dwave.embedding.chain_breaks.weighted_random

weighted_random(samples, chains)

Unembed samples using weighed random choice for broken chains.

Parameters:

• samples (samples_like) – A collection of samples. samples_like is an extension of NumPy’s array-like. See dimod.as_samples().

• chains (list[array-like]) – List of chains, where each chain is an array-like collection of the variables in the same order as their representation in the given samples.

Returns:

A 2-tuple containing:

numpy.ndarray: Unembedded samples as an \(nS \times nC\) array of dtype int8, where \(nC\) is the number of chains and \(nS\) the number of samples. Broken chains are resolved by setting the sample value to a random value weighted by frequency of the value in the chain.

numpy.ndarray: Indices of the samples. Equivalent to np.arange(nS) because all samples are kept and no samples are added.

Return type:

tuple

Examples

This example unembeds samples from a target graph that chains nodes 0 and 1 to represent one source node and nodes 2, 3, and 4 to represent another. The sample has broken chains for both source nodes.

>>> import numpy as np
>>> import dimod
...
>>> chains = [(0, 1), (2, 3, 4)]
>>> samples = np.array([[1, 0, 1, 0, 1]], dtype=np.int8)
>>> unembedded, idx = dwave.embedding.weighted_random(samples, chains)
>>> unembedded
array([[1, 1]], dtype=int8)
>>> idx
array([0, 1])