From the course: Data-Centric Visual AI
Join today to access over 25,500 courses taught by industry experts.
Advanced model evaluation
From the course: Data-Centric Visual AI
Advanced model evaluation
- [Instructor] Hey, everyone. In this Notebook, we're going to be talking about advanced model evaluation. Evaluating models is a critical routine in any model production. It's easy to get caught up in those overall summary statistics, but, however, if a model cannot perform well in critical situations or edge cases, it likely will not be good enough to land in production yet, so in this module in Notebook, we're going to be talking about how you can find and test these edge cases, save them for later. That way, you can test those same edge cases for every model you create. Let's hop in with installation. We're going to be using the open source library FiftyOne again. However, we're also going to be using umap-learn and scikit-learn as well. These will come later on to help us kind of understand our data a little bit better. To load our data set, we're going to be using that VisDrone data set from the past Notebook. Again, for this example, for the sake of the time and with those…
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.