Qubit Operator converted from fermionic op and majoranic op does not match

[Gopal-Dahale]

Describe the issue

For small examples (eg. H2 with sto3g), they match but for the given example they don't. Is this supposed to be the case?

openfermionpyscf version is 0.5.

What version of this software are you using?

1.7.1

How can the issue be reproduced?

from openfermion import MolecularData, jordan_wigner
from openfermion.transforms import (
    get_fermion_operator, 
    get_majorana_operator,
)
from openfermionpyscf import run_pyscf

geometry = [
    ('H', (-0.1129, 0.3052, 0.0000)),
    ('Be',(1.0381, -0.3601, 0.0000)),
    ('H', (2.1898, -1.0243, 0.0000))
]
basis = '3-21g'
multiplicity = 1
charge = 0

# Make molecule and print out a few interesting facts about it.
molecule = MolecularData(geometry, basis, multiplicity, charge)
molecule = run_pyscf(molecule, run_scf=True)

molecular_hamiltonian = molecule.get_molecular_hamiltonian()
fermionic_hamiltonian = get_fermion_operator(molecular_hamiltonian)
majorana_hamiltonian = get_majorana_operator(fermionic_hamiltonian)

qubit_op = jordan_wigner(fermionic_hamiltonian)
qubit_op_majorana = jordan_wigner(majorana_hamiltonian)

qubit_op.compress()
qubit_op_majorana.compress()

print(qubit_op == qubit_op_majorana)
print("Number of qubit terms from fermionic hamiltonian:", len(qubit_op.terms))
print("Number of qubit terms from majorana hamiltonian:", len(qubit_op_majorana.terms))

Result:

False
Number of qubit terms from fermionic hamiltonian: 18916
Number of qubit terms from majorana hamiltonian: 19888

[ncrubin]

do the operators you get match in their spectrum? if there are a bunch of terms with really small coefficients close to whatever the compress threshold is then that could be it. but their spectrum should be equivalent.