DataHour: Introduction to Serving Machine Learning Models as Microservices

Deploying machine learning models can be a challenging task, especially when managing multiple models simultaneously. Traditional deployment approaches may struggle to handle such complexity. However, there are helpful tools like MLServer and Seldon Core available to address these challenges. MLServer, a Python library, facilitates the creation of machine learning microservices capable of hosting one or more models within the same service. On the other hand, Seldon Core enables the construction of inference graphs to manage A/B testing, shadow and canary deployment, feature transformations, and model monitoring. In this workshop, we'll explore how to leverage these open-source tools to deploy microservices tailored to various use cases and model configurations. Join us on this hands-on journey through the essential steps in the machine learning workflow, using diverse examples beyond the realm of the finance, music, and healthcare industries.