dgl.ops.gsddmm๏
- dgl.ops.gsddmm(g, op, lhs_data, rhs_data, lhs_target='u', rhs_target='v')[source]๏
Generalized Sampled-Dense-Dense Matrix Multiplication interface. It computes edge features by
op
lhs features and rhs features.where
is the returned feature on edges and , refers tou
,v
respectively. is the binary operatorop
, and is the graph we apply gsddmm on:g
. and are one of โs.- Parameters:
g (DGLGraph) โ The input graph.
op (str) โ Binary operator, could be
add
,sub
,mul
,div
,dot
,copy_lhs
,copy_rhs
.lhs_data (tensor or None) โ The left operand, could be None if itโs not required by op.
rhs_data (tensor or None) โ The right operand, could be None if itโs not required by op.
lhs_target (str) โ Choice of ``u``(source), ``e``(edge) or ``v``(destination) for left operand.
rhs_target (str) โ Choice of ``u``(source), ``e``(edge) or ``v``(destination) for right operand.
- Returns:
The result tensor.
- Return type:
tensor