class GraphConvolution(*args: Any, **kwargs: Any)

Bases: Module

GCN graph convolution layer

forward(x: torch.Tensor) → torch.Tensor

Forward

Parameters:

x – Tensor

Returns:

Tensor

reset_parameters() → None

Reset model parameters

weight

self.weight = Parameter(torch.FloatTensor(in_features, out_features)) if bias:

self.bias = Parameter(torch.FloatTensor(out_features))

else:

self.register_parameter(‘bias’, None)

self.reset_parameters()