Automated Machine Learning with Microsoft Azure: Build highly accurate and scalable end-to-end AI solutions with Azure AutoML

Description

A practical, step-by-step guide to using Microsoft's AutoML technology on the Azure Machine Learning service for developers and data scientists working with the Python programming language Key FeaturesCreate, deploy, productionalize, and scale automated machine learning solutions on Microsoft AzureImprove the accuracy of your ML models through automatic data featurization and model trainingIncrease productivity in your organization by using artificial intelligence to solve common problemsBook Description Automated Machine Learning with Microsoft Azure helps you to build high-performing, accurate machine learning models in record time. It allows anyone to easily harness the power of artificial intelligence and increase the productivity and profitability of your business. With a series of clicks on a guided user interface (GUI), novices and seasoned data scientists alike can train and deploy machine learning solutions to production with ease. This book will teach you how to use Azure AutoML with both the GUI as well as the AzureML Python software development kit (SDK) in a careful, step-by-step way. First, you'll learn how to prepare data, train models, and register them to your Azure Machine Learning workspace. You'll then discover how to take those models and use them to create both automated batch solutions using machine learning pipelines and real-time scoring solutions using Azure Kubernetes Service (AKS). Finally, you will be able to use AutoML on your own data to not only train regression, classification, and forecasting models but also use them to solve a wide variety of business problems. By the end of this Azure book, you'll be able to show your business partners exactly how your ML models are making predictions through automatically generated charts and graphs, earning their trust and respect. What you will learnUnderstand how to train classification, regression, and forecasting ML algorithms with Azure AutoMLPrepare data for Azure AutoML to ensure smooth model training and deploymentAdjust AutoML configuration settings to make your models as accurate as possibleDetermine when to use a batch-scoring solution versus a real-time scoring solutionProductionalize your AutoML solution with Azure Machine Learning pipelinesCreate real-time scoring solutions with AutoML and Azure Kubernetes ServiceDiscover how to quickly deliver value and earn business trust using AutoMLTrain a large number of AutoML models at once using the AzureML Python SDKWho this book is for Data scientists, aspiring data scientists, machine learning engineers, or anyone interested in applying artificial intelligence or machine learning in their business will find this book useful. You need to have beginner-level knowledge of artificial intelligence and a technical background in computer science, statistics, or information technology before getting started with this machine learning book. Familiarity with Python will help you implement this book's more advanced features, but even data analysts and SQL experts will be able to train ML models after finishing this book. Table of ContentsIntroducing AutoMLGetting Started with Azure Machine Learning ServiceTraining Your First AutoML ModelBuilding an AutoML Regression SolutionBuilding an AutoML Classification SolutionBuilding an AutoML Forecasting SolutionUsing the Many Models Solution AcceleratorChoosing Real-Time versus Batch ScoringImplementing a Batch Scoring SolutionCreating End-to-End AutoML SolutionsImplementing a Real-Time Scoring SolutionRealizing Business Value with AutoML Cover Title Page Copyright and Credits Dedication Contributors Table of Contents Preface Section 1: AutoML Explained – Why, What, and How Chapter 1: Introducing AutoML Explaining data science's ROI problem Defining machine learning, data science, and AI Machine learning versus traditional software The five steps to machine learning success Putting it all together Analyzing why AI projects fail slowly Solving the ROI problem with AutoML Summary Chapter 2: Getting Started with Azure Machine Learning Service Technical requirements Creating your first AMLS workspace Creating an Azure account Creating an AMLS workspace Creating an AMLS workspace with code Navigating AML studio Building compute to run your AutoML jobs Creating a compute instance Creating a compute cluster Creating a compute cluster and compute instance with the Azure CLI Working with data in AMLS Creating a dataset using the GUI Creating a dataset using code Understanding how AutoML works on Azure Ensuring data quality with data guardrails Improving data with intelligent feature engineering Normalizing data for ML with iterative data transformation Training models quickly with iterative ML model building Getting the best results with ML model ensembling Summary Chapter 3: Training Your First AutoML Model Technical requirements Loading data into AMLS for AutoML Creating an AutoML solution Interpreting your AutoML results Understanding data guardrails Understanding model metrics Explaining your AutoML model Obtaining better AutoML performance Summary Section 2: AutoML for Regression, Classification, and Forecasting – A Step-by-Step Guide Chapter 4: Building an AutoML Regression Solution Technical requirements Preparing data for AutoML regression Setting up your Jupyter environment Preparing your data for AutoML Training an AutoML regression model Registering your trained regression model Fine-tuning your AutoML regression model Improving AutoML regression models Understanding AutoML regression algorithms Summary Chapter 5: Building an AutoML Classification Solution Technical requirements Prepping data for AutoML classification Navigating to your Jupyter environment Loading and transforming your data Training an AutoML classification model Registering your trained classification model Training an AutoML multiclass model Fine-tuning your AutoML classification model Improving AutoML classification models Understanding AutoML classification algorithms Summary Chapter 6: Building an AutoML Forecasting Solution Technical requirements Prepping data for AutoML forecasting Navigating to your Jupyter environment Loading and transforming your data Training an AutoML forecasting model Training a forecasting model with standard algorithms Training a forecasting model with Prophet and ARIMA Registering your trained forecasting model Fine-tuning your AutoML forecasting model Improving AutoML forecasting models Understanding AutoML forecasting algorithms Summary Chapter 7: Using the Many Models Solution Accelerator Technical requirements Installing the many models solution accelerator Creating a new notebook in your Jupyter environment Installing the MMSA from GitHub Prepping data for many models Prepping the sample OJ dataset Prepping a pandas dataframe Training many models simultaneously Training the sample OJ dataset Training your sample dataset with the MMSA Scoring new data for many models Scoring OJ sales data with the MMSA Scoring your sample dataset with many models Improving your many models results Summary Section 3: AutoML in Production – Automating Real-Time and Batch Scoring Solutions Chapter 8: Choosing Real-Time versus Batch Scoring Technical requirements Architecting batch scoring solutions Understanding the five-step batch scoring process Scheduling your batch scoring solution Scoring data in batches and delivering results Choosing batch over real time Architecting real-time scoring solutions Understanding the four-step real-time scoring process Training a model for real-time deployment   Delivering results in real time Knowing when to use real-time scoring Choosing real-time over batch solutions Determining batch versus real-time scoring scenarios Scenarios for real-time or batch scoring Answers for the type of solution appropriate for each scenario Summary Chapter 9: Implementing a Batch Scoring Solution Technical requirements Creating an ML pipeline Coding the first three steps of your ML scoring pipeline Creating a Python script to score data in your ML pipeline Creating and containerizing an environment Configuring and running your ML scoring pipeline Accessing your scored predictions via AML studio Creating a parallel scoring pipeline Coding the first three steps of your ML parallel scoring pipeline Creating Python scripts to score data in your ML parallel pipeline Configuring and running your ML parallel scoring pipeline Creating an AutoML training pipeline Coding the first two steps of your AutoML training pipeline Configuring your AutoML model training settings and step Creating a Python script to register your model Configuring and running your AutoML training pipeline Triggering and scheduling your ML pipelines Triggering your published pipeline from the GUI Triggering and scheduling a published pipeline through code Summary Chapter 10: Creating End-to-End AutoML Solutions Technical requirements Connecting AMLS to ADF Creating an ADF Creating a service principal and granting access Creating a linked service to connect ADF with AMLS Scheduling a machine learning pipeline in ADF Transferring data using ADF Installing a self-hosted integration runtime Creating an Azure Blob storage linked service Creating a linked service to your PC Creating an ADF pipeline to copy data Automating an end-to-end scoring solution Editing an ML pipeline to score new data Creating an ADF pipeline to run your ML pipeline Adding a trigger to your ADF pipeline Automating an end-to-end training solution Creating a pipeline to copy data into Azure Editing an ML pipeline to train with new data Adding a Machine Learning Execute Pipeline activity to your ADF pipeline Summary Chapter 11: Implementing a Real-Time Scoring Solution Technical requirements Creating real-time endpoints through the UI Creating an ACI-hosted endpoint through the UI Creating an AKS cluster through the UI Creating an AKS-hosted endpoint through the UI Creating real-time endpoints through the SDK Creating and testing a real-time endpoint with ACI through Python Creating an AKS cluster through Python Creating and testing a real-time endpoint with AKS through Python Improving performance on your AKS cluster Summary Chapter 12: Realizing Business Value with AutoML Technical requirements Architecting AutoML solutions Making key architectural decisions for AutoML solutions Architecting a batch solution Architecting a real-time solution Visualizing AutoML modeling results Visualizing the results of classification Visualizing the results of forecasting and regression Explaining AutoML results to your business Using AutoML in other Microsoft products Using AutoML within PowerBI Using AutoML within Azure Synapse Analytics Using AutoML with ML.NET Using AutoML on SQL Server, HDInsight, and Azure Databricks Realizing business value Getting the business to adopt a new, automated solution Getting the business to replace an older, automated process Getting the business to adopt a new, decision-assistance tool Getting the business to replace an old decision assistance tool Summary Why subscribe? About Packt Other Books You May Enjoy Index