Chemqulacs#
Chemqulacs includes quantum chemistry code running on quantum circuit simulators and quantum computers.
Documentation#
Installation#
Chemqulacs requires Python 3.9.8 or later.
pip install git+https://github.com/wmizukami/chemqulacs.git
Usage#
VQECASCI, VQECASSCF and VQECI#
You can perform CASCI and CASSCF with VQE by VQECASCI and VQECASSCF classes, which are implemented using PySCF.
from pyscf import gto, scf
from chemqulacs.util import utils
from chemqulacs.vqe.vqemcscf import VQECASCI, VQECASSCF
# Retrieve geometry of a water molecule from PubChem
geom_water = utils.get_geometry_from_pubchem("water")
# Create PySCF's Mole object
mol = gto.M(atom=geom_water, basis="sto-3g")
# Run SCF calculation (RHF)
mf = scf.RHF(mol)
mf.run()
# Run CASCI with VQE with CAS(2e, 2o)
vqe_casci = VQECASCI(mf, 2, 2)
vqe_casci.kernel()
print(f"VQE-CASCI Energy: {vqe_casci.e_tot}")
# Run CASSCF with VQE with CAS(2e, 2o)
vqe_casscf = VQECASSCF(mf, 2, 2)
vqe_casscf.kernel()
print(f"VQE-CASSCF Energy: {vqe_casscf.e_tot}")
The default options for VQE are as follows:
Fermion-qubit mapping: Jordan-Wigner
Optimizer: Adam
Quantum backend: Qulacs
Ansatz: ParticleConservingU1 with 2 layers
You can change those options via arguments for VQECASCI and VQECASSCF objects, which are then passed to VQECI objects. Please see chemqulacs.vqe.vqeci module for more details.
You can also change number of excited states to calculate. SSVQE is used for excited-state calculations by the VQECASCI and VQECASSCF classes.
from pyscf import gto, scf
from chemqulacs.util import utils
from chemqulacs.vqe.vqemcscf import VQECASCI, VQECASSCF
from chemqulacs.vqe.vqeci import Ansatz
from quri_parts.algo.optimizer import LBFGS
# Retrieve geometry of a water molecule from PubChem
geom_water = utils.get_geometry_from_pubchem("water")
# Create PySCF's Mole object
mol = gto.M(atom=geom_water, basis="sto-3g")
# Run SCF calculation (RHF)
mf = scf.RHF(mol)
mf.run()
# Run CASCI with VQE with CAS(2e, 2o)
vqe_casci = VQECASCI(mf, 2, 2, optimizer=LBFGS(), ansatz=Ansatz.GateFabric, layers=2, excitation_number=5)
vqe_casci.kernel()
print(f"VQE-CASCI Energis: {vqe_casci.fcisolver.energies}")
vqe_casci.print_energies()
# Calculated electronic
# state Energy(Hartree) Excitation energy(eV) Occupancy indicates\n
# S0 -74.964448 0.0000
# S1 -74.567104 10.8123 1 -> 2
# S2 -74.528864 11.8528 0 -> 2
# S3 -74.528864 11.8528 1 -> 3
# Run CASSCF with VQE with CAS(2e, 2o)
vqe_casscf = VQECASSCF(mf, 2, 2, optimizer=LBFGS(), ansatz=Ansatz.GateFabric, layers=2, excitation_number=5)
vqe_casscf.kernel()
print(f"VQE-CASSCF Energis: {vqe_casci.fcisolver.energies}")
vqe_casscf.print_energies()
QSE#
You can run QSE by with VQE as follows.
from pyscf import gto, scf
from chemqulacs.util import utils
from chemqulacs.qse.qse import QSE
from chemqulacs.vqe.vqemcscf import VQECASCI, VQECASSCF
# Retrieve geometry of a water molecule from PubChem
geom_water = utils.get_geometry_from_pubchem("water")
# Create PySCF's Mole object
mol = gto.M(atom=geom_water, basis="sto-3g")
# Run SCF calculation (RHF)
mf = scf.RHF(mol)
mf.run()
# Run CASCI with VQE with CAS(2e, 2o)
vqe_casci = VQECASCI(mf, 2, 2)
vqe_casci.kernel()
print(f"VQE-CASCI Energy: {vqe_casci.e_tot}")
# Run QSE with VQE
qse = QSE(vqe_casci.fcisolver)
qse.gen_excitation_operators("ee", 1)
qse.solve()
Development#
Poetry is used for dependency management.
git clone https://github.com/wmizukami/chemqulacs
cd chemqulacs
# If quri_parts 0.14 should be used:
cp pyproject.14.toml pyproject.toml
cp poetry.14.lock poetry.lock
poetry install
poetry run pytest
See Makefile for other commands for development.