gammagl.models.MixHopModel

class MixHopModel(feature_dim, hidden_dim, out_dim, p, drop_rate, num_layers=3, norm='both', name=None)[source]

MixHop proposed in “MixHop: Higher-Order Graph Convolutional Architectures via Sparsified Neighborhood Mixing” paper.

Parameters:

  • feature_dim (int) – input feature dimension.

  • hidden_dim (int) – hidden dimension.

  • out_dim (int) – The number of classes for prediction.

  • p (list) – The list of integer adjacency powers.

  • drop_rate (float) – dropout rate.

  • num_layers (int, optional) – Number of Mixhop Graph Convolutional Layers.

  • norm (str, optional) – apply the normalizer.

  • name (str, optional) – model name.

forward(x, edge_index, edge_weight, num_nodes=None)[source]