TilingComposite

Class

Tiles many smaller problems across a larger Chimera-structured sampler.

class TilingComposite(sampler, sub_m, sub_n, t=4)

Composite to tile a small problem across a Chimera-structured sampler. A problem that can fit on a small Chimera graph can be replicated across a larger Chimera graph to get samples from multiple areas of the system in one call. For example, a 2x2 Chimera lattice could be tiled 64 times (8x8) on a fully-yielded D-Wave 2000Q system (16x16).

Parameters:

• sampler (dimod.Sampler) – A structured dimod sampler to be wrapped.
• sub_m (int) – The number of rows in the sub-Chimera lattice.
• sub_n (int) – The number of columns in the sub-Chimera lattice.
• t (int) – The size of the shore within each Chimera cell.

Sampler Properties

TilingComposite.properties	dict – Contains one key 'child_properties' which has a copy of the child sampler’s properties.
TilingComposite.parameters	dict[str, list] – The keys are the keyword parameters accepted by the child sampler.

Composite Properties

TilingComposite.children	list – Contains the single wrapped structured sampler.
TilingComposite.child	The first child in children.

Structure Properties

TilingComposite.nodelist	list – The nodes available to the sampler.
TilingComposite.edgelist	list – The edges available to the sampler.
TilingComposite.adjacency	dict[variable, set] – The adjacency structure.
TilingComposite.structure	A namedtuple Structure(nodelist, edgelist, adjacency)

Methods

TilingComposite.sample(bqm, **parameters)	Samples from the given bqm using the instantiated sample method.
TilingComposite.sample_ising(h, J, **kwargs)	Sample from the sub-Chimera lattice.
TilingComposite.sample_qubo(Q, **parameters)	Samples from the given QUBO using the instantiated sample method.