From the course: Deploying Scalable Machine Learning for Data Science
Join today to access over 24,700 courses taught by industry experts.
Best practices for scaling ML
From the course: Deploying Scalable Machine Learning for Data Science
Best practices for scaling ML
- [Instructor] Now that we've reviewed the main parts of this course, it's a good time to review some best practices for scaling machine learning models. One of the first things we wanna keep in mind is that we often want to take advantage of horizontal scaling. Again, this is different from vertical scaling. Horizontal scaling works by adding resources like servers or containers to our cluster and then removing them as load drops down. This has the advantage over vertical scaling in that we don't actually have to change our underlying hardware or try to fit everything into one server. In addition, we wanna deploy our models with an API and the reason for that is an API makes our models accessible to other programs. Machine learning models in production don't stand alone. They're not islands. They inter-operate and work with other applications. APIs are the key to making them accessible to those other applications. We also wanna use containers. Containers allow us to distribute our…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
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.