BlockDecomposition

class BlockDecomposition(input_dim, num_relations, num_blocks=None, output_dim=None)[source]

Represent relation-specific weight matrices via block-diagonal matrices.

The block-diagonal decomposition restricts each transformation matrix to a block-diagonal-matrix, i.e.,

\[\mathbf{W}_r^l = diag(\mathbf{B}_{r,1}^l, \ldots, \mathbf{B}_{r,B}^l)\]

where \(\mathbf{B}_{r,i} \in \mathbb{R}^{(d^{(l) }/ B) \times (d^{(l)} / B)}\).

Initialize the layer.

Parameters

input_dim (int) – >0 The input dimension.
num_relations (int) – >0 The number of relations.
num_blocks (Optional[int]) – >0 The number of blocks to use. Has to be a divisor of input_dim.
output_dim (Optional[int]) – >0 The output dimension. If None is given, defaults to input_dim.

Methods Summary

`forward`(x, source, target, edge_type[, ...])	Relation-specific message passing from source to target.
`reset_parameters`()	Reset the parameters of this layer.

Methods Documentation

forward(x, source, target, edge_type, edge_weights=None, accumulator=None)[source]

Relation-specific message passing from source to target.

Parameters

x (FloatTensor) – shape: (num_nodes, input_dim) The node representations.
source (LongTensor) – shape: (num_edges,) The source indices.
target (LongTensor) – shape: (num_edges,) The target indices.
edge_type (LongTensor) – shape: (num_edges,) The edge types.
edge_weights (Optional[FloatTensor]) – shape: (num_edges,) Precomputed edge weights.
accumulator (Optional[FloatTensor]) – shape: (num_nodes, output_dim) a pre-allocated output accumulator. may be used if multiple different message passing steps are performed and accumulated by sum. If none is given, create an accumulator filled with zeroes.

Return type

FloatTensor

Returns

shape: (num_nodes, output_dim) The enriched node embeddings.