From the course: DevOps for Data Scientists
Join today to access over 22,600 courses taught by industry experts or purchase this course individually.
Overview of DevOps best practices for data science
From the course: DevOps for Data Scientists
Overview of DevOps best practices for data science
- [Instructor] Now let's close out this course with an overview of best practices. We'll focus first on development practices. First, be sure to use version control. This will help you track your code, and it makes collaboration much easier than working without version control. Number two, if you plan to exchange models or use different platforms for development and production, consider using the PMML or Predictive Model Markup Language as an interchange format. And then finally, use a continuous integration tool like Jenkins. This will improve the speed and the pace at which you can release new code. Now, with regards to operation practices, use staging environments and canary deploys, test your code, and roll it out in a limited way. This will isolate or minimize the disruption if code has to be rolled back. Also, keep in mind that data models are software and all software needs to be secured. Think in terms of authentication and authorization, as well as good security development…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.