Book Image

Machine Learning with Amazon SageMaker Cookbook

By : Joshua Arvin Lat
Book Image

Machine Learning with Amazon SageMaker Cookbook

By: Joshua Arvin Lat

Overview of this book

Amazon SageMaker is a fully managed machine learning (ML) service that helps data scientists and ML practitioners manage ML experiments. In this book, you'll use the different capabilities and features of Amazon SageMaker to solve relevant data science and ML problems. This step-by-step guide features 80 proven recipes designed to give you the hands-on machine learning experience needed to contribute to real-world experiments and projects. You'll cover the algorithms and techniques that are commonly used when training and deploying NLP, time series forecasting, and computer vision models to solve ML problems. You'll explore various solutions for working with deep learning libraries and frameworks such as TensorFlow, PyTorch, and Hugging Face Transformers in Amazon SageMaker. You'll also learn how to use SageMaker Clarify, SageMaker Model Monitor, SageMaker Debugger, and SageMaker Experiments to debug, manage, and monitor multiple ML experiments and deployments. Moreover, you'll have a better understanding of how SageMaker Feature Store, Autopilot, and Pipelines can meet the specific needs of data science teams. By the end of this book, you'll be able to combine the different solutions you've learned as building blocks to solve real-world ML problems.

Related Content you might be interested in

Current Title:

Small book image
Machine Learning with Amazon SageMaker Cookbook
Joshua Arvin Lat
Oct, 2021 | 762 pages
Small book image
Machine Learning Engineering on AWS
Joshua Arvin Lat
Oct, 2022 | 530 pages
Small book image
Learn Amazon SageMaker
Julien Simon
Nov, 2021 | 554 pages
Small book image
Learn Amazon SageMaker
Julien Simon
Aug, 2020 | 490 pages
Table of Contents (11 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
Code in Action
Download the color images
Conventions used
Sections
Get in touch
Share Your Thoughts
1
Chapter 1: Getting Started with Machine Learning Using Amazon SageMaker
Chapter 1: Getting Started with Machine Learning Using Amazon SageMaker
Technical requirements
Launching an Amazon SageMaker Notebook Instance
Checking the versions of the SageMaker Python SDK and the AWS CLI
Preparing the Amazon S3 bucket and the training dataset for the linear regression experiment
Visualizing and understanding your data in Python
Training your first model in Python
Loading a linear learner model with Apache MXNet in Python
Evaluating the model in Python
Deploying your first model in Python
Invoking an Amazon SageMaker model endpoint with the SageMakerRuntime client from boto3
Free Chapter
2
Chapter 2: Building and Using Your Own Algorithm Container Image
Chapter 2: Building and Using Your Own Algorithm Container Image
Technical requirements
Launching and preparing the Cloud9 environment
Setting up the Python and R experimentation environments
Preparing and testing the train script in Python
Preparing and testing the serve script in Python
Building and testing the custom Python algorithm container image
Pushing the custom Python algorithm container image to an Amazon ECR repository
Using the custom Python algorithm container image for training and inference with Amazon SageMaker Local Mode
Preparing and testing the train script in R
Preparing and testing the serve script in R
Building and testing the custom R algorithm container image
Pushing the custom R algorithm container image to an Amazon ECR repository
Using the custom R algorithm container image for training and inference with Amazon SageMaker Local Mode
3
Chapter 3: Using Machine Learning and Deep Learning Frameworks with Amazon SageMaker
Chapter 3: Using Machine Learning and Deep Learning Frameworks with Amazon SageMaker
Technical requirements
Preparing the SageMaker notebook instance for multiple deep learning local experiments
Generating a synthetic dataset for deep learning experiments
Preparing the entrypoint TensorFlow and Keras training script
Training and deploying a TensorFlow and Keras model with the SageMaker Python SDK
Preparing the entrypoint PyTorch training script
Preparing the entrypoint PyTorch inference script
Training and deploying a PyTorch model with the SageMaker Python SDK
Preparing the entrypoint scikit-learn training script
Training and deploying a scikit-learn model with the SageMaker Python SDK
Debugging disk space issues when using local mode
Debugging container execution issues when using local mode
4
Chapter 4: Preparing, Processing, and Analyzing the Data
Chapter 4: Preparing, Processing, and Analyzing the Data
Technical requirements
Generating a synthetic dataset for anomaly detection experiments
Training and deploying an RCF model
Invoking machine learning models with Amazon Athena using SQL queries
Analyzing data with Amazon Athena in Python
Generating a synthetic dataset for analysis and transformation
Performing dimensionality reduction with the built-in PCA algorithm
Performing cluster analysis with the built-in KMeans algorithm
Converting CSV data into protobuf recordIO format
Training a KNN model using the protobuf recordIO training input type
Preparing the SageMaker Processing prerequisites using the AWS CLI
Managed data processing with SageMaker Processing in Python
Managed data processing with SageMaker Processing in R
5
Chapter 5: Effectively Managing Machine Learning Experiments
Chapter 5: Effectively Managing Machine Learning Experiments
Technical requirements
Synthetic data generation for classification problems
Identifying issues with SageMaker Debugger
Inspecting SageMaker Debugger logs and results
Running and managing multiple experiments with SageMaker Experiments
Experiment analytics with SageMaker Experiments
Inspecting experiments, trials, and trial components with SageMaker Experiments
6
Chapter 6: Automated Machine Learning in Amazon SageMaker
Chapter 6: Automated Machine Learning in Amazon SageMaker
Technical requirements
Onboarding to SageMaker Studio
Generating a synthetic dataset with additional columns containing random values
Creating and monitoring a SageMaker Autopilot experiment in SageMaker Studio (console)
Creating and monitoring a SageMaker Autopilot experiment using the SageMaker Python SDK
Inspecting the SageMaker Autopilot experiment's results and artifacts
Performing Automatic Model Tuning with the SageMaker XGBoost built-in algorithm
Analyzing the Automatic Model Tuning job results
7
Chapter 7: Working with SageMaker Feature Store, SageMaker Clarify, and SageMaker Model Monitor
Chapter 7: Working with SageMaker Feature Store, SageMaker Clarify, and SageMaker Model Monitor
Technical requirements
Generating a synthetic dataset and using SageMaker Feature Store for storage and management
Querying data from the offline store of SageMaker Feature Store and uploading it to Amazon S3
Detecting pre-training bias with SageMaker Clarify
Detecting post-training bias with SageMaker Clarify
Enabling ML explainability with SageMaker Clarify
Deploying an endpoint from a model and enabling data capture with SageMaker Model Monitor
Baselining and scheduled monitoring with SageMaker Model Monitor
8
Chapter 8: Solving NLP, Image Classification, and Time-Series Forecasting Problems with Built-in Algorithms
Chapter 8: Solving NLP, Image Classification, and Time-Series Forecasting Problems with Built-in Algorithms
Technical requirements
Generating a synthetic dataset for text classification problems
Preparing the test dataset for batch transform inference jobs
Training and deploying a BlazingText model
Using Batch Transform for inference
Preparing the datasets for image classification using the Apache MXNet Vision Datasets classes
Training and deploying an image classifier using the built-in Image Classification Algorithm in SageMaker
Generating a synthetic time series dataset
Performing the train-test split on a time series dataset
Training and deploying a DeepAR model
Performing probabilistic forecasting with a deployed DeepAR model
9
Chapter 9: Managing Machine Learning Workflows and Deployments
Chapter 9: Managing Machine Learning Workflows and Deployments
Technical requirements
Working with Hugging Face models
Preparing the prerequisites of a multi-model endpoint deployment
Hosting multiple models with multi-model endpoints
Setting up A/B testing on multiple models with production variants
Preparing the Step Functions execution role
Managing ML workflows with AWS Step Functions and the Data Science SDK
Managing ML workflows with SageMaker Pipelines
Why subscribe?
10
Other Books You May Enjoy
Other Books You May Enjoy
Packt is searching for authors like you
Share Your Thoughts

What this book covers

Chapter 1, Getting Started with Machine Learning Using Amazon SageMaker, focuses on getting our feet wet on training and deploying an ML model using Amazon SageMaker. You will perform a simplified end-to-end ML experiment using Amazon SageMaker and the Linear Learner built-in algorithm.

Chapter 2, Building and Using Your Own Algorithm Container Image, is devoted to helping us understand how model training and deployment with Amazon SageMaker works internally. You will work on creating and using your own algorithm container images and scripts to train and deploy a custom ML model in SageMaker.

Chapter 3, Using Machine Learning and Deep Learning Frameworks with Amazon SageMaker, teaches you how to define, train, and deploy your own models using several ML and deep learning frameworks with the Amazon SageMaker Python SDK. This will allow you to use any custom models you have prepared using libraries and frameworks such as TensorFlow, Keras, scikit-learn, and PyTorch, and port these to SageMaker.

Chapter 4, Preparing, Processing, and Analyzing the Data, explores the different techniques and solutions you can use to handle your different data processing and analysis requirements. You will work with recipes that make use of SageMaker Processing, Amazon Athena, and several unsupervised built-in SageMaker algorithms to perform various data preparation and processing tasks.

Chapter 5, Effectively Managing Machine Learning Experiments, provides practical solutions and examples on debugging and managing ML experiments. You will use SageMaker Debugger to detect issues in your training jobs. In addition to this, you will work with SageMaker Experiments to manage and track multiple experiments at the same time.

Chapter 6, Automated Machine Learning in Amazon SageMaker, reveals the capabilities and features of SageMaker that help us build, train, and tune ML models automatically. You will take a closer look at using AutoML in SageMaker using SageMaker Autopilot. In addition to this, you will use and configure the automatic model tuning capability to search for the optimal set of hyperparameter values for our model.

Chapter 7, Working with SageMaker Feature Store, SageMaker Clarify, and SageMaker Model Monitor, exposes a few more capabilities of SageMaker that have great integration with SageMaker Studio – SageMaker Feature Store, SageMaker Clarify, and SageMaker Model Monitor. These capabilities help data scientists and ML practitioners handle requirements that involve using online and offline feature stores, detecting bias in the data, enabling ML explainability, and monitoring a deployed model.

Chapter 8, Solving NLP, Image Classification, and Time-Series Forecasting Problems with Built-in Algorithms, is devoted to different solutions and recipes that make use of several built-in SageMaker algorithms to solve natural language processing (NLP), image classification, and time-series forecasting problems.

Chapter 9, Managing Machine Learning Workflows and Deployments, explores several intermediate solutions for real-time endpoint deployments and automated workflows. You will work on recipes that focus on deep learning model deployments for Hugging Face models, multi-model endpoint deployments, and ML workflows.

Previous Section
Next Section