gammagl.models.GINModel

class GINModel(in_channels, hidden_channels, out_channels, num_layers=4, name='GIN')[source]

The FiLM graph convolutional operator from the “GNN-FiLM: Graph Neural Networks with Feature-wise Linear Modulation” paper.

\[\mathbf{x}^{\prime}_i = \sum_{r \in \mathcal{R}} \sum_{j \in \mathcal{N}(i)} \sigma \left( \boldsymbol{\gamma}_{r,i} \odot \mathbf{W}_r \mathbf{x}_j + \boldsymbol{\beta}_{r,i} \right)\]
Parameters:

  • in_channels (int, tuple) – Size of each input sample, or -1 to derive the size from the first input(s) to the forward method. A tuple corresponds to the sizes of source and target dimensionalities.

  • hidden_channels (int) – Size of hidden layers of mlp.

  • out_channels (int) – Size of each output sample.

  • num_layers (int, optional) – Number of layers of GINConv. (default: 1)

  • name (str, optional) – name of the model.

forward(x, edge_index, batch)[source]