def test_all_energies(self):
sampler = PolyScaleComposite(HigherOrderComposite(ExactSolver()))
h = {'a': -1, 'b': 4}
J = {'abc': -1, 'ab': 1, 'aaa': .5}
sampleset = sampler.sample_hising(h, J, discard_unsatisfied=True)
for sample, energy in sampleset.data(['sample', 'energy']):
en = 0
for v, bias in h.items():
en += sample[v] * bias
for term, bias in J.items():
val = bias
for v in term:
val *= sample[v]
en += val
self.assertAlmostEqual(energy, en)
def test_empty(self):
sampler = PolyScaleComposite(RangeLimitedSampler())
samples = sampler.sample_hubo({})
self.assertEqual(len(samples.variables), 0)
sampler = PolyScaleComposite(RangeLimitedSampler())
samples = sampler.sample_hubo({'a': 4}, scalar=.25)
self.assertEqual(samples.first.energy, 4)
def test_fail_scale(self):
sampler = PolyScaleComposite(RangeLimitedSampler())
with self.assertRaises(RuntimeError):
sampler.sample_hising({'a': 4}, {}, scalar=1)
def test_scale(self):
sampler = PolyScaleComposite(RangeLimitedSampler())
samples = sampler.sample_hising({'a': 4}, {}, scalar=.25)
self.assertEqual(samples.first.energy, 4)
def test_normalizing(self):
sampler = PolyScaleComposite(RangeLimitedSampler())
samples = sampler.sample_hising({'a': 4}, {})
self.assertEqual(samples.first.energy, 4)
def test_empty(self):
sampler = PolyScaleComposite(RangeLimitedSampler())
samples = sampler.sample_hising({}, {})
self.assertEqual(len(samples.variables), 0)
def test_all_zero(self):
sampler = PolyScaleComposite(RangeLimitedSampler())
sampler.sample_hising({'a': 0}, {'ab': 0, 'bc': 0, 'abc': 0})
def test_wrap_bqm(self):
with self.assertRaises(TypeError):
PolyScaleComposite(ExactSolver())
def test_typical(self):
sampler = PolyScaleComposite(HigherOrderComposite(ExactSolver()))
self.assertTrue(hasattr(sampler, 'sample_poly'))
self.assertTrue(hasattr(sampler, 'sample_hising'))
self.assertTrue(hasattr(sampler, 'sample_hubo'))