Use this graph operation to transform, inspect, classify, or aggregate a graph whose directedness, edge semantics, weights, node identifiers, and duplicate-edge policy are defined.

Weighted candidates

Required object input.

Return parameter p

Required number input.

In-out parameter q

Required number input.

Connection threshold

Required number input.

Demonstrate weighted node2vec-style transitions with a threshold for strong previous-node connections. The tool evaluates the supplied inputs together and returns the named outputs below; it does not infer omitted operating conditions or change the units shown.^[1]

Weighted transitions

The resulting weighted transitions returned as an object.

• Graph results depend on representation and conventions for self-loops, parallel edges, direction, isolated nodes, weights, normalization, and traversal order. Learned embeddings additionally depend on sampling and training parameters.
• Use finite inputs in the displayed units and preserve more precision than the final presentation requires. Independently verify safety-critical, financial, compliance, or production decisions.

Validate node and edge counts before and after transformation, test small known graphs, and preserve an explicit graph schema with algorithm parameters.

• PinSage Aggregation Demonstrator

  Demonstrate PinSage-style importance pooling from random-walk visit counts and node features.

• Node Classification Demonstrator

  Combine a node's class scores with weighted neighbor scores and predict a class.

• Node2vec Transition Demonstrator

  Calculate second-order node2vec transition probabilities from return, neighborhood, and outward candidates.