networkx.algorithms.approximation.maxcut.randomized_partitioning¶

randomized_partitioning
(G, seed=None, p=0.5, weight=None)[source]¶ Compute a random partitioning of the graph nodes and its cut value.
A partitioning is calculated by observing each node and deciding to add it to the partition with probability
p
, returning a random cut and its corresponding value (the sum of weights of edges connecting different partitions). Parameters
 GNetworkX graph
 seedinteger, random_state, or None (default)
Indicator of random number generation state. See Randomness.
 pscalar
Probability for each node to be part of the first partition. Should be in [0,1]
 weightobject
Edge attribute key to use as weight. If not specified, edges have weight one.
 Returns
 cut_sizescalar
Value of the minimum cut.
 partitionpair of node sets
A partitioning of the nodes that defines a minimum cut.