Book Image

Engineering MLOps

By : Emmanuel Raj
Book Image

Engineering MLOps

By: Emmanuel Raj

Overview of this book

Engineering MLps presents comprehensive insights into MLOps coupled with real-world examples in Azure to help you to write programs, train robust and scalable ML models, and build ML pipelines to train and deploy models securely in production. The book begins by familiarizing you with the MLOps workflow so you can start writing programs to train ML models. Then you’ll then move on to explore options for serializing and packaging ML models post-training to deploy them to facilitate machine learning inference, model interoperability, and end-to-end model traceability. You’ll learn how to build ML pipelines, continuous integration and continuous delivery (CI/CD) pipelines, and monitor pipelines to systematically build, deploy, monitor, and govern ML solutions for businesses and industries. Finally, you’ll apply the knowledge you’ve gained to build real-world projects. By the end of this ML book, you'll have a 360-degree view of MLOps and be ready to implement MLOps in your organization.
Table of Contents (18 chapters)
1
Section 1: Framework for Building Machine Learning Models
7
Section 2: Deploying Machine Learning Models at Scale
13
Section 3: Monitoring Machine Learning Models in Production

Why package ML models?

MLOps enables a systematic approach to train and evaluate models. After models are trained and evaluated, the next steps are to bring them to production. As we know, ML doesn't work like traditional software engineering, which is deterministic in nature and where a piece of code or module is imported into the existing system and it works. Engineering ML solutions is non-deterministic and involves serving ML models to make predictions or analyze data.

In order to serve the models, they need to be packed into software artifacts to be shipped to the testing or production environments. Usually, these software artifacts are packaged into a file or a bunch of files or containers. This allows the software to be environment- and deployment-agnostic. ML models need to be packaged for the following reasons:

Portability

Packaging ML models into software artifacts enables them to be shipped or transported from one environment to another. This can be done...