RelGraphConvΒΆ
-
class
dgl.nn.pytorch.conv.
RelGraphConv
(in_feat, out_feat, num_rels, regularizer=None, num_bases=None, bias=True, activation=None, self_loop=True, dropout=0.0, layer_norm=False)[source]ΒΆ Bases:
torch.nn.modules.module.Module
Relational graph convolution layer from Modeling Relational Data with Graph Convolutional Networks
It can be described in as below:
\[h_i^{(l+1)} = \sigma(\sum_{r\in\mathcal{R}} \sum_{j\in\mathcal{N}^r(i)}e_{j,i}W_r^{(l)}h_j^{(l)}+W_0^{(l)}h_i^{(l)})\]where \(\mathcal{N}^r(i)\) is the neighbor set of node \(i\) w.r.t. relation \(r\). \(e_{j,i}\) is the normalizer. \(\sigma\) is an activation function. \(W_0\) is the self-loop weight.
The basis regularization decomposes \(W_r\) by:
\[W_r^{(l)} = \sum_{b=1}^B a_{rb}^{(l)}V_b^{(l)}\]where \(B\) is the number of bases, \(V_b^{(l)}\) are linearly combined with coefficients \(a_{rb}^{(l)}\).
The block-diagonal-decomposition regularization decomposes \(W_r\) into \(B\) number of block diagonal matrices. We refer \(B\) as the number of bases.
The block regularization decomposes \(W_r\) by:
\[W_r^{(l)} = \oplus_{b=1}^B Q_{rb}^{(l)}\]where \(B\) is the number of bases, \(Q_{rb}^{(l)}\) are block bases with shape \(R^{(d^{(l+1)}/B)*(d^{l}/B)}\).
- Parameters
in_feat (int) β Input feature size; i.e, the number of dimensions of \(h_j^{(l)}\).
out_feat (int) β Output feature size; i.e., the number of dimensions of \(h_i^{(l+1)}\).
num_rels (int) β Number of relations.
regularizer (str, optional) β
Which weight regularizer to use (βbasisβ, βbddβ or
None
):βbasisβ is for basis-decomposition.
βbddβ is for block-diagonal-decomposition.
None
applies no regularization.
Default:
None
.num_bases (int, optional) β Number of bases. It comes into effect when a regularizer is applied. If
None
, it uses number of relations (num_rels
). Default:None
. Note thatin_feat
andout_feat
must be divisible bynum_bases
when applying βbddβ regularizer.bias (bool, optional) β True if bias is added. Default:
True
.activation (callable, optional) β Activation function. Default:
None
.self_loop (bool, optional) β True to include self loop message. Default:
True
.dropout (float, optional) β Dropout rate. Default:
0.0
layer_norm (bool, optional) β True to add layer norm. Default:
False
Examples
>>> import dgl >>> import numpy as np >>> import torch as th >>> from dgl.nn import RelGraphConv >>> >>> g = dgl.graph(([0,1,2,3,2,5], [1,2,3,4,0,3])) >>> feat = th.ones(6, 10) >>> conv = RelGraphConv(10, 2, 3, regularizer='basis', num_bases=2) >>> etype = th.tensor([0,1,2,0,1,2]) >>> res = conv(g, feat, etype) >>> res tensor([[ 0.3996, -2.3303], [-0.4323, -0.1440], [ 0.3996, -2.3303], [ 2.1046, -2.8654], [-0.4323, -0.1440], [-0.1309, -1.0000]], grad_fn=<AddBackward0>)
-
forward
(g, feat, etypes, norm=None, *, presorted=False)[source]ΒΆ Forward computation.
- Parameters
g (DGLGraph) β The graph.
feat (torch.Tensor) β A 2D tensor of node features. Shape: \((|V|, D_{in})\).
etypes (torch.Tensor or list[int]) β An 1D integer tensor of edge types. Shape: \((|E|,)\).
norm (torch.Tensor, optional) β An 1D tensor of edge norm value. Shape: \((|E|,)\).
presorted (bool, optional) β Whether the edges of the input graph have been sorted by their types. Forward on pre-sorted graph may be faster. Graphs created by
to_homogeneous()
automatically satisfy the condition. Also seereorder_graph()
for sorting edges manually.
- Returns
New node features. Shape: \((|V|, D_{out})\).
- Return type
torch.Tensor