Chapter 4: Continuous Integration and Continuous Delivery (CI/CD) for Machine Learning

While working through the code examples, in both Chapter 2, Automating Machine Learning Model Development Using SageMaker Autopilot, and Chapter 3, Automating Complicated Model Development with AutoGluon, for the age calculator use case, you would've noticed a common trend that highlighted a drawback in using either the Autopilot or AutoGluon methodologies – specifically, that there is a disconnect in both processes between creating a production-grade ML (machine learning) model and then actually deploying the model into production.

Whether an ML practitioner leverages the CRISP-DM methodology or an AutoML methodology, the scope of their responsibilities ends once they have produced an optimal ML model. After their task is complete, the ML practitioner simply hands the model over to the various teams responsible for deploying and managing the model in production. This handover creates...