torch-geometric - Build and train graph neural networks with PyG
Develop and train graph neural networks for node and graph classification, link prediction, molecular properties, and geometric data.
Tags
Updated: 2026-10-08Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Build graph neural networks
- Classify nodes and graphs
- Predict graph links
- Process heterogeneous graphs
- Batch irregular graph data
- Implement message passing layers
- Load benchmark datasets
- Create custom datasets
- Train on large-scale graphs
Inputs
- Graph data
- Node features
- Edge indices
- Labels
- Dataset files
- Learning task specifications
Outputs
- Model predictions
- Trained model parameters
- Processed graph datasets
- Batched graph objects
Requirements
- Python environment
- PyTorch installation
- torch-geometric installation