Model Deployment and Maintenance for Data Scientists
The machine learning pipeline doesn’t end at just building the model. This course will teach you how to deploy your machine learning models as application programming interface (API) endpoints, and the maintenance required to support the model.
What you'll learn
Machine learning models only become useful once they begin to support the business through a deployed application.
In this course, Model Deployment and Maintenance for Data Scientists, you’ll gain the ability to run, monitor, and optimize machine learning models in production.
First, you’ll explore options for deploying machine learning models as an API endpoint.
Next, you’ll discover metrics and KPIs for the model you will need to monitor.
Finally, you’ll learn how to iterate and improve on your model as time goes on.
When you’re finished with this course, you’ll have the skills and knowledge of deploying and maintaining machine learning models needed to productionalize your machine learning pipeline.
Table of contents
- Course Introduction and Prerequisites 3m
- Review of the Machine Learning Pipeline 3m
- The Insurance Dataset 3m
- Machine Learning Inferencing 4m
- Machine Learning and Containers 3m
- Demo: Machine Learning and Containers 2m
- Training and Inferencing 2m
- Demo: Deploying a Machine Learning Endpoint 4m
- Module Summary 1m
About the author
Miguel Saavedra is an author, solutions architect, and Instructor specializing in AWS, big data, automation, and security. He has worked in several companies in the finance/fintech, education, and medical industries as well as some government projects. He has published and conducted research on both cloud and on-premise solutions for big data focusing on network analytics, and machine learning. He also designs highly available and automated CICD toolchains for high throughput microservices on AW... more