In this chapter, we covered ModelOps – a set of practices for automating a common set of operations that arise in data science projects. We explored how ModelOps relates to DevOps and described major steps in the ModelOps pipeline. We looked at strategies for managing code, versioning data, and sharing project environments between team members. We also examined the importance of experiment tracking and automated testing for data science projects. As a conclusion, we outlined the full CI/CD pipeline with continuous model training and explored a set of tools that can be used to build such pipelines.
In the next chapter, we will look at how to build and manage a data science technology stack.