API Reference

• dgl
  • Graph Create Ops
  • Subgraph Extraction Ops
  • Graph Transform Ops
  • Graph Partition Utilities
  • Batching and Reading Out Ops
  • Adjacency Related Utilities
  • Graph Traversal & Message Propagation
  • Utilities
• dgl.data
  • Base Dataset Class
  • Node Prediction Datasets
  • Edge Prediction Datasets
  • Graph Prediction Datasets
  • Utilities
• dgl.dataloading
  • DataLoaders
  • Neighbor Sampler
  • Collators
  • Negative Samplers for Link Prediction
  • Async Copying to/from GPUs
• dgl.DGLGraph
  • Querying metagraph structure
  • Querying graph structure
  • Querying and manipulating sparse format
  • Querying and manipulating node/edge ID type
  • Using Node/edge features
  • Transforming graph
  • Adjacency and incidence matrix
  • Computing with DGLGraph
  • Querying and manipulating batch information
  • Mutating topology
  • Device Control
  • Misc
• dgl.distributed
  • Initialization
  • Distributed Graph
  • Distributed Tensor
  • Distributed Node Embedding
  • Distributed embedding optimizer
  • Distributed workload split
  • Distributed Sampling
  • Partition
• dgl.function
  • DGL Built-in Function
  • Message functions
  • Reduce functions
• dgl.nn
  • NN Modules (PyTorch)
  • NN Modules (MXNet)
  • NN Modules (Tensorflow)
• dgl.ops
  • GSpMM functions
  • GSDDMM functions
  • Segment Reduce Module
  • Supported Data types
  • Relation with Message Passing APIs
• dgl.sampling
  • Random walk
  • Neighbor sampling
• User-defined Functions
  • Edge-wise User-defined Function
  • Node-wise User-defined Function
  • Degree Bucketing for Message Passing with User Defined Functions