Efficient Graph Deep Learning in TensorFlow with tf_geometric

We introduce tfgeometric, an efficient and friendly library for graph deep learning, which is compatible with both TensorFlow 1.x and 2.x. tfgeometric provides kernel libraries for building Graph Neural Networks (GNNs) as well as implementations of popular GNNs. The kernel libraries consist of infrastructures for building efficient GNNs, including graph data structures, graph map-reduce framework, graph mini-batch strategy, etc. These infrastructures enable tfgeometric to support single-graph computation, multi-graph computation, graph mini-batch, distributed training, etc.; therefore, tfgeometric can be used for a variety of graph deep learning tasks, such as transductive node classification, inductive node classification, link prediction, and graph classification. Based on the kernel libraries, tfgeometric implements a variety of popular GNN models for different tasks. To facilitate the implementation of GNNs, tfgeometric also provides some other libraries for dataset management, graph sampling, etc. Different from existing popular GNN libraries, tfgeometric provides not only Object-Oriented Programming (OOP) APIs, but also Functional APIs, which enable tfgeometric to handle advanced graph deep learning tasks such as graph meta-learning. The APIs of tfgeometric are friendly, and they are suitable for both beginners and experts. In this paper, we first present an overview of tfgeometric's framework. Then, we conduct experiments on some benchmark datasets and report the performance of several popular GNN models implemented by tf_geometric.