Practical Automated Machine Learning on Azure

Table of contents

1. Foreword
2. Preface
   1. Conventions Used in This Book
   2. Using Code Examples
   3. O’Reilly Online Learning
   4. How to Contact Us
   5. Acknowledgments
3. I. Automated Machine Learning
4. 1. Machine Learning: Overview and Best Practices
   1. Machine Learning: A Quick Refresher
      1. Model Parameters
      2. Hyperparameters
   2. Best Practices for Machine Learning Projects
      1. Understand the Decision Process
      2. Establish Performance Metrics
      3. Focus on Transparency to Gain Trust
      4. Embrace Experimentation
      5. Don’t Operate in a Silo
   3. An Iterative and Time-Consuming Process
      1. Feature Engineering
      2. Algorithm Selection
      3. Hyperparameter Tuning
      4. The End-to-End Process
   4. Growing Demand
   5. Conclusion
5. 2. How Automated Machine Learning Works
   1. What Is Automated Machine Learning?
      1. Understanding Data
      2. Detecting Tasks
      3. Choosing Evaluation Metrics
      4. Feature Engineering
      5. Selecting a Model
      6. Monitoring and Retraining
      7. Bringing It All Together
   2. Automated ML
      1. How Automated ML Works
      2. Preserving Privacy
      3. Enabling Transparency
      4. Guardrails
      5. End-to-End Model Life-Cycle Management
   3. Conclusion
6. II. Automated ML on Azure
7. 3. Getting Started with Microsoft Azure Machine Learning and Automated ML
   1. The Machine Learning Process
      1. Collaboration and Monitoring
      2. Deployment
   2. Setting Up an Azure Machine Learning Workspace for Automated ML
      1. Azure Notebooks
      2. Notebook VM
   3. Conclusion
8. 4. Feature Engineering and Automated Machine Learning
   1. Data Preprocessing Methods Available in Automated ML
   2. Auto-Featurization for Automated ML
      1. Auto-Featurization for Classification and Regression
      2. Auto-Featurization for Time-Series Forecasting
   3. Conclusion
9. 5. Deploying Automated Machine Learning Models
   1. Deploying Models
      1. Registering the Model
      2. Creating the Container Image
      3. Deploying the Model for Testing
      4. Testing a Deployed Model
      5. Deploying to AKS
   2. Swagger Documentation for the Web Service
   3. Debugging a Deployment
      1. Web Service Deployment Fails
   4. Conclusion
10. 6. Classification and Regression
    1. What Is Classification and Regression?
       1. Classification and Regression Algorithms
       2. Using Automated ML for Classification and Regression
    2. Conclusion
11. III. How Enterprises Are Using Automated Machine Learning
12. 7. Model Interpretability and Transparency with Automated ML
    1. Model Interpretability
       1. Model Interpretability with Azure Machine Learning
    2. Model Transparency
       1. Understanding the Automated ML Model Pipelines
       2. Guardrails
    3. Conclusion
13. 8. Automated ML for Developers
    1. Azure Databricks and Apache Spark
    2. ML.NET
    3. SQL Server
    4. Conclusion
14. 9. Automated ML for Everyone
    1. Azure Portal UI
    2. Power BI
       1. Preparing the Data
       2. Automated ML Training
       3. Understanding the Best Model
       4. Understanding the Automated ML Training Process
       5. Model Deployment and Inferencing
    3. Enabling Collaboration
       1. Azure Machine Learning to Power BI
       2. Power BI Automated ML to Azure Machine Learning
    4. Conclusion
15. Index