Skip to content

Latest commit

 

History

History
109 lines (96 loc) · 3.58 KB

softmax.md

File metadata and controls

109 lines (96 loc) · 3.58 KB

tfgnn.softmax

View source on GitHub

Computes softmax over a many-to-one relationship in a GraphTensor.

tfgnn.softmax(
    graph_tensor: tfgnn.GraphTensor,
    per_tag: tfgnn.IncidentNodeOrContextTag,
    *,
    edge_set_name: Union[Sequence[EdgeSetName], EdgeSetName, None] = None,
    node_set_name: Union[Sequence[NodeSetName], NodeSetName, None] = None,
    feature_value: Union[Sequence[Field], Field, None] = None,
    feature_name: Optional[FieldName] = None
) -> Union[Sequence[Field], Field]

This function can be used to compute a softmax normalization...

  • of edge values, across the edges with a common incident node at per_tag (e.g., SOURCE or TARGET);
  • of node values, across all the nodes in the same graph component;
  • of edge values, across all the edges in the same graph component.

For non-scalar values, the softmax function is applied element-wise.

Args

graph_tensor A scalar GraphTensor.
per_tag tfgnn.CONTEXT for normalization per graph component, or an incident node tag (e.g., tfgnn.SOURCE or tfgnn.TARGET) for normalization per common incident node.
edge_set_name The name of the edge set on which values are normalized, or a non-empty sequence of such names. Unless from_tag=tfgnn.CONTEXT, all named edge sets must have the same incident node set at the given tag.
node_set_name The name of the node set on which values are normalized, or a non-empty sequence of such names. Can only be passed together with from_tag=tfgnn.CONTEXT. Exactly one of edge_set_name or node_set_name must be set.
feature_value A tensor or list of tensors, parallel to the node_set_names or edge_set_names, to supply the input values of softmax. Each tensor has shape [num_items, *feature_shape], where num_items is the number of edges in the given edge set or nodes in the given node set, and *feature_shape is the same across all inputs.
feature_name The name of a feature stored on each graph piece from which pooling is done, for use instead of an explicity passed feature_value. Exactly one of feature_name or feature_value must be set.

Returns

A tensor or a list of tensors with the softmaxed values. The dimensions of the tensors and the length of the list do not change from the input.