- Using Git for version control
- Incorporating model testing into the deployment process
- Working with the Predictive Model Markup Language
- Securing the data science models in production
- Monitoring models in production
- Creating a Dockerfile for data science models
Skill Level Intermediate
- [Instructor] Welcome to DevOps for Data Science. This course is designed to provide data scientists with an introduction to DevOps or development and operations. DevOps incorporates many ideas from agile software methodologies. We'll start by reviewing common data science tasks such as collecting and organizing data, engineering features, building models and then validating them. We'll review a few ways to help streamline putting models into production with a look at version control for managing code with Git, the predictive model markup language for exchanging models between platforms and the continuous integration tool, Jenkins, that can help with testing and deploying code.
You'll learn about deployment practices such as using staging environments and canary deployments for integration testing and limited releases. We'll also discuss security and monitoring practices that are required when deploying to production. You'll also learn about Docker, a popular platform from running applications like data science models within containers. Containers are increasingly popular in DevOps because they streamline system deployments and operational management. So let's get started with DevOps for data science.