From the course: Executive Guide to Predictive Modeling Strategy at Scale
Join today to access over 22,600 courses taught by industry experts or purchase this course individually.
Scoring traditional ML models
From the course: Executive Guide to Predictive Modeling Strategy at Scale
Scoring traditional ML models
- [Instructor] There's a good reason why traditional machine learning methods have stood the test of time. They still work quite well and they tend to be highly scalable. Let's take a look. One of advantage of these more traditional techniques like logistic regression let's say or decision trees is that they're very transparent. They really tell you a story about the data. Also, they're fast. They're fast models to build generally speaking and they're fast at scoring. Also, super important to us in this course, they're easy to migrate. These are straightforward formulas, so it's not ridiculous even to contemplate manually moving it from one system to another or writing a simple parser to do so. For all these reasons, they are still among the most common choices. They might not be the newest or the most exciting, but they really do work in many situations. Let's go a level deeper. Our statistical models, again like binary logistic regression is a super common one, are going to produce…
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.