ChebConv

class dgl.nn.pytorch.conv.ChebConv(in_feats, out_feats, k, activation=<function relu>, bias=True)[source]

Bases: Module

来自图上的快速局部谱滤波卷积神经网络的Chebyshev谱图卷积层

\[ \begin{align}\begin{aligned}h_i^{l+1} &= \sum_{k=0}^{K-1} W^{k, l}z_i^{k, l}\\Z^{0, l} &= H^{l}\\Z^{1, l} &= \tilde{L} \cdot H^{l}\\Z^{k, l} &= 2 \cdot \tilde{L} \cdot Z^{k-1, l} - Z^{k-2, l}\\\tilde{L} &= 2\left(I - \tilde{D}^{-1/2} \tilde{A} \tilde{D}^{-1/2}\right)/\lambda_{max} - I\end{aligned}\end{align} \]

其中 \(\tilde{A}\)\(A\) + \(I\)\(W\) 是可学习的权重。

Parameters:
  • in_feats (int) – 输入特征的维度;即,\(h_i^{(l)}\)的维度数。

  • out_feats (int) – 输出特征的维度 \(h_i^{(l+1)}\)

  • k (int) – 切比雪夫滤波器大小 \(K\)

  • activation (function, optional) – 激活函数。默认 ReLu

  • bias (bool, optional) – If True, adds a learnable bias to the output. Default: True.

示例

>>> import dgl
>>> import numpy as np
>>> import torch as th
>>> from dgl.nn import ChebConv
>>
>>> g = dgl.graph(([0,1,2,3,2,5], [1,2,3,4,0,3]))
>>> feat = th.ones(6, 10)
>>> conv = ChebConv(10, 2, 2)
>>> res = conv(g, feat)
>>> res
tensor([[ 0.6163, -0.1809],
        [ 0.6163, -0.1809],
        [ 0.6163, -0.1809],
        [ 0.9698, -1.5053],
        [ 0.3664,  0.7556],
        [-0.2370,  3.0164]], grad_fn=<AddBackward0>)
forward(graph, feat, lambda_max=None)[source]

计算ChebNet层。

Parameters:
  • graph (DGLGraph) – The graph.

  • feat (torch.Tensor) – 输入特征的形状为 \((N, D_{in})\),其中 \(D_{in}\) 是输入特征的大小,\(N\) 是节点的数量。

  • lambda_max (listtensorNone, 可选.) –

    一个长度为 \(B\) 的列表(tensor),存储每个单独图的归一化拉普拉斯矩阵的最大特征值, 其中 \(B\) 是输入图的批量大小。默认值:None。

    如果为 None,此方法将默认值设置为 2。 可以使用 dgl.laplacian_lambda_max() 来计算此值。

Returns:

The output feature of shape \((N, D_{out})\) where \(D_{out}\) is size of output feature.

Return type:

torch.Tensor