Chapter 13: Model Governance and MLOps

In the previous chapters, we learned how to build, understand, and deploy models. We will now learn how to govern these models and how to responsibly use these models in operations. In earlier chapters, we discussed the methods for understanding the business problem, the system in which the model will operate, and the potential consequences of using a model's predictions. MLOps is a word made up of machine learning and DevOps. It is made of processes and practices to efficiently, reliably, and effectively operationalize the production of machine learning (ML) models within an enterprise. MLOps aims to ensure commercial value and regulatory requirements are met continuously by ensuring production models' outcomes are of good quality and automation is in place. It provides a centralized system to manage the entire life cycle of all ML models in production.

Activities within MLOps cover all aspects of model deployment, provide real-time...