# This code is part of a Qiskit project.
#
# (C) Copyright IBM 2020, 2023.
#
# This code is licensed under the Apache License, Version 2.0. You may
# obtain a copy of this license in the LICENSE.txt file in the root directory
# of this source tree or at http://www.apache.org/licenses/LICENSE-2.0.
#
# Any modifications or derivative works of this code must retain this
# copyright notice, and modified files need to carry a notice indicating
# that they have been altered from the originals.
"""
This module implements the abstract base class for electronic structure driver modules.
"""
from __future__ import annotations
from abc import abstractmethod
from dataclasses import dataclass
from enum import Enum
from typing import Any, Sequence
from pathlib import Path
import numpy as np
import h5py
from qiskit_nature.second_q.formats.qcschema import (
QCModel,
QCProperties,
QCProvenance,
QCSchema,
QCTopology,
QCWavefunction,
)
from qiskit_nature.second_q.operators import Tensor
from qiskit_nature.second_q.problems import ElectronicBasis, ElectronicStructureProblem
from .base_driver import BaseDriver
@dataclass
class _QCSchemaData:
"""An internal data container to simplify the construction of QCSchema objects."""
hij: np.ndarray | None = None
hij_b: np.ndarray | None = None
eri: np.ndarray | None = None
hij_mo: np.ndarray | None = None
hij_mo_b: np.ndarray | None = None
eri_mo: np.ndarray | None = None
eri_mo_ba: np.ndarray | None = None
eri_mo_bb: np.ndarray | None = None
e_nuc: float | None = None
e_ref: float | None = None
overlap: np.ndarray | None = None
mo_coeff: np.ndarray | None = None
mo_coeff_b: np.ndarray | None = None
mo_energy: np.ndarray | None = None
mo_energy_b: np.ndarray | None = None
mo_occ: np.ndarray | None = None
mo_occ_b: np.ndarray | None = None
dip_x: np.ndarray | None = None
dip_y: np.ndarray | None = None
dip_z: np.ndarray | None = None
dip_mo_x_a: np.ndarray | None = None
dip_mo_y_a: np.ndarray | None = None
dip_mo_z_a: np.ndarray | None = None
dip_mo_x_b: np.ndarray | None = None
dip_mo_y_b: np.ndarray | None = None
dip_mo_z_b: np.ndarray | None = None
dip_nuc: tuple[float, float, float] | None = None
dip_ref: tuple[float, float, float] | None = None
symbols: Sequence[str] | None = None
coords: Sequence[float] | None = None
multiplicity: int | None = None
charge: int | None = None
masses: Sequence[float] | None = None
method: str | None = None
basis: str | None = None
creator: str | None = None
version: str | None = None
routine: str | None = None
nbasis: int | None = None
nmo: int | None = None
nalpha: int | None = None
nbeta: int | None = None
keywords: dict[str, Any] | None = None
def to_hdf5(self, h5py_data: str | Path | h5py.Group) -> None:
"""Converts the schema data object to HDF5.
Args:
h5py_data: the h5py group into which to store the object.
"""
if isinstance(h5py_data, h5py.Group):
self._to_hdf5(h5py_data)
return
with h5py.File(h5py_data, "w") as file:
self._to_hdf5(file)
def _to_hdf5(self, group: h5py.Group) -> None:
"""This internal method deals with actually converting the schema data object to HDF5.
Args:
group: the h5py group into which to store the object.
"""
# we use __dict__ here because we do not want the recursive behavior of asdict()
for key, value in self.__dict__.items():
if value is None:
continue
if isinstance(value, np.ndarray):
group.create_dataset(key, data=value)
else:
group.attrs[key] = value
@classmethod
def from_hdf5(cls, h5py_data: str | Path | h5py.Group) -> _QCSchemaData:
"""Constructs a schema object from an HDF5 object.
Args:
h5py_data: can be either the path to a file or an `h5py.Group`.
Returns:
An instance of the schema data object.
"""
if isinstance(h5py_data, h5py.Group):
return cls._from_hdf5_group(h5py_data)
with h5py.File(h5py_data, "r") as file:
return cls._from_hdf5_group(file)
@classmethod
def _from_hdf5_group(cls, h5py_group: h5py.Group) -> _QCSchemaData:
"""This internal method deals with actually constructing a schema
data object from an `h5py.Group`.
Args:
h5py_group: the actual `h5py.Group`.
Returns:
An instance of the schema data object.
"""
data = dict(h5py_group.attrs.items())
for key, value in h5py_group.items():
data[key] = value[...]
for key, value in data.items():
if key in {"symbols", "coords", "masses"}:
data[key] = value.tolist()
elif key in {"dip_nuc", "dip_ref"}:
data[key] = (value[0], value[1], value[2])
return cls(**data)
[documentos]class MethodType(Enum):
"""MethodType Enum
The HF-style methods are common names which are likely available everywhere.
The KS-style methods are not available for all drivers. Please check the specific driver
documentation for details.
"""
RHF = "rhf"
ROHF = "rohf"
UHF = "uhf"
RKS = "rks"
ROKS = "roks"
UKS = "uks"
[documentos]class ElectronicStructureDriver(BaseDriver):
"""
Base class for Qiskit Nature's electronic structure drivers.
"""
[documentos] @abstractmethod
def run(self) -> ElectronicStructureProblem:
"""Returns an :class:`.ElectronicStructureProblem` output as produced by the driver."""
pass
[documentos] @abstractmethod
def to_qcschema(self, *, include_dipole: bool = True) -> QCSchema:
"""Extracts all available information after the driver was run into a :class:`.QCSchema`
object.
Args:
include_dipole: whether or not to include the custom dipole integrals in the QCSchema.
Returns:
A :class:`.QCSchema` storing all extracted system data computed by the driver.
"""
[documentos] @abstractmethod
def to_problem(
self,
*,
basis: ElectronicBasis = ElectronicBasis.MO,
include_dipole: bool = True,
) -> ElectronicStructureProblem:
"""Extends the :meth:`to_qcschema` method and translates the :class:`.QCSchema` object to an
:class:`.ElectronicStructureProblem`.
Args:
basis: the :class:`.ElectronicBasis` in which to construct the problem.
include_dipole: whether or not to include an :class:`.ElectronicDipoleMoment` property
in the generated problem (if the data is available).
Returns:
An :class:`.ElectronicStructureProblem`.
"""
@staticmethod
def _to_qcschema(data: _QCSchemaData, *, include_dipole: bool = True) -> QCSchema:
molecule = QCTopology(
schema_name="qcschema_molecule",
schema_version=2,
symbols=data.symbols,
geometry=data.coords,
molecular_charge=data.charge,
molecular_multiplicity=data.multiplicity,
masses=data.masses,
)
properties = QCProperties()
properties.calcinfo_natom = len(data.symbols) if data.symbols is not None else None
properties.calcinfo_nbasis = data.nbasis
properties.calcinfo_nmo = data.nmo
properties.calcinfo_nalpha = data.nalpha
properties.calcinfo_nbeta = data.nbeta
properties.return_energy = data.e_ref
properties.nuclear_repulsion_energy = data.e_nuc
properties.nuclear_dipole_moment = data.dip_nuc
properties.scf_dipole_moment = data.dip_ref
def format_np_array(arr):
if isinstance(arr, Tensor):
# NOTE: this also deals with symmetry-reduced integral classes and ensures that
# they are not automatically unfolded to 1-fold symmetry
arr = arr.array
return arr.ravel().tolist()
wavefunction = QCWavefunction(basis=data.basis)
if data.overlap is not None:
wavefunction.overlap = "scf_overlap"
wavefunction.scf_overlap = format_np_array(data.overlap)
if data.mo_coeff is not None:
wavefunction.orbitals_a = "scf_orbitals_a"
wavefunction.scf_orbitals_a = format_np_array(data.mo_coeff)
if data.mo_coeff_b is not None:
wavefunction.orbitals_b = "scf_orbitals_b"
wavefunction.scf_orbitals_b = format_np_array(data.mo_coeff_b)
if data.mo_occ is not None:
wavefunction.occupations_a = "scf_occupations_a"
wavefunction.scf_occupations_a = format_np_array(data.mo_occ)
if data.mo_occ_b is not None:
wavefunction.occupations_b = "scf_occupations_b"
wavefunction.scf_occupations_b = format_np_array(data.mo_occ_b)
if data.mo_energy is not None:
wavefunction.eigenvalues_a = "scf_eigenvalues_a"
wavefunction.scf_eigenvalues_a = format_np_array(data.mo_energy)
if data.mo_energy_b is not None:
wavefunction.eigenvalues_b = "scf_eigenvalues_b"
wavefunction.scf_eigenvalues_b = format_np_array(data.mo_energy_b)
if data.hij is not None:
wavefunction.fock_a = "scf_fock_a"
wavefunction.scf_fock_a = format_np_array(data.hij)
if data.hij_b is not None:
wavefunction.fock_b = "scf_fock_b"
wavefunction.scf_fock_b = format_np_array(data.hij_b)
if data.hij_mo is not None:
wavefunction.fock_mo_a = "scf_fock_mo_a"
wavefunction.scf_fock_mo_a = format_np_array(data.hij_mo)
if data.hij_mo_b is not None:
wavefunction.fock_mo_b = "scf_fock_mo_b"
wavefunction.scf_fock_mo_b = format_np_array(data.hij_mo_b)
if data.eri is not None:
wavefunction.eri = "scf_eri"
wavefunction.scf_eri = format_np_array(data.eri)
if data.eri_mo is not None:
wavefunction.eri_mo_aa = "scf_eri_mo_aa"
wavefunction.scf_eri_mo_aa = format_np_array(data.eri_mo)
if data.eri_mo_ba is not None:
wavefunction.eri_mo_ba = "scf_eri_mo_ba"
wavefunction.scf_eri_mo_ba = format_np_array(data.eri_mo_ba)
if data.eri_mo_bb is not None:
wavefunction.eri_mo_bb = "scf_eri_mo_bb"
wavefunction.scf_eri_mo_bb = format_np_array(data.eri_mo_bb)
if include_dipole:
if data.dip_x is not None:
wavefunction.dipole_x = "scf_dipole_x"
wavefunction.scf_dipole_x = format_np_array(data.dip_x)
if data.dip_y is not None:
wavefunction.dipole_y = "scf_dipole_y"
wavefunction.scf_dipole_y = format_np_array(data.dip_y)
if data.dip_z is not None:
wavefunction.dipole_z = "scf_dipole_z"
wavefunction.scf_dipole_z = format_np_array(data.dip_z)
if data.dip_mo_x_a is not None:
wavefunction.dipole_mo_x_a = "scf_dipole_mo_x_a"
wavefunction.scf_dipole_mo_x_a = format_np_array(data.dip_mo_x_a)
if data.dip_mo_y_a is not None:
wavefunction.dipole_mo_y_a = "scf_dipole_mo_y_a"
wavefunction.scf_dipole_mo_y_a = format_np_array(data.dip_mo_y_a)
if data.dip_mo_z_a is not None:
wavefunction.dipole_mo_z_a = "scf_dipole_mo_z_a"
wavefunction.scf_dipole_mo_z_a = format_np_array(data.dip_mo_z_a)
if data.dip_mo_x_b is not None:
wavefunction.dipole_mo_x_b = "scf_dipole_mo_x_b"
wavefunction.scf_dipole_mo_x_b = format_np_array(data.dip_mo_x_b)
if data.dip_mo_y_b is not None:
wavefunction.dipole_mo_y_b = "scf_dipole_mo_y_b"
wavefunction.scf_dipole_mo_y_b = format_np_array(data.dip_mo_y_b)
if data.dip_mo_z_b is not None:
wavefunction.dipole_mo_z_b = "scf_dipole_mo_z_b"
wavefunction.scf_dipole_mo_z_b = format_np_array(data.dip_mo_z_b)
qcschema = QCSchema(
schema_name="qcschema",
schema_version=3,
molecule=molecule,
driver="energy",
model=QCModel(
method=data.method,
basis=data.basis,
),
keywords=data.keywords if data.keywords is not None else {},
provenance=QCProvenance(
creator=data.creator,
version=data.version,
routine=data.routine if data.routine is not None else "",
),
return_result=data.e_ref,
success=True,
properties=properties,
wavefunction=wavefunction,
)
return qcschema