML Engineering on AWS
Chapter 1: Introduction to ML Engineering on AWS
Technical requirements
What is expected from ML engineers?
How ML engineers can get the most out of AWS
Essential prerequisites
Preparing the dataset
AutoML with AutoGluon
Getting started with SageMaker and SageMaker Studio
No-code machine learning with SageMaker Canvas
AutoML with SageMaker Autopilot
Summary
Further reading
3
Chapter 2: Deep Learning AMIs
Chapter 2: Deep Learning AMIs
Technical requirements
Getting started with Deep Learning AMIs
Launching an EC2 instance using a Deep Learning AMI
Downloading the sample dataset
Training an ML model
Loading and evaluating the model
Cleaning up
Understanding how AWS pricing works for EC2 instances
Summary
Further reading
4
Chapter 3: Deep Learning Containers
Chapter 3: Deep Learning Containers
Technical requirements
Getting started with AWS Deep Learning Containers
Essential prerequisites
Using AWS Deep Learning Containers to train an ML model
Serverless ML deployment with Lambda’s container image support
Summary
Further reading
5
Part 2:Solving Data Engineering and Analysis Requirements
Part 2:Solving Data Engineering and Analysis Requirements
6
Chapter 4: Serverless Data Management on AWS
Chapter 4: Serverless Data Management on AWS
Technical requirements
Getting started with serverless data management
Preparing the essential prerequisites
Running analytics at scale with Amazon Redshift Serverless
Setting up Lake Formation
Using Amazon Athena to query data in Amazon S3
Summary
Further reading
7
Chapter 5: Pragmatic Data Processing and Analysis
Chapter 5: Pragmatic Data Processing and Analysis
Technical requirements
Getting started with data processing and analysis
Preparing the essential prerequisites
Automating data preparation and analysis with AWS Glue DataBrew
Preparing ML data with Amazon SageMaker Data Wrangler
Summary
Further reading
8
Part 3: Diving Deeper with Relevant Model Training and Deployment Solutions
Part 3: Diving Deeper with Relevant Model Training and Deployment Solutions
9
Chapter 6: SageMaker Training and Debugging Solutions
Chapter 6: SageMaker Training and Debugging Solutions
Technical requirements
Getting started with the SageMaker Python SDK
Preparing the essential prerequisites
Training an image classification model with the SageMaker Python SDK
Using the Debugger Insights Dashboard
Utilizing Managed Spot Training and Checkpoints
Cleaning up
Summary
Further reading
10
Chapter 7: SageMaker Deployment Solutions
Chapter 7: SageMaker Deployment Solutions
Technical requirements
Getting started with model deployments in SageMaker
Preparing the pre-trained model artifacts
Preparing the SageMaker script mode prerequisites
Deploying a pre-trained model to a real-time inference endpoint
Deploying a pre-trained model to a serverless inference endpoint
Deploying a pre-trained model to an asynchronous inference endpoint
Cleaning up
Deployment strategies and best practices
Summary
Further reading
11
Part 4:Securing, Monitoring, and Managing Machine Learning Systems and Environments
Part 4:Securing, Monitoring, and Managing Machine Learning Systems and Environments
12
Chapter 8: Model Monitoring and Management Solutions
Chapter 8: Model Monitoring and Management Solutions
Technical prerequisites
Registering models to SageMaker Model Registry
Deploying models from SageMaker Model Registry
Enabling data capture and simulating predictions
Scheduled monitoring with SageMaker Model Monitor
Analyzing the captured data
Deleting an endpoint with a monitoring schedule
Cleaning up
Summary
Further reading
13
Chapter 9: Security, Governance, and Compliance Strategies
Chapter 9: Security, Governance, and Compliance Strategies
Managing the security and compliance of ML environments
Preserving data privacy and model privacy
Establishing ML governance
Summary
Further reading
14
Part 5:Designing and Building End-to-end MLOps Pipelines
Part 5:Designing and Building End-to-end MLOps Pipelines
15
Chapter 10: Machine Learning Pipelines with Kubeflow on Amazon EKS
Chapter 10: Machine Learning Pipelines with Kubeflow on Amazon EKS
Technical requirements
Diving deeper into Kubeflow, Kubernetes, and EKS
Preparing the essential prerequisites
Setting up Kubeflow on Amazon EKS
Running our first Kubeflow pipeline
Using the Kubeflow Pipelines SDK to build ML workflows
Cleaning up
Recommended strategies and best practices
Summary
Further reading
16
Chapter 11: Machine Learning Pipelines with SageMaker Pipelines
Chapter 11: Machine Learning Pipelines with SageMaker Pipelines
Technical requirements
Diving deeper into SageMaker Pipelines
Preparing the essential prerequisites
Running our first pipeline with SageMaker Pipelines
Creating Lambda functions for deployment
Testing our ML inference endpoint
Completing the end-to-end ML pipeline
Cleaning up
Recommended strategies and best practices
Summary
Further reading
17
Index
Index
Why subscribe?
18
Other Books You May Enjoy
Other Books You May Enjoy
Packt is searching for authors like you
Share Your Thoughts

Creating Lambda functions for deployment

Our second (and more complete pipeline) will require a few additional resources to help us deploy our ML model. In this section, we will create the following Lambda functions:

  • check-if-endpoint-exists – This is a Lambda function that accepts the name of the ML inference endpoint as input and returns True if the endpoint exists already.
  • deploy-model-to-new-endpoint – This is a Lambda function that accepts the model package ARN as input (along with the role and the endpoint name) and deploys the model into a new inference endpoint
  • deploy-model-to-existing-endpoint – This is a Lambda function that accepts the model package ARN as input (along with the role and the endpoint name) and deploys the model into an existing inference endpoint (by updating the deployed model inside the ML instance)

We will use these functions later in the Completing the end-to-end ML pipeline section to deploy the ML model we...

Previous Section
End of Section 6
Next Section