Agent skill · research science · davila7
torch-geometric
Graph Neural Networks (PyG). Node/graph classification, link prediction, GCN, GAT, GraphSAGE, heterogeneous graphs, molecular property prediction, for geometric deep learning.
Why this skill is useful
Provides domain-specific knowledge on graph neural networks and their applications that the AI wouldn't reliably generate on its own.
What it needs
Requires pyg_lib, torch_cluster, torch_geometric, torch_scatter, torch_sparse, torch_spline_conv installed locally. About 8k tokens when loaded. Last updated 2026-08-06. 30,138 stars on the source repository.
What this skill does
PyTorch Geometric (PyG) Overview PyTorch Geometric is a library built on PyTorch for developing and training Graph Neural Networks (GNNs). Apply this skill for deep learning on graphs and irregular structures, including mini-batch processing, multi-GPU training, and geometric deep learning applications. When to Use This Skill This skill should be used when working with: Graph-based machine learning: Node classification, graph classification, link prediction Molecular property prediction: Drug discovery, chemical property prediction Social network analysis: Community detection, influence prediction Citation networks: Paper classification, recommendation systems 3D geometric data: Point clouds, meshes, molecular structures Heterogeneous graphs: Multi-type nodes and edges (e.g., knowledge graphs) Large-scale graph learning: Neighbor sampling, distributed training Quick Start Installation For additional dependencies (sparse operations, clustering): Basic Graph Creation Loading a Benchmark Dataset Core Concepts Data Structure PyG represents graphs using the torchgeometric.data.Data class with these key attributes: data.x: Node feature matrix [numnodes, numnodefeatures] data.edgeindex: Graph connectivity in COO format [2, numedges] data.edgeattr: Edge feature matrix [numedges, numedgefeatures] (optional) data.y: Target labels for nodes or graphs data.pos: Node spatial positions [numnodes, numdimensions] (optional) Custom attributes: Can add any attribute (e.g., data.trainmask, data.batch) Important: These attributes are not mandatory—extend Data objects with custom attributes as needed. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills davila7/torch_geometric