From the course: Building Recommender Systems with Machine Learning and AI
Unlock this course with a free trial
Join today to access over 23,200 courses taught by industry experts.
Running SVD and SVD++ on MovieLens - Python Tutorial
From the course: Building Recommender Systems with Machine Learning and AI
Running SVD and SVD++ on MovieLens
- [Instructor] That's definitely enough theory, let's see SVD in action and recommend some movies with it. So, open up Spyder. And close out any existing windows so we don't get confused. A shortcut for that, by the way, is CTRL-SHIFT-W. Open up everything inside the Matrixfactorization folder inside your course materials. Let's start by looking at the SVDbakeoff.py file. Surpriselib includes a couple of different SVD implementations. One is SVD more or less as we've described it, the other is called SVD++. It's a slight variant of SVD that ended up being part of the winning system for the Netflix prize. The difference is fairly subtle. It has to do with the actual loss function that is used while running stochastic gradient descent. In SVD++, this loss function takes into account the idea that merely rating an item at all, is some sort of implicit interest in the item, no matter what the rating was. But that was enough to make SVD good enough to win. There's not much to talk about in…
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.