dwave-hybrid: A Python repository from JoelPasvolsky

D-Wave Hybrid

A general, minimal Python framework for building hybrid asynchronous decomposition samplers for quadratic unconstrained binary optimization (QUBO) problems.

dwave-hybrid facilitates three aspects of solution development:

Hybrid approaches to combining quantum and classical compute resources
Evaluating a portfolio of algorithmic components and problem-decomposition strategies
Experimenting with workflow structures and parameters to obtain the best application results

The framework enables rapid development and insight into expected performance of productized versions of its experimental prototypes.

Your optimized algorithmic components and other contributions to this project are welcome!

Installation or Building

Install from a package on PyPI:

pip install dwave-hybrid

or from source:

git clone https://github.com/dwavesystems/dwave-hybrid.git
cd dwave-hybrid
pip install -r requirements.txt
python setup.py install

Testing

Install test requirements and run unittest:

pip install -r tests/requirements.txt
python -m unittest

Example

import dimod
import hybrid

# Construct a problem
bqm = dimod.BinaryQuadraticModel({}, {'ab': 1, 'bc': -1, 'ca': 1}, 0, dimod.SPIN)

# Define the workflow
iteration = hybrid.RacingBranches(
    hybrid.InterruptableTabuSampler(),
    hybrid.EnergyImpactDecomposer(size=2)
    | hybrid.QPUSubproblemAutoEmbeddingSampler()
    | hybrid.SplatComposer()
) | hybrid.ArgMin()
workflow = hybrid.LoopUntilNoImprovement(iteration, convergence=3)

# Solve the problem
init_state = hybrid.State.from_problem(bqm)
final_state = workflow.run(init_state).result()

# Print results
print("Solution: sample={.samples.first}".format(final_state))

License

Released under the Apache License 2.0. See LICENSE file.

Contributing