Description

Hands-On Graph Neural Networks Using Python is a comprehensive guide for machine learning practitioners, data scientists, and students interested in learning about graph neural networks and their applications. The book begins with the fundamentals of graph theory and shows how to create graph datasets from tabular data. It then explores major GNN architectures and essential concepts such as graph convolution, self-attention, link prediction, and heterogeneous graphs. The book concludes with applications to solve real-life problems, enabling readers to build a professional portfolio. By the end of this book, readers will have learned to create graph datasets, implement GNNs using Python and PyTorch Geometric, and apply them to solve real-world problems, along with building and training GNN models for node and graph classification, link prediction, and much more!

The entire code with examples and use cases is freely available on GitHub.

Additionally, we recommend Maxime's blog posts on Graph Neural Networks:

• Graph Convolutional Networks: Introduction to GNNs
• Graph Attention Networks: Self-Attention for GNNs
• GraphSAGE: Scaling up Graph Neural Networks
• GIN: How to Design the Most Powerful Graph Neural Network