Shypd.aiThe largest open index of AI toolsPyTorch Geometric (PyG) is a library built on PyTorch for easily writing and training Graph Neural Networks (GNNs) for a wide range of applications related to structured data.
Trending
About
PyTorch Geometric (PyG) is a powerful and flexible library designed for deep learning on graphs and other irregular structures, often referred to as geometric deep learning. Built upon PyTorch, it provides an easy-to-use API for implementing and training various Graph Neural Network (GNN) architectures. Key features include mini-batch loaders for both small and large graphs, multi-GPU support, `torch.compile` and DataPipe integration, and a comprehensive collection of benchmark datasets. PyG also offers helpful transforms for learning on arbitrary graphs, 3D meshes, or point clouds, making it suitable for a diverse set of applications.
Capabilities
Pricing & Plans
Open Source
Free
FAQs
Trending
