cugraph-pyg API Reference#
cugraph-pyg
Graph Storage#
|
Duck-typed version of PyG's GraphStore and FeatureStore that uses Dask to distribute the graph structure across GPUs and a cugraph.gnn.FeatureStore to store node/edge features. |
|
cuGraph-backed PyG GraphStore implementation that distributes the graph across workers. |
Feature Storage#
|
A basic implementation of the PyG FeatureStore interface that stores feature data in a single TensorDict. |
|
A basic implementation of the PyG FeatureStore interface that stores feature data in WholeGraph WholeMemory. |
Data Loaders#
|
Duck-typed version of the PyG NeighborLoader interface that uses Dask to sample nodes using the uniform neighbor sampling algorithm. |
|
Iterator that executes sampling using Dask and cuGraph and loads sampled minibatches from disk. |
|
Duck-typed version of torch_geometric.loader.NodeLoader. |
|
Duck-typed version of torch_geometric.loader.NeighborLoader |
Samplers#
|
Methods |
|
Iterator that processes results from the cuGraph distributed sampler. |
|
Subclass of SampleReader that reads homogeneous output samples produced by the cuGraph distributed sampler. |
|
Iterator that combines output graphs with their features to produce final output minibatches that can be fed into a GNN model. |