Qiskit code throws when z2symmetry_reduction argument in QubitConverter is set to 'auto'

Question

Following Qiskit code throws: AttributeError: 'ListOp' object has no attribute 'to_circuit'. This seems to be related to: z2symmetry_reduction='auto'. If the z2symmetry_reduction argument in QubitConverter is set to 'None', the code runs fine.

from qiskit import transpile
from qiskit.transpiler import PassManager, StagedPassManager
from qiskit.providers.fake_provider import FakeWashington

from qiskit_nature.converters.second_quantization import QubitConverter
# from qiskit_nature.second_q.mappers import QubitConverter (same result)
from qiskit_nature.second_q.circuit.library.ansatzes import UCC, UCCSD
from qiskit_nature.second_q.drivers import PySCFDriver
from qiskit_nature.second_q.formats.molecule_info import MoleculeInfo
from qiskit_nature.second_q.mappers import ParityMapper
from qiskit_nature.second_q.transformers import FreezeCoreTransformer

qubit_converter = QubitConverter(ParityMapper(), two_qubit_reduction=True, z2symmetry_reduction='auto')

molecule = MoleculeInfo(["Li", "H"], [(0.0, 0.0, 0.0), (0.0, 0.0, 1.595)])
driver = PySCFDriver.from_molecule(molecule, basis="sto3g")

transformer = FreezeCoreTransformer()
electronic_structure_problem = transformer.transform(driver.run())

second_quantized_hamiltonian = electronic_structure_problem.second_q_ops()
pauli_sum_operator = qubit_converter.convert(second_quantized_hamiltonian[0], num_particles=electronic_structure_problem.num_particles)

num_particles = electronic_structure_problem.num_particles
num_spatial_orbitals = electronic_structure_problem.num_spatial_orbitals
num_spin_orbitals = electronic_structure_problem.num_spin_orbitals

ucc_ansatz = UCC(num_spatial_orbitals, num_particles, excitations='sd', qubit_converter=qubit_converter, 
                 alpha_spin=True, beta_spin=True, max_spin_excitation=1, generalized=True, preserve_spin=True, reps=5)

decomposed_circuit = ucc_ansatz.decompose().decompose().decompose()
print("depth: ", decomposed_circuit.depth())

transpiled_circuit = transpile(ucc_ansatz, FakeWashington(), optimization_level=3)
print("depth transpiled: ", transpiled_circuit.depth())
print(dict(transpiled_circuit.count_ops()))

print("ansatz parameters: ", len(ucc_ansatz.parameters.data))

My software version:

Answer 1

The auto symmetry detection relies on the problem defining the sector. When using a GroundStateSolver this conversion is done for you taking that into account. Doing things more manually, as per your example, update this line

pauli_sum_operator = qubit_converter.convert(second_quantized_hamiltonian[0], num_particles=electronic_structure_problem.num_particles)

to pass in the sector locator method from the problem. (Use the second_q converter, not the other older deprecated one, otherwise you may hit other errors, like the ListOp error you show)

pauli_sum_operator = qubit_converter.convert(second_quantized_hamiltonian[0], num_particles=electronic_structure_problem.num_particles,
sector_locator=electronic_structure_problem.symmetry_sector_locator)

So it can automatically locate the sector as it needs to get that via the problem.

I will note that in the upcoming 0.6.0 release of Nature, that should be released in the near future, the QubitConverter is deprecated, to be removed in the future, in favor of having a TaperedQubitmapper so mappers are used directly going forwards without the 'wrapping' by the converter. The old way with the converter will still be supported, but deprecated. Check the migration guide in the docs after the release for more information.