def test_swap_network_trotter_ansatz_value_not_implemented():
ansatz = SwapNetworkTrotterAnsatz(test_hamiltonian)
study = HamiltonianVariationalStudy('study', ansatz, test_hamiltonian)
trial_result = cirq.TrialResult(params=ansatz.param_resolver(
ansatz.default_initial_params()),
measurements={},
repetitions=1)
with pytest.raises(NotImplementedError):
_ = study.value(trial_result)
def test_hamiltonian_variational_study_optimize():
ansatz = SwapNetworkTrotterAnsatz(test_hamiltonian)
study = HamiltonianVariationalStudy('study', ansatz, test_fermion_op)
study.optimize(OptimizationParams(ExampleAlgorithm(),
cost_of_evaluate=1.0),
'run',
reevaluate_final_params=True)
result = study.results['run']
assert all(result.data_frame['optimal_parameters'].apply(study.evaluate) ==
result.data_frame['optimal_value'])
assert result.params.cost_of_evaluate == 1.0
def test_hamiltonian_variational_study_noise_bounds():
ansatz = SwapNetworkTrotterAnsatz(test_hamiltonian)
study = HamiltonianVariationalStudy('study', ansatz, test_hamiltonian)
numpy.random.seed(38017)
a, b = study.noise_bounds(1e4)
c, d = study.noise_bounds(1e2)
numpy.testing.assert_allclose(10 * a, c)
numpy.testing.assert_allclose(10 * b, d)
a, b = study.noise_bounds(1e4, confidence=0.95)
c, d = study.noise_bounds(1e2, confidence=0.95)
numpy.testing.assert_allclose(10 * a, c)
numpy.testing.assert_allclose(10 * b, d)
numpy.testing.assert_allclose(study.noise_bounds(1e2),
study.noise_bounds(1e2, 0.99))
with pytest.raises(ValueError):
_ = study.noise_bounds(1.0, 1.0)
with pytest.raises(ValueError):
_ = study.noise_bounds(1.0, -1.0)
def test_trotter_ansatzes_evaluate_order_2(ansatz_factory, trotter_algorithm,
hamiltonian, atol):
"""Check that a Trotter ansatz with two iterations and default parameters
is consistent with time evolution with two Trotter steps."""
ansatz = ansatz_factory(hamiltonian, iterations=2)
qubits = ansatz.qubits
preparation_circuit = cirq.Circuit.from_ops(
prepare_gaussian_state(qubits,
openfermion.QuadraticHamiltonian(
hamiltonian.one_body),
occupied_orbitals=range(len(qubits) // 2)))
study = HamiltonianVariationalStudy(
'study', ansatz, hamiltonian, preparation_circuit=preparation_circuit)
simulator = cirq.google.XmonSimulator()
# Compute value using ansatz
val = study.evaluate(study.default_initial_params())
# Compute value by simulating time evolution
quarter_way_hamiltonian = openfermion.DiagonalCoulombHamiltonian(
one_body=hamiltonian.one_body, two_body=0.25 * hamiltonian.two_body)
three_quarters_way_hamiltonian = openfermion.DiagonalCoulombHamiltonian(
one_body=hamiltonian.one_body, two_body=0.75 * hamiltonian.two_body)
simulation_circuit = cirq.Circuit.from_ops(
simulate_trotter(qubits,
quarter_way_hamiltonian,
time=0.5 * ansatz.adiabatic_evolution_time,
n_steps=1,
order=1,
algorithm=trotter_algorithm),
simulate_trotter(qubits,
three_quarters_way_hamiltonian,
time=0.5 * ansatz.adiabatic_evolution_time,
n_steps=1,
order=1,
algorithm=trotter_algorithm))
circuit = preparation_circuit + simulation_circuit
result = simulator.simulate(circuit)
final_state = result.final_state
correct_val = openfermion.expectation(study._hamiltonian_linear_op,
final_state).real
numpy.testing.assert_allclose(val, correct_val, atol=atol)
def test_hamiltonian_variational_study_save_load():
datadir = 'tmp_ffETr2rB49RGP8WE8jer'
study_name = 'test_hamiltonian_study'
ansatz = SwapNetworkTrotterAnsatz(test_hamiltonian)
study = HamiltonianVariationalStudy(study_name,
ansatz,
test_fermion_op,
datadir=datadir)
study.optimize(
OptimizationParams(ExampleAlgorithm(), cost_of_evaluate=1.0),
'example')
study.save()
loaded_study = HamiltonianVariationalStudy.load(study_name,
datadir=datadir)
assert loaded_study.name == study.name
assert str(loaded_study.circuit) == str(study.circuit)
assert loaded_study.datadir == datadir
assert loaded_study.hamiltonian == test_fermion_op
assert len(loaded_study.results) == 1
result = loaded_study.results['example']
assert isinstance(result, OptimizationTrialResult)
assert result.repetitions == 1
assert result.params.cost_of_evaluate == 1.0
# Clean up
os.remove('{}/{}.study'.format(datadir, study_name))
os.rmdir(datadir)
def test_hamiltonian_variational_study_noise():
ansatz = SwapNetworkTrotterAnsatz(test_hamiltonian)
study = HamiltonianVariationalStudy('study', ansatz, test_hamiltonian)
numpy.random.seed(10821)
assert (abs(study.noise()) < abs(study.noise(1e6)) < abs(study.noise(1e5))
< abs(study.noise(1e4)) < abs(study.noise(1e3)))
def test_hamiltonian_variational_study_init_qubit_operator():
ansatz = SwapNetworkTrotterAnsatz(test_hamiltonian)
study = HamiltonianVariationalStudy('study', ansatz,
openfermion.QubitOperator((0, 'X')))
assert study.hamiltonian == openfermion.QubitOperator((0, 'X'))