From the course: Deploying Scalable Machine Learning for Data Science
Unlock the full course today
Join today to access over 22,600 courses taught by industry experts or purchase this course individually.
Containers bundle ML model components
From the course: Deploying Scalable Machine Learning for Data Science
Containers bundle ML model components
- [Narrator] One of the important tasks that we have to consider when scaling machine learning models is how will we deploy our code. Let's review some of our requirements with regards to scaling machine learning models. We want our applications to scale, but we also want to use computing resources efficiently. So we'll use horizontal scaling. Horizontal scaling involves using clusters of servers. And the size of the clusters can change depending on demand. That's how we get both the resources we need to scale and the cost-efficiencies of only deploying resources that we need. Let's consider two ways we could deploy code to this kind of environment. With option one, we install each piece of software that we need on a server or virtual machine as we deploy those machines. Option two is that we use a container that is prebuilt with all the components that we need. Now, let's look more at each of these options. If we opt for option one, there are several things we need to do. We need to…
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.