Take advantage of the power of cloud and the latest AI techniques. Whether you're an experienced developer wanting to improve your app with AI-powered features or you want to make a business process smarter by getting AI to do some of the work, this book's got you covered. Authors Anand Raman, Chris Hoder, Simon Bisson, and Mary Branscombe show you how to build practical intelligent applications for the cloud, mobile, browsers, and edge devices using a hands-on approach.
This book shows you how cloud AI services fit in alongside familiar software development approaches, walks you through key Microsoft AI services, and provides real-world examples of AI-oriented architectures that integrate different Azure AI services. All you need to get started is a working knowledge of basic cloud concepts.
• Become familiar with Azure AI offerings and capabilities
• Build intelligent applications using Azure Cognitive Services
• Train, tune, and deploy models with Azure Machine Learning, PyTorch, and the Open Neural Network Exchange (ONNX)
• Learn to solve business problems using AI in the Power Platform
• Use transfer learning to train vision, speech, and language models in minutes
Author(s): Simon Bisson, Mary Branscombe, Chris Hoder, Anand Raman
Edition: 1
Publisher: O'Reilly Media
Year: 2022
Language: English
Commentary: Vector PDF
Pages: 228
City: Sebastopol, CA
Tags: Microsoft Azure; Artificial Intelligence; Machine Learning; Python; PyTorch; Azure Cognitive Services
Copyright
Table of Contents
Preface
Who This Book Is For
How to Use This Book
Conventions Used in This Book
Using Code Examples
O’Reilly Online Learning
How to Contact Us
Acknowledgments
Part I. Understanding AI-Oriented Architecture
Chapter 1. An Introduction to AI-Oriented Architecture
What You Can Do with AI
From Milestones to Models to Architectures
Ready to Jump In?
Part II. Tools and Services to Help You Build AI-Oriented Architectures
Chapter 2. Understanding AI Offerings and Capabilities
AI Services for All Types of Users
Microsoft’s AI Offerings
Managed AI Services and Infrastructure Options in Azure
Business Platforms with Extensible AI
AI for Big Data and Relational Data
Making Machine Learning More Portable
Cognitive Services
How to Determine What Tool Is Best for You
Chapter 3. Train, Tune, and Deploy Models with Azure Machine Learning, ONNX, and PyTorch
Understanding Azure Machine Learning
Understanding Azure Machine Learning Studio
Getting Started with Azure Machine Learning
Setting Up a Machine Learning Environment
Integration with Azure Services
Using Visual Studio Code
The Azure Machine Learning Python SDK for Local Development
Azure Machine Learning and R
Build Your First Model Using Azure Machine Learning Studio
Use Automated Machine Learning
Using Designer
Using Azure Machine Learning with Notebooks and Python
Working with Azure Machine Learning Using Different Machine Learning Frameworks
An Introduction to MLOps
Logging in Azure Machine Learning
Tuning Using Hyperparameters
Exporting with ONNX
Using ONNX with WinML
Using ONNX in Machine Learning Container Runtimes
Wrapping It Up
Chapter 4. Using Azure Cognitive Services to Build Intelligent Applications
Using Prebuilt AI
The Core Azure Cognitive Services
Language
Azure OpenAI Service
Speech
Vision
Decision Making
Wrapping It Up
Chapter 5. Using Azure Applied AI Services for Common Scenarios
Azure Applied AI Services
Azure Video Analyzer
Cognitive Search
Azure Form Recognizer
Azure Bot Service
Immersive Reader
Use Transfer Learning to Train Vision, Speech, and Language Models in Minutes
Creating a Custom Vision Model
Creating a Custom Speech Model
Wrapping It Up
Chapter 6. Machine Learning for Everyone: Low-Code and No-Code Experiences
The Microsoft Power Platform
Power BI and AI
AI Visualizations in Power BI
Using AI for Data Preparation in Power BI
Working with Custom Machine Learning Models in Power BI
Building Your Own Custom Models in Power BI
AI Builder
Training a Custom Form Processing Model
Using AI Builder Models
Using Cognitive Services and Other AI Models in Power Automate
Logic Apps and AI
Wrapping It Up
Chapter 7. Responsible AI Development and Use
Understanding Responsible AI
Responsible AI Improves Performance and Outcomes
Experiment and Iterate
Tools for Delivering Responsible AI
Tools for Transparency
Tools for AI Fairness
Tools for Reliability and Understanding Error
Human in the Loop Oversight
Wrapping It Up
Further Resources
Chapter 8. Best Practices for Machine Learning Projects
Working Well with Data
Sharing Data
Data Provenance and Governance
Making Machine Learning Projects Successful
Preparing Your Dataset
Establish Performance Metrics
Transparency and Trust
Experiment, Update, and Move On
Collaboration, Not Silos
Wrapping It Up
Part III. AI-Oriented Architectures in the Real World
Chapter 9. How Microsoft Runs Cognitive Services for Millions of Users
AI for Anyone
Clusters and Containers
Chapter 10. Seeing AI: Using Azure Machine Learning and Cognitive Services in a Mobile App at Scale
Custom and Cloud Models
The Seeing AI Backend
Getting the Interface Right
Chapter 11. Translating Multiple Languages at Scale for International Organizations
Delivering Translations for an International Parliament
Connecting to Existing Audio-Visual (AV) Systems
Using Custom Speech Recognition for Specialized Vocabularies
From Specialized Prototype to General Application
Working within Constraints
Chapter 12. Bringing Reinforcement Learning from the Lab to the Convenience Store
Two APIs, Eight Weeks, 100% Uplift
Afterword
Index
About the Authors
Colophon
