From the course: Learning Graph Neural Networks
Join today to access over 24,400 courses taught by industry experts.
Exercise: The CiteSeer dataset for node classification
From the course: Learning Graph Neural Networks
Exercise: The CiteSeer dataset for node classification
- [Instructor] In this demo, we'll work with citation data represented as a graph and we'll use it in a node classification task. The nodes of the graph will represent research papers and every research paper is categorized into a class or category. So this citation dataset that we are going to be working with is similar to the Cora dataset that we'd explored earlier. In our node classification problem, we'll first perform classification using a dense neural network that only takes into account the node features. We'll use that as our baseline model. We'll then perform classification using a graph convolutional network and see how the classification model is much improved when we use GNS instead of regular DNS. Now here I'm in a new notebook, NodeClassificationWithCiteseerData. I'm going to change the runtime type so that I'm no longer using the CPU. Instead, I use the T4 GPU to run my code. Now you can run using the CPU if you want to. The dataset is small enough that it does not…
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.
