ebook img

Azure AI Services at Scale for Cloud, Mobile, and Edge: Building Intelligent Apps with Azure Cognitive Services and Machine Learning PDF

227 Pages·2022·46.641 MB·English
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Azure AI Services at Scale for Cloud, Mobile, and Edge: Building Intelligent Apps with Azure Cognitive Services and Machine Learning

Azure AI Services at Scale for Cloud, Mobile, and Edge Building Intelligent Apps with Azure Cognitive Services and Machine Learning B is s o Hn od, B era r &ns Aco nm a nb de , Simon Bisson, Mary Branscombe, Chris Hoder & Anand Raman Azure AI Services at Scale for Cloud, Mobile, and Edge Take advantage of the power of cloud and the latest AI techniques. Whether you’re an experienced developer “An invaluable wanting to improve your app with AI-powered features or you want to make a business process smarter by getting AI introduction to to do some of the work, this book’s got you covered. Simon Microsoft’s family Bisson, Mary Branscombe, Chris Hoder, and Anand Raman of AI services.” show you how to build practical intelligent applications for —Rik Hepworth the cloud, mobile, browsers, and edge devices using a hands- Chief Consulting Officer, Black Marble on approach. This book shows you how cloud AI services fit in alongside familiar software development approaches, walks you Simon Bisson is a freelance writer, through key Microsoft AI services, and provides real-world specializing in enterprise technologies examples of AI-oriented architectures that integrate different and development. He writes for several Azure AI services. All you need to get started is a working publications, including InfoWorld, ZDNet, and Computerworld. knowledge of basic cloud concepts. Mary Branscombe has worked as • Become familiar with Azure AI offerings and capabilities a freelance technology journalist • Build intelligent applications using Azure Cognitive Services for many publications over three decades. Recently she’s specialized • Train, tune, and deploy models with Azure Machine Learning, in AI, enterprise technology, and PyTorch, and the Open Neural Network Exchange (ONNX) development. • Learn to solve business problems using AI in the Power Chris Hoder is a principal program Platform manager on the cognitive services team at Microsoft. He’s responsible • Use transfer learning to train vision, speech, and language for product development of the Azure models in minutes OpenAI Service. B is Anand Raman leads program s o management for the AI services Hn platform at Microsoft. Previously, he od, B was the chief of staff for the Microsoft era Azure AI and Data Group. r &ns Aco nm a nb de , MACHINE LEARNING Twitter: @oreillymedia linkedin.com/company/oreilly-media US $59.99 CAN $74.99 youtube.com/oreillymedia ISBN: 978-1-098-10804-5 Praise for Azure AI Services at Scale for Cloud, Mobile, and Edge Creating modern applications powered by machine learning and AI is too hard, even when you have the tools at hand to do it. In this book, Bisson, Branscombe, Hoder, and Anand show how much simpler it can be, with up-to-date and clear guidance on how to use the capabilities already built into Azure. —John Montgomery, CVP, Azure AI Products, Microsoft This book is an easy read for anyone looking to understand what Microsoft’s AI services can do for them, how to use the AI services in their business and applications, and how they should use AI responsibly for good. Highly recommended for anyone looking to build intelligent applications at scale using Microsoft Azure AI services. —Sundar Srinivasan, VP, Search Technology Center India (STCI), Microsoft An invaluable introduction to Microsoft’s family of AI services, with the added bonus of a thought-provoking section on using AI responsibly. A genuine something for everyone reference. Recommended without reservation. —Rik Hepworth, Chief Consulting Officer, Black Marble This book on Azure AI services is a must read for anyone looking for guidance on building intelligent applications and implementing an AI-oriented architecture for their organizations. Congratulations to the authors for delivering an insightful AI book that has fused the nuts and bolts of AI with years of practical on-the-ground experience. —Wee Hyong Tok, O’Reilly Book Author and Principal GPM, Microsoft The time has clearly come for building AI-oriented applications, to tap into their immense potential, and to do so in a trustworthy, responsible way. This book lays out a compelling path that will help innovators create the next generation of intelligent systems using Azure’s popular AI services in a way that will be both highly effective and sustainable. —Dion Hinchcliffe, VP and Principal Analyst, Constellation Research This book is a one-stop-shop for what, how, why, and “whether you should” questions for those looking to benefit from the democratization of AI offered by Microsoft’s cloud services. —Eric Boyd, CVP, Azure AI Platform, Microsoft Azure AI Services at Scale for Cloud, Mobile, and Edge Building Intelligent Apps with Azure , Cognitive Services and Machine Learning Simon Bisson, Mary Branscombe, Chris Hoder, and Anand Raman BBeeiijjiinngg BBoossttoonn FFaarrnnhhaamm SSeebbaassttooppooll TTookkyyoo Azure AI Services at Scale for Cloud, Mobile, and Edge by Simon Bisson, Mary Branscombe, Chris Hoder, and Anand Raman Copyright © 2022 O’Reilly Media, Inc. All rights reserved. Printed in the United States of America. Published by O’Reilly Media, Inc., 1005 Gravenstein Highway North, Sebastopol, CA 95472. O’Reilly books may be purchased for educational, business, or sales promotional use. Online editions are also available for most titles (http://oreilly.com). For more information, contact our corporate/institutional sales department: 800-998-9938 or [email protected]. Acquisitions Editor: Rebecca Novack Indexer: Ellen Troutman-Zaig Development Editor: Gary O’Brien Interior Designer: David Futato Production Editor: Beth Kelly Cover Designer: Karen Montgomery Copyeditor: nSight, Inc. Illustrator: Kate Dullea Proofreader: Piper Editorial Consulting, LLC April 2022: First Edition Revision History for the First Edition 2022-04-11: First Release See http://oreilly.com/catalog/errata.csp?isbn=9781098108045 for release details. The O’Reilly logo is a registered trademark of O’Reilly Media, Inc. Azure AI Services at Scale for Cloud, Mobile, and Edge, the cover image, and related trade dress are trademarks of O’Reilly Media, Inc. The views expressed in this work are those of the authors, and do not represent the publisher’s views. While the publisher and the authors have used good faith efforts to ensure that the information and instructions contained in this work are accurate, the publisher and the authors disclaim all responsibility for errors or omissions, including without limitation responsibility for damages resulting from the use of or reliance on this work. Use of the information and instructions contained in this work is at your own risk. If any code samples or other technology this work contains or describes is subject to open source licenses or the intellectual property rights of others, it is your responsibility to ensure that your use thereof complies with such licenses and/or rights. This work is part of a collaboration between O’Reilly and Microsoft. See our statement of editorial independence. 978-1-098-10804-5 [LSI] Table of Contents Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Part I. Understanding AI-Oriented Architecture 1. An Introduction to AI-Oriented Architecture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 What You Can Do with AI 3 From Milestones to Models to Architectures 4 Ready to Jump In? 7 Part II. Tools and Services to Help You Build AI-Oriented Architectures 2. Understanding AI Offerings and Capabilities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 AI Services for All Types of Users 11 Microsoft’s AI Offerings 14 Managed AI Services and Infrastructure Options in Azure 15 Business Platforms with Extensible AI 17 AI for Big Data and Relational Data 17 Making Machine Learning More Portable 18 Cognitive Services 19 How to Determine What Tool Is Best for You 21 3. Train, Tune, and Deploy Models with Azure Machine Learning, ONNX, and PyTorch. . 27 Understanding Azure Machine Learning 27 Understanding Azure Machine Learning Studio 28 Getting Started with Azure Machine Learning 29 Setting Up a Machine Learning Environment 29 v Integration with Azure Services 31 Using Visual Studio Code 32 The Azure Machine Learning Python SDK for Local Development 33 Azure Machine Learning and R 35 Build Your First Model Using Azure Machine Learning Studio 37 Use Automated Machine Learning 37 Using Designer 38 Using Azure Machine Learning with Notebooks and Python 39 Working with Azure Machine Learning Using Different Machine Learning Frameworks 41 An Introduction to MLOps 43 Logging in Azure Machine Learning 45 Tuning Using Hyperparameters 45 Exporting with ONNX 46 Using ONNX with WinML 47 Using ONNX in Machine Learning Container Runtimes 47 Wrapping It Up 48 4. Using Azure Cognitive Services to Build Intelligent Applications. . . . . . . . . . . . . . . . . . 49 Using Prebuilt AI 49 The Core Azure Cognitive Services 54 Language 55 Azure OpenAI Service 61 Speech 64 Vision 69 Decision Making 74 Wrapping It Up 77 5. Using Azure Applied AI Services for Common Scenarios. . . . . . . . . . . . . . . . . . . . . . . . . . 79 Azure Applied AI Services 79 Azure Video Analyzer 81 Cognitive Search 83 Azure Form Recognizer 86 Azure Bot Service 91 Immersive Reader 93 Use Transfer Learning to Train Vision, Speech, and Language Models in Minutes 94 Creating a Custom Vision Model 95 Creating a Custom Speech Model 99 Wrapping It Up 101 vi | Table of Contents 6. Machine Learning for Everyone: Low-Code and No-Code Experiences. . . . . . . . . . . . 103 The Microsoft Power Platform 104 Power BI and AI 105 AI Visualizations in Power BI 107 Using AI for Data Preparation in Power BI 109 Working with Custom Machine Learning Models in Power BI 110 Building Your Own Custom Models in Power BI 111 AI Builder 113 Training a Custom Form Processing Model 116 Using AI Builder Models 119 Using Cognitive Services and Other AI Models in Power Automate 126 Logic Apps and AI 134 Wrapping It Up 138 7. Responsible AI Development and Use. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139 Understanding Responsible AI 141 Responsible AI Improves Performance and Outcomes 142 Experiment and Iterate 143 Tools for Delivering Responsible AI 143 Tools for Transparency 145 Tools for AI Fairness 150 Tools for Reliability and Understanding Error 152 Human in the Loop Oversight 153 Wrapping It Up 155 Further Resources 156 8. Best Practices for Machine Learning Projects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 Working Well with Data 157 Sharing Data 158 Data Provenance and Governance 158 Making Machine Learning Projects Successful 162 Preparing Your Dataset 163 Establish Performance Metrics 165 Transparency and Trust 167 Experiment, Update, and Move On 167 Collaboration, Not Silos 168 Wrapping It Up 168 Table of Contents | vii Part III. AI-Oriented Architectures in the Real World 9. How Microsoft Runs Cognitive Services for Millions of Users. . . . . . . . . . . . . . . . . . . . 171 AI for Anyone 172 Clusters and Containers 174 10. Seeing AI: Using Azure Machine Learning and Cognitive Services in a Mobile App at Scale. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 Custom and Cloud Models 178 The Seeing AI Backend 180 Getting the Interface Right 181 11. Translating Multiple Languages at Scale for International Organizations. . . . . . . . . 183 Delivering Translations for an International Parliament 183 Connecting to Existing Audio-Visual (AV) Systems 184 Using Custom Speech Recognition for Specialized Vocabularies 184 From Specialized Prototype to General Application 185 Working within Constraints 186 12. Bringing Reinforcement Learning from the Lab to the Convenience Store. . . . . . . . 187 Two APIs, Eight Weeks, 100% Uplift 188 Afterword. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195 viii | Table of Contents

See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.