Source: ZHUAN ZHI
This article serves as a tutorial and is recommended for a 5-minute read.
This tutorial's practical section will be based on TF-GNN, a library for handling graph-structured data in TensorFlow.
Graphs are universal data structures that can represent information from various domains (social, biomedical, online transactions, etc.). Graph Neural Networks (GNNs) are an exciting approach that uses graph-structured data within neural network models, which has gained significant popularity recently. However, implementing and running GNNs on large (and complex) datasets still poses many challenges for machine learning platforms.
Thank you for your interest in our tutorial! The main goal of this tutorial is to assist practitioners and researchers in implementing GNNs in a TensorFlow setting. Specifically, this tutorial will be primarily practical and will guide the audience through the process of running existing GNNs on heterogeneous graph data and introduce how to implement new GNN models. The practical section of this tutorial will be based on TF-GNN, a library for handling graph-structured data in TensorFlow.
https://github.com/tensorflow/gnn/tree/main/examples/tutorials/neurips_2022
