GraphToGlobalMPM

class deeplay.models.gnn.gtogmpm.GraphToGlobalMPM(*args, **kwargs)

Bases: MPM

Graph-to-Global Message Passing Neural Network (MPN) model.

Parameters

hidden_features: list[int]

Number of hidden units in each Message Passing Layer.

out_features: int

Number of output features.

pool: template-like

Specification for the pooling of the model. Default: nn.Identity.

out_activation: template-like

Specification for the output activation of the model. Default: nn.Identity.

Configurables

• hidden_features (list[int]): Number of hidden units in each Message Passing Layer.

• out_features (int): Number of output features.

• pool (template-like): Specification for the pooling of the model. Default: nn.Identity.

• out_activation (template-like): Specification for the output activation of the model. Default: nn.Identity.

• encoder (template-like): Specification for the encoder of the model. Default: dl.Parallel consisting of two MLPs to process node and edge features.

• backbone (template-like): Specification for the backbone of the model. Default: dl.MessagePassingNeuralNetwork.

• selector (template-like): Specification for the selector of the model. Default: dl.FromDict(“x”) selecting the node features.

• head (template-like): Specification for the head of the model. Default: dl.MultiLayerPerceptron.

Constraints

• input: Dict[str, Any] or torch-geometric Data object containing the following attributes:

  • x: torch.Tensor of shape (num_nodes, node_in_features).

  • edge_index: torch.Tensor of shape (2, num_edges).

  • edge_attr: torch.Tensor of shape (num_edges, edge_in_features).

  • batch: torch.Tensor of shape (num_nodes,).

  NOTE: node_in_features and edge_in_features are inferred from the input data.

• output: torch.Tensor of shape (batch_size, out_features)

Examples

>>> model = GraphToGlobalMPM([64, 64], 1).create()
>>> inp = {}
>>> inp["x"] = torch.randn(10, 16)
>>> inp["edge_index"] = torch.randint(0, 10, (2, 20))
>>> inp["edge_attr"] = torch.randn(20, 8)
>>> inp["batch"] = torch.Tensor([0, 0, 0, 0, 1, 1, 1, 1, 1, 1]).long()
>>> model(inp).shape
torch.Size([2, 1])