Operators for Message-Passing#
Simple Neighborhood Aggregator (SAGEConv)#
Computes the forward pass for simple aggregation (agg_simple) using node features in an |
|
Computes the backward pass for a simple aggregation (agg_simple) using node features in an |
|
Computes the forward pass for simple aggregation (agg_simple) using edge features in an |
|
Computes the backward pass for a simple aggregation (agg_simple) using edge features in an |
|
Computes the forward pass for both a simple aggregation (agg_simple) using node features in an |
|
Computes the backward pass for both a simple aggregation (agg_simple) using node features in an |
|
Computes the forward pass for simple aggregation using node features in an |
|
Computes the backward pass for a simple aggregation using node features in an |
|
Computes the forward pass for simple aggregation using edge features in an |
|
Computes the backward pass for a simple aggregation using edge features in an |
|
Computes the forward pass for both a simple aggregation using node features in an |
|
Computes the backward pass for both a simple aggregation (agg_simple) using node features in an |
Weighted Neighborhood Aggregation#
Computes the forward pass for a weighted aggregation (agg_weighted) using node features in an |
|
Computes the backward pass for a weighted aggregation (agg_weighted) using node features in an |
|
Computes the forward pass for a weighted aggregation (agg_weighted) using node features in an |
|
Computes the backward pass for a weighted aggregation (agg_weighted) using node features in an |
Heterogenous Aggregator using Basis Decomposition (RGCNConv)#
Computes the forward pass for node-to-node full-graph RGCN-like basis regularized |
|
Computes the backward pass for node-to-node full-graph RGCN-like basis regularized |
Graph Attention (GATConv/GATv2Conv)#
Computes the forward pass for a multi-head attention layer (GAT-like) |
|
Computes the forward pass for a multi-head attention layer (GAT-like) |
|
Computes the forward pass for a multi-head attention layer (GAT-like) |
|
Computes the backward pass for a multi-head attention layer (GAT-like) |
|
Computes the backward pass for a multi-head attention layer (GAT-like) |
|
Computes the backward pass for a multi-head attention layer (GAT-like) |
|
Computes the backward pass for a multi-head attention layer (GAT-like) |
|
Computes the backward pass for a multi-head attention layer (GAT-like) |
|
Computes the backward pass for a multi-head attention layer (GAT-like) |
|
Computes the backward pass for a multi-head attention layer (GAT-like) |
|
Computes the forward pass for a multi-head attention layer (GAT-like) |
|
Computes the forward pass for a multi-head attention layer (GAT-like) |
|
Computes the forward pass for a multi-head attention layer (GAT-like) |
|
Computes the backward pass for a multi-head attention layer (GAT-like) |
|
Computes the backward pass for a multi-head attention layer (GAT-like) |
|
Computes the backward pass for a multi-head attention layer (GAT-like) |
|
Computes the backward pass for a multi-head attention layer (GAT-like) |
|
Computes the backward pass for a multi-head attention layer (GAT-like) |
|
Computes the backward pass for a multi-head attention layer (GAT-like) |
|
Computes the backward pass for a multi-head attention layer (GAT-like) |
|
Computes the forward pass for a multi-head attention layer (GAT-like) |
|
Computes the forward pass for a multi-head attention layer (GAT-like) |
|
Computes the forward pass for a multi-head attention layer (GAT-like) |
|
Computes the backward pass for a multi-head attention layer (GAT-like) |
Transformer-like Graph Attention (TransformerConv)#
Computes the forward pass for a multi-head attention layer (GAT-like) |
|
Computes the forward pass for a multi-head attention layer (GAT-like) |
|
Computes the forward pass for a multi-head attention layer (GAT-like) |
|
Computes the backward pass for a multi-head attention layer (GAT-like) |
Directional Message-Passing (DMPNN)#
Computes the forward pass for D-MPNN-like full graph aggregation |
|
Computes the backward pass for D-MPNN-like full graph aggregation |
Update Edges: Concatenation or Sum of Edge and Node Features#
Computes the forward pass for an operation which concatenates |
|
Computes the backward pass for an operation which concatenates |
|
Computes the forward pass for an operation which sums |
|
Computes the backward pass for an operation which sums |
|
Computes the forward pass for an operation which concatenates |
|
Computes the backward pass for an operation which concatenates |
|
Computes the forward pass for an operation which sums |
|
Computes the backward pass for an operation which sums |