0%

Getting your models into production is the fundamental challenge of machine learning. MLOps offers a set of proven principles aimed at solving this problem in a reliable and automated way. This insightful guide takes you through what MLOps is (and how it differs from DevOps) and shows you how to put it into practice to operationalize your machine learning models.

Current and aspiring machine learning engineers--or anyone familiar with data science and Python--will build a foundation in MLOps tools and methods (along with AutoML and monitoring and logging), then learn how to implement them in AWS, Microsoft Azure, and Google Cloud. The faster you deliver a machine learning system that works, the faster you can focus on the business problems you're trying to crack. This book gives you a head start.

You'll discover how to:

  • Apply DevOps best practices to machine learning
  • Build production machine learning systems and maintain them
  • Monitor, instrument, load-test, and operationalize machine learning systems
  • Choose the correct MLOps tools for a given machine learning task
  • Run machine learning models on a variety of platforms and devices, including mobile phones and specialized hardware

Table of Contents

  1. Preface
    1. Why We Wrote This Book
    2. How This Book Is Organized
    3. Chapters
    4. Appendix
    5. Exercise Questions
    6. Discussion Questions
    7. Origin of Chapter Quotes
    8. Conventions Used in This Book
    9. Using Code Examples
    10. O’Reilly Online Learning
    11. How to Contact Us
    12. Acknowledgements
    13. From Noah
    14. From Alfredo
  2. 1. Introduction to MLOps
    1. Rise of the Machine Learning Engineer and MLOps
    2. What is MLOps
    3. DevOps and MLOps
    4. A MLOps Hierarchy of Needs
    5. Implementing DevOps
    6. DataOps and Data Engineering
    7. Platform Automation
    8. MLOps
    9. Conclusion
    10. Exercises
    11. Critical Thinking Discussion Questions
  3. 2. MLOps Foundations
    1. Bash and the Linux Command Line
    2. Cloud Shell Development Environments
    3. Bash Shell and Commands
    4. List Files
    5. Run Commands
    6. Files and Navigation
    7. Input/Output
    8. Configuration
    9. Writing a Script
    10. Cloud Computing Foundations & Building Blocks
    11. Getting Started with Cloud Computing
    12. Python Crash Course
    13. Minimilistic Python Tutorial
    14. Math for Programmers Crash Course
    15. Descriptive Statistics and Normal Distributions
    16. Optimization
    17. Machine Learning Key Concepts
    18. Doing Data Science
    19. Build an MLOps Pipeline from Zero
    20. Conclusion
    21. Exercises
    22. Critical Thinking Discussion Questions
  4. 3. MLOps for Containers and Edge Devices
    1. Containers
    2. Container Runtime
    3. Creating a Container
    4. Running a Container
    5. Best practices
    6. Serving a trained model over HTTP
    7. Edge Devices
    8. Coral
    9. Azure Percept
    10. TFHub
    11. Porting over non-TPU models
    12. Containers for Managed ML Systems
    13. Containers in Monetizing MLOps
    14. Build Once Run Many MLOps Workflow
    15. Conclusion
    16. Exercises
    17. Critical Thinking Discussion Questions
  5. 4. Continuous Delivery for Machine Learning Models
    1. Packaging for ML Models
    2. Infrastructure as Code for Continuous Delivery of ML Models
    3. Using Cloud Pipelines
    4. Controlled Rollout of models
    5. Testing techniques for Model Deployment
    6. Conclusion
    7. Exercises
    8. Critical Thinking Discussion Questions
  6. 5. AutoML and KaizenML
    1. AutoML
    2. MLOps Industrial Revolution
    3. Kaizen vs KaizenML
    4. Feature Stores
    5. Apple’s Ecosystem
    6. Apple’s AutoML: CreateML
    7. Apple’s Core ML Tools
    8. Google’s AutoML and Edge Computer Vision
    9. Azure’s AutoML
    10. AWS AutoML
    11. Open Source AutoML Solutions
    12. Ludwig
    13. FLAML
    14. Model Explainability
    15. Conclusion
    16. Exercises
    17. Critical Thinking Discussion Questions
  7. 6. Monitoring and Logging
    1. Observability for Cloud MLOps
    2. Introduction to Logging
    3. Logging in Python
    4. Modifying log levels
    5. Logging different applications
    6. Monitoring and Observability
    7. Basics of Model Monitoring
    8. Monitoring Drift with AWS SageMaker
    9. Monitoring Drift with Azure ML
    10. Conclusion
    11. Exercises
    12. Critical Thinking Discussion Questions
  8. 7. MLOps For AWS
    1. Introduction to AWS
    2. Getting Started with AWS Services
    3. MLOps on AWS
    4. MLOps Cookbook on AWS
    5. CLI Tools
    6. Flask Microservice
    7. AWS Lambda Recipes
    8. Deploy to AWS Lambda with SAM
    9. Applying AWS Machine Learning to the Real World
    10. Case Study: Sports Social Network
    11. Case Study: Career Advice with Julien Simon, AWS Machine Learning Evangelist
    12. Conclusion
    13. Exercises
    14. Critical Thinking Discussion Questions
  9. 8. MLOps for Azure
    1. Azure CLI and Python SDK
    2. Authentication
    3. Service Principal
    4. Authenticating API Services
    5. Compute Instances
    6. Deploying
    7. Registering Models
    8. Versioning datasets
    9. Deploying Models to a Compute Cluster
    10. Configuring a Cluster
    11. Deploying a Model
    12. Troubleshooting Deployment Issues
    13. Retrieve Logs
    14. Application Insights
    15. Debugging Locally
    16. Azure ML Pipelines
    17. Publishing Pipelines
    18. Azure Machine Learning Designer
    19. ML Lifecycle
    20. Conclusion
    21. Exercises
    22. Critical Thinking Discussion Questions
  10. 9. MLOps For GCP
    1. Google Cloud Platform Overview
    2. Continuous Integration and Continuous Delivery
    3. Kubernetes Hello World
    4. Cloud-Native Database Choice and Design
    5. DataOps on GCP: Applied Data Engineering
    6. Operationalizing ML Models
    7. Conclusion
    8. Exercises
    9. Critical Thinking Discussion Questions
  11. 10. Machine Learning Interoperability
    1. Why interoperability is critical
    2. ONNX: Open Neural Network Exchange
    3. ONNX Model Zoo
    4. Convert PyTorch into ONNX
    5. Create a generic ONNX checker
    6. Convert TensorFlow into ONNX
    7. Deploy ONNX to Azure
    8. Apple Core ML
    9. Edge Integration
    10. Conclusion
    11. Exercises
    12. Critical Thinking Discussion Questions
  12. 11. Building MLOps command-line tools and Microservices
    1. Python Packaging
    2. The requirements file
    3. Command-line Tools
    4. Creating a dataset linter
    5. Modularizing a command-line tool
    6. Microservices
    7. Creating a serverless function
    8. Authenticating to Cloud Functions
    9. Building a cloud-based CLI
    10. Machine Learning CLI Workflows
    11. Conclusion
    12. Exercises
    13. Critical Thinking Discussion Questions
  13. 12. Machine Learning Engineering and MLOps Case Studies
    1. Unlikely Benefits of Ignorance in Building Machine Learning Models
    2. MLOps Projects at Sqor Sports Social Network
    3. The Perfect Technique vs. The Real World
    4. Critical Challenges in MLOps
    5. Ethical and Unintended Consequences
    6. Lack of Operational Excellence
    7. Focus on Prediction Accuracy vs. the Big Picture
    8. Final Recommendations to Implement MLOPs
    9. Data Governance and Cybersecurity
    10. MLOps Design Patterns
    11. Conclusion
    12. Exercises
    13. Critical Thinking Discussion Questions
  14. A. Technology Certifications
    1. AWS Certifications
    2. AWS Cloud Practitioner and AWS Solutions Architect
    3. AWS Certified Machine Learning - Specialty
    4. Other Cloud Certifications
    5. Azure Data Scientist and AI Engineer
    6. GCP
    7. SQL Related Certifications
  15. B. Remote Work
    1. Equipment for Working Remote
    2. Network
    3. Home Work Area
    4. Location, Location, Location
  16. C. Think Like a VC for Your Career
    1. Pear Revenue Strategy
    2. Passive
    3. Positive
    4. Autonomy
    5. Exponential
    6. Rule of 25%
    7. NOTES
  17. D. Building a Technical Portfolio for MLOps
    1. Project: Continuous Delivery of Flask Data Engineering API
    2. Project: Docker & Kubernetes Container Project
    3. Project: Serverless AI Data Engineering Pipeline
    4. Project: Build Edge ML Solution
    5. Deliverables
    6. Project: Build Cloud-Native ML Application or API
    7. Project Selection
    8. Getting a Job: Don’t Storm the Castle, Walk in the backdoor
  18. E. Data Science Case Study: Intermittent Fasting
    1. Notes on Intermittent Fasting, Blood Glucose, and Food
  19. F. Key Terms
  20. G. Additional Educational Resources
    1. Additional MLOps Critical Thinking Questions
    2. Additional MLOps Educational Materials
    3. Education Disruption
    4. Current State of Higher Education That Will Be Disrupted
    5. Ten Times Better Education
    6. Conclusion
  21. H. Technical Project Management
    1. Project Plan
    2. Weekly Demo
    3. Task Tracking
34.238.138.162