gigl.types.LoadedGraphTensors

class gigl.types.data.LoadedGraphTensors(node_ids: Union[torch.Tensor, dict[NewType.<locals>.new_type, torch.Tensor]], node_features: Union[torch.Tensor, dict[NewType.<locals>.new_type, torch.Tensor], NoneType], edge_index: Union[torch.Tensor, dict[gigl.src.common.types.graph_data.EdgeType, torch.Tensor]], edge_features: Union[torch.Tensor, dict[gigl.src.common.types.graph_data.EdgeType, torch.Tensor], NoneType], positive_label: Union[torch.Tensor, dict[gigl.src.common.types.graph_data.EdgeType, torch.Tensor], NoneType], negative_label: Union[torch.Tensor, dict[gigl.src.common.types.graph_data.EdgeType, torch.Tensor], NoneType])

Bases: object

Methods

__init__
treat_labels_as_edges	Convert positive and negative labels to edges.

__eq__(other)

Return self==value.

__hash__ = None

__init__(node_ids: Tensor | dict[NodeType, Tensor], node_features: Tensor | dict[NodeType, Tensor] | None, edge_index: Tensor | dict[EdgeType, Tensor], edge_features: Tensor | dict[EdgeType, Tensor] | None, positive_label: Tensor | dict[EdgeType, Tensor] | None, negative_label: Tensor | dict[EdgeType, Tensor] | None) → None

__repr__()

Return repr(self).

__weakref__

list of weak references to the object (if defined)

treat_labels_as_edges() → None

Convert positive and negative labels to edges. Converts this object in-place to a “heterogeneous” representation.

This requires the following conditions and will throw if they are not met:

1. The node_ids, node_features, edge_index, and edge_features are not dictionaries (we loaded a homogeneous graph).

2. The positive_label and negative_label are not None and are Tensors, not dictionaries.