pygmtools.utils.build_aff_mat_from_networkx

pygmtools.utils.build_aff_mat_from_networkx(G1: Graph, G2: Graph, node_aff_fn=None, edge_aff_fn=None, backend=None)[source]

Convert networkx object to affinity matrix

Parameters

  • G1 – networkx object, whose type must be networkx.Graph

  • G2 – networkx object, whose type must be networkx.Graph

  • node_aff_fn – (default: inner_prod_aff_fn) the node affinity function with the characteristic node_aff_fn(2D Tensor, 2D Tensor) -> 2D Tensor, which accepts two node feature tensors and outputs the node-wise affinity tensor. See inner_prod_aff_fn() as an example.

  • edge_aff_fn – (default: inner_prod_aff_fn) the edge affinity function with the characteristic edge_aff_fn(2D Tensor, 2D Tensor) -> 2D Tensor, which accepts two edge feature tensors and outputs the edge-wise affinity tensor. See inner_prod_aff_fn() as an example.

  • backend – (default: pygmtools.BACKEND variable) the backend for computation.

Returns

the affinity matrix corresponding to the networkx object G1 and G2

Example
>>> import networkx as nx
>>> import pygmtools as pygm
>>> pygm.set_backend('numpy')

# Generate networkx images
>>> G1 = nx.DiGraph()
>>> G1.add_weighted_edges_from([(1, 2, 0.5), (2, 3, 0.8), (3, 4, 0.7)])
>>> G2 = nx.DiGraph()
>>> G2.add_weighted_edges_from([(1, 2, 0.3), (2, 3, 0.6), (3, 4, 0.9), (4, 5, 0.4)])

# Obtain Affinity Matrix
>>> K = pygm.utils.build_aff_mat_from_networkx(G1, G2)
>>> K.shape
(20, 20)

# The affinity matrices K can be further processed by GM solvers