1 Overview of graph neural networks (10.49 MB)
2 Prerequisites (1.55 MB)
1 Message passing in GNNs (6.62 MB)
2 Aggregation and transformation math (7.15 MB)
3 Aggregation and transformation math in matrix form (6.2 MB)
1 Introducing graph attention (5.47 MB)
2 Computing the attention coefficient (6.58 MB)
3 Including attention in GNN layers (4.61 MB)
4 Getting set up with Colab and the PyTorch Geometric library (5.95 MB)
5 Exploring the Cora dataset (12.09 MB)
6 Setting up the graph convolutional network (9.67 MB)
7 Training a graph convolutional network (13.84 MB)
8 Node classification using a graph attention network (15.09 MB)
9 Using the GATv2Conv layer for attention (9.69 MB)
1 Understanding graph classification (10.04 MB)
2 Exploring the PROTEINS Dataset for graph classification (9.17 MB)
3 Minibatching graph data (5.96 MB)
4 Setting up a graph classification model (9.84 MB)
5 Training a GNN for graph classification (8.81 MB)
6 Eliminating neighborhood normalization and skip connections (7.27 MB)
1 A quick overview of autoencoders (5.23 MB)
2 Introducing graph autoencoders (4.52 MB)
3 Splitting link prediction data (12.08 MB)
4 Understanding link splits (13.76 MB)
5 Designing an autoencoder for link prediction (12.33 MB)
6 Training the autoencoder (15.04 MB)
1 Summary and next steps (2.64 MB)
Ex Files Advanced Graph Neural Networks (644.04 KB)