__all__ = [
"NormalEquationsInversion",
"RegularizedOperator",
"RegularizedInversion",
"PreconditionedInversion",
]
import logging
from typing import TYPE_CHECKING, Optional, Sequence, Tuple
import numpy as np
from scipy.sparse.linalg import cg as sp_cg
from scipy.sparse.linalg import lsqr
from pylops.basicoperators import Diagonal, VStack
from pylops.optimization.basesolver import Solver
from pylops.optimization.basic import cg, cgls
from pylops.utils.backend import get_array_module
from pylops.utils.typing import NDArray
if TYPE_CHECKING:
from pylops.linearoperator import LinearOperator
logging.basicConfig(format="%(levelname)s: %(message)s", level=logging.WARNING)
def _check_regularization_dims(
Regs: Sequence["LinearOperator"],
dataregs: Optional[Sequence[NDArray]] = None,
epsRs: Optional[Sequence[float]] = None,
) -> None:
"""check Regs, dataregs, and epsRs have same dimensions"""
nRegs = len(Regs)
ndataregs = nRegs if dataregs is None else len(dataregs)
nepsRs = nRegs if epsRs is None else len(epsRs)
if not nRegs == ndataregs == nepsRs:
raise ValueError("Regs, dataregs, and epsRs must have the same size")
[docs]class NormalEquationsInversion(Solver):
r"""Inversion of normal equations.
Solve the regularized normal equations for a system of equations
given the operator ``Op``, a data weighting operator ``Weight`` and
optionally a list of regularization terms ``Regs`` and/or ``NRegs``.
Parameters
----------
Op : :obj:`pylops.LinearOperator`
Operator to invert of size :math:`[N \times M]`.
See Also
--------
RegularizedInversion: Regularized inversion
PreconditionedInversion: Preconditioned inversion
Notes
-----
Solve the following normal equations for a system of regularized equations
given the operator :math:`\mathbf{Op}`, a data weighting operator
:math:`\mathbf{W}`, a list of regularization terms (:math:`\mathbf{R}_i`
and/or :math:`\mathbf{N}_i`), the data :math:`\mathbf{y}` and
regularization data :math:`\mathbf{y}_{\mathbf{R}_i}`, and the damping factors
:math:`\epsilon_I`, :math:`\epsilon_{\mathbf{R}_i}` and :math:`\epsilon_{\mathbf{N}_i}`:
.. math::
( \mathbf{Op}^T \mathbf{W} \mathbf{Op} +
\sum_i \epsilon_{\mathbf{R}_i}^2 \mathbf{R}_i^T \mathbf{R}_i +
\sum_i \epsilon_{\mathbf{N}_i}^2 \mathbf{N}_i +
\epsilon_I^2 \mathbf{I} ) \mathbf{x}
= \mathbf{Op}^T \mathbf{W} \mathbf{y} + \sum_i \epsilon_{\mathbf{R}_i}^2
\mathbf{R}_i^T \mathbf{y}_{\mathbf{R}_i}
Note that the data term of the regularizations :math:`\mathbf{N}_i` is
implicitly assumed to be zero.
"""
def _print_setup(self) -> None:
self._print_solver(nbar=55)
strreg = f"Regs={self.Regs}"
streps = f"\nepsRs={self.epsRs} epsI={self.epsI}"
print(strreg + streps)
print("-" * 55)
def _print_finalize(self) -> None:
print(f"\nTotal time (s) = {self.telapsed:.2f}")
print("-" * 55 + "\n")
def setup(
self,
y: NDArray,
Regs: Sequence["LinearOperator"],
Weight: Optional["LinearOperator"] = None,
dataregs: Optional[Sequence[NDArray]] = None,
epsI: float = 0,
epsRs: Optional[Sequence[float]] = None,
NRegs: Optional[Sequence["LinearOperator"]] = None,
epsNRs: Optional[Sequence[float]] = None,
show: bool = False,
) -> None:
r"""Setup solver
Parameters
----------
y : :obj:`np.ndarray`
Data of size :math:`[N \times 1]`
Regs : :obj:`list`
Regularization operators (``None`` to avoid adding regularization)
Weight : :obj:`pylops.LinearOperator`, optional
Weight operator
dataregs : :obj:`list`, optional
Regularization data (must have the same number of elements
as ``Regs``)
epsI : :obj:`float`, optional
Tikhonov damping
epsRs : :obj:`list`, optional
Regularization dampings (must have the same number of elements
as ``Regs``)
NRegs : :obj:`list`
Normal regularization operators (``None`` to avoid adding
regularization). Such operators must apply the chain of the
forward and the adjoint in one go. This can be convenient in
cases where a faster implementation is available compared to applying
the forward followed by the adjoint.
epsNRs : :obj:`list`, optional
Regularization dampings for normal operators (must have the same
number of elements as ``NRegs``)
show : :obj:`bool`, optional
Display setup log
"""
self.y = y
self.Regs = Regs
self.epsI = epsI
self.epsRs = epsRs
self.dataregs = dataregs
self.ncp = get_array_module(y)
# check consistency in regularization terms
if Regs is not None:
_check_regularization_dims(Regs, dataregs, epsRs)
# store adjoint
self.OpH = self.Op.H
# create dataregs and epsRs if not provided
if dataregs is None and Regs is not None:
self.dataregs = [
self.ncp.zeros(int(Reg.shape[0]), dtype=Reg.dtype) for Reg in Regs
]
if epsRs is None and Regs is not None:
self.epsRs = [1] * len(Regs)
# normal equations
if Weight is not None:
self.y_normal = self.Op.rmatvec(Weight.matvec(y))
else:
self.y_normal = self.Op.rmatvec(y)
if Weight is not None:
self.Op_normal = self.OpH @ Weight @ self.Op
else:
self.Op_normal = self.OpH @ self.Op
# add regularization terms
if epsI > 0:
self.Op_normal += epsI**2 * Diagonal(
self.ncp.ones(self.Op.dims, dtype=self.Op.dtype),
dtype=self.Op.dtype,
)
if (
self.epsRs is not None
and self.Regs is not None
and self.dataregs is not None
):
for epsR, Reg, datareg in zip(self.epsRs, self.Regs, self.dataregs):
self.y_normal += epsR**2 * Reg.rmatvec(datareg)
self.Op_normal += epsR**2 * Reg.H @ Reg
if epsNRs is not None and NRegs is not None:
for epsNR, NReg in zip(epsNRs, NRegs):
self.Op_normal += epsNR**2 * NReg
# print setup
if show:
self._print_setup()
def step(self) -> None:
raise NotImplementedError(
"NormalEquationsInversion uses as default the"
" scipy.sparse.linalg.cg solver, therefore the "
"step method is not implemented. Use directly run or solve."
)
def run(
self,
x: NDArray,
engine: str = "scipy",
show: bool = False,
**kwargs_solver,
) -> Tuple[NDArray, int]:
r"""Run solver
Parameters
----------
x : :obj:`np.ndarray`
Current model vector to be updated by multiple steps of the solver.
If ``None``, x is assumed to be a zero vector
engine : :obj:`str`, optional
Solver to use (``scipy`` or ``pylops``)
show : :obj:`bool`, optional
Display iterations log
**kwargs_solver
Arbitrary keyword arguments for chosen solver
(:py:func:`scipy.sparse.linalg.cg` and
:py:func:`pylops.optimization.solver.cg` are used as default for numpy
and cupy `data`, respectively)
.. note::
When user does not supply ``atol``, it is set to "legacy".
Returns
-------
xinv : :obj:`numpy.ndarray`
Inverted model.
istop : :obj:`int`
Convergence information (only when using :py:func:`scipy.sparse.linalg.cg`):
``0``: successful exit
``>0``: convergence to tolerance not achieved, number of iterations
``<0``: illegal input or breakdown
"""
if x is not None:
self.y_normal = self.y_normal - self.Op_normal.matvec(x)
if engine == "scipy" and self.ncp == np:
if "atol" not in kwargs_solver:
kwargs_solver["atol"] = "legacy"
xinv, istop = sp_cg(self.Op_normal, self.y_normal, **kwargs_solver)
elif engine == "pylops" or self.ncp != np:
if show:
kwargs_solver["show"] = True
xinv = cg(
self.Op_normal,
self.y_normal,
self.ncp.zeros(self.Op_normal.shape[1], dtype=self.Op_normal.dtype),
**kwargs_solver,
)[0]
istop = None
else:
raise NotImplementedError("Engine must be scipy or pylops")
if x is not None:
xinv = x + xinv
return xinv, istop
def solve(
self,
y: NDArray,
Regs: Sequence["LinearOperator"],
x0: Optional[NDArray] = None,
Weight: Optional["LinearOperator"] = None,
dataregs: Optional[Sequence[NDArray]] = None,
epsI: float = 0,
epsRs: Optional[Sequence[float]] = None,
NRegs: Optional[Sequence["LinearOperator"]] = None,
epsNRs: Optional[Sequence[float]] = None,
engine: str = "scipy",
show: bool = False,
**kwargs_solver,
) -> Tuple[NDArray, int]:
r"""Run entire solver
Parameters
----------
y : :obj:`np.ndarray`
Data of size :math:`[N \times 1]`
Regs : :obj:`list`
Regularization operators (``None`` to avoid adding regularization)
x0 : :obj:`np.ndarray`, optional
Initial guess of size :math:`[M \times 1]`. If ``None``, initialize
internally as zero vector
Weight : :obj:`pylops.LinearOperator`, optional
Weight operator
dataregs : :obj:`list`, optional
Regularization data (must have the same number of elements
as ``Regs``)
epsI : :obj:`float`, optional
Tikhonov damping
epsRs : :obj:`list`, optional
Regularization dampings (must have the same number of elements
as ``Regs``)
NRegs : :obj:`list`
Normal regularization operators (``None`` to avoid adding
regularization). Such operators must apply the chain of the
forward and the adjoint in one go. This can be convenient in
cases where a faster implementation is available compared to applying
the forward followed by the adjoint.
epsNRs : :obj:`list`, optional
Regularization dampings for normal operators (must have the same
number of elements as ``NRegs``)
engine : :obj:`str`, optional
Solver to use (``scipy`` or ``pylops``)
show : :obj:`bool`, optional
Display setup log
Returns
-------
x : :obj:`np.ndarray`
Estimated model of size :math:`[N \times 1]`
istop : :obj:`int`
Convergence information (only when using :py:func:`scipy.sparse.linalg.cg`):
``0``: successful exit
``>0``: convergence to tolerance not achieved, number of iterations
``<0``: illegal input or breakdown
"""
self.setup(
y=y,
Regs=Regs,
Weight=Weight,
dataregs=dataregs,
epsI=epsI,
epsRs=epsRs,
NRegs=NRegs,
epsNRs=epsNRs,
show=show,
)
x, istop = self.run(x0, engine=engine, show=show, **kwargs_solver)
self.finalize(show)
return x, istop
def RegularizedOperator(
Op: "LinearOperator",
Regs: Sequence["LinearOperator"],
epsRs: Sequence[float] = (1,),
) -> "LinearOperator":
r"""Regularized operator.
Creates a regularized operator given the operator ``Op``
and a list of regularization terms ``Regs``.
Parameters
----------
Op : :obj:`pylops.LinearOperator`
Operator to invert
Regs : :obj:`tuple` or :obj:`list`
Regularization operators
epsRs : :obj:`tuple` or :obj:`list`, optional
Regularization dampings
Returns
-------
OpReg : :obj:`pylops.LinearOperator`
Regularized operator
See Also
--------
RegularizedInversion: Regularized inversion
Notes
-----
Create a regularized operator by augumenting the problem operator
:math:`\mathbf{Op}`, by a set of regularization terms :math:`\mathbf{R_i}`
and their damping factors and :math:`\epsilon_{{R}_i}`:
.. math::
\begin{bmatrix}
\mathbf{Op} \\
\epsilon_{\mathbf{R}_1} \mathbf{R}_1 \\
... \\
\epsilon_{R_N} \mathbf{R}_N
\end{bmatrix}
"""
OpReg = VStack(
[Op] + [epsR * Reg for epsR, Reg in zip(epsRs, Regs)], dtype=Op.dtype
)
return OpReg
[docs]class RegularizedInversion(Solver):
r"""Regularized inversion.
Solve a system of regularized equations given the operator ``Op``,
a data weighting operator ``Weight``, and a list of regularization
terms ``Regs``.
Parameters
----------
Op : :obj:`pylops.LinearOperator`
Operator to invert of size :math:`[N \times M]`.
See Also
--------
RegularizedOperator: Regularized operator
NormalEquationsInversion: Normal equations inversion
PreconditionedInversion: Preconditioned inversion
Notes
-----
Solve the following system of regularized equations given the operator
:math:`\mathbf{Op}`, a data weighting operator :math:`\mathbf{W}^{1/2}`,
a list of regularization terms :math:`\mathbf{R}_i`,
the data :math:`\mathbf{y}` and
regularization data :math:`\mathbf{y}_{\mathbf{R}_i}`, and the damping factors
:math:`\epsilon_\mathbf{I}`: and :math:`\epsilon_{\mathbf{R}_i}`:
.. math::
\begin{bmatrix}
\mathbf{W}^{1/2} \mathbf{Op} \\
\epsilon_{\mathbf{R}_1} \mathbf{R}_1 \\
\vdots \\
\epsilon_{\mathbf{R}_N} \mathbf{R}_N
\end{bmatrix} \mathbf{x} =
\begin{bmatrix}
\mathbf{W}^{1/2} \mathbf{y} \\
\epsilon_{\mathbf{R}_1} \mathbf{y}_{\mathbf{R}_1} \\
\vdots \\
\epsilon_{\mathbf{R}_N} \mathbf{y}_{\mathbf{R}_N} \\
\end{bmatrix}
where the ``Weight`` provided here is equivalent to the
square-root of the weight in
:py:func:`pylops.optimization.leastsquares.NormalEquationsInversion`. Note
that this system is solved using the :py:func:`scipy.sparse.linalg.lsqr`
and an initial guess ``x0`` can be provided to this solver, despite the
original solver does not allow so.
"""
def _print_setup(self) -> None:
self._print_solver(nbar=65)
strreg = f"Regs={self.Regs}"
streps = f"\nepsRs={self.epsRs}"
print(strreg + streps)
print("-" * 65)
def _print_finalize(self) -> None:
print(f"\nTotal time (s) = {self.telapsed:.2f}")
print("-" * 65 + "\n")
def setup(
self,
y: NDArray,
Regs: Sequence["LinearOperator"],
Weight: Optional["LinearOperator"] = None,
dataregs: Optional[Sequence[NDArray]] = None,
epsRs: Optional[Sequence[float]] = None,
show: bool = False,
) -> None:
r"""Setup solver
Parameters
----------
y : :obj:`np.ndarray`
Data of size :math:`[N \times 1]`
Regs : :obj:`list`
Regularization operators (``None`` to avoid adding regularization)
Weight : :obj:`pylops.LinearOperator`, optional
Weight operator
dataregs : :obj:`list`, optional
Regularization data (must have the same number of elements
as ``Regs``)
epsRs : :obj:`list`, optional
Regularization dampings (must have the same number of elements
as ``Regs``)
show : :obj:`bool`, optional
Display setup log
"""
self.y = y
self.Regs = Regs
self.epsRs = epsRs
self.dataregs = dataregs
self.ncp = get_array_module(y)
# check consistency in regularization terms
if Regs is not None:
_check_regularization_dims(Regs, dataregs, epsRs)
# create regularization data
if dataregs is None and Regs is not None:
self.dataregs = [
self.ncp.zeros(int(Reg.shape[0]), dtype=Reg.dtype) for Reg in Regs
]
if self.epsRs is None and Regs is not None:
self.epsRs = [1] * len(Regs)
# create regularization operators
self.RegOp: LinearOperator
if Weight is not None:
if Regs is None:
self.RegOp = Weight @ self.Op
else:
self.RegOp = RegularizedOperator(
Weight @ self.Op, Regs, epsRs=self.epsRs
)
else:
if Regs is None:
self.RegOp = self.Op
else:
self.RegOp = RegularizedOperator(self.Op, Regs, epsRs=self.epsRs)
# augumented data
if Weight is not None:
self.datatot: NDArray = Weight.matvec(self.y.copy())
else:
self.datatot = self.y.copy()
# augumented operator
if self.epsRs is not None and self.dataregs is not None:
for epsR, datareg in zip(self.epsRs, self.dataregs):
self.datatot = np.hstack((self.datatot, epsR * datareg))
# print setup
if show:
self._print_setup()
def step(self) -> None:
raise NotImplementedError(
"RegularizedInversion uses as default the"
" scipy.sparse.linalg.lsqr solver, therefore the "
"step method is not implemented. Use directly run or solve."
)
def run(
self,
x: NDArray,
engine: str = "scipy",
show: bool = False,
**kwargs_solver,
) -> Tuple[NDArray, int, int, float, float]:
r"""Run solver
Parameters
----------
x : :obj:`np.ndarray`
Current model vector to be updated by multiple steps of the solver.
If ``None``, x is assumed to be a zero vector
engine : :obj:`str`, optional
Solver to use (``scipy`` or ``pylops``)
show : :obj:`bool`, optional
Display iterations log
**kwargs_solver
Arbitrary keyword arguments for chosen solver
(:py:func:`scipy.sparse.linalg.lsqr` and
:py:func:`pylops.optimization.solver.cgls` are used for engine ``scipy``
and ``pylops``, respectively)
Returns
-------
xinv : :obj:`numpy.ndarray`
Inverted model.
istop : :obj:`int`
Gives the reason for termination
``1`` means :math:`\mathbf{x}` is an approximate solution to
:math:`\mathbf{y} = \mathbf{Op}\,\mathbf{x}`
``2`` means :math:`\mathbf{x}` approximately solves the least-squares
problem
itn : :obj:`int`
Iteration number upon termination
r1norm : :obj:`float`
:math:`||\mathbf{r}||_2^2`, where
:math:`\mathbf{r} = \mathbf{y} - \mathbf{Op}\,\mathbf{x}`
r2norm : :obj:`float`
:math:`\sqrt{\mathbf{r}^T\mathbf{r} +
\epsilon^2 \mathbf{x}^T\mathbf{x}}`.
Equal to ``r1norm`` if :math:`\epsilon=0`
"""
if x is not None:
self.datatot = self.datatot - self.RegOp.matvec(x)
if engine == "scipy" and self.ncp == np:
if show:
kwargs_solver["show"] = 1
xinv, istop, itn, r1norm, r2norm = lsqr(
self.RegOp, self.datatot, **kwargs_solver
)[0:5]
elif engine == "pylops" or self.ncp != np:
if show:
kwargs_solver["show"] = True
xinv, istop, itn, r1norm, r2norm = cgls(
self.RegOp,
self.datatot,
self.ncp.zeros(self.RegOp.dims, dtype=self.RegOp.dtype),
**kwargs_solver,
)[0:5]
else:
raise NotImplementedError("Engine must be scipy or pylops")
if x is not None:
xinv = x + xinv
return xinv, istop, itn, r1norm, r2norm
def solve(
self,
y: NDArray,
Regs: Sequence["LinearOperator"],
x0: Optional[NDArray] = None,
Weight: Optional["LinearOperator"] = None,
dataregs: Optional[Sequence[NDArray]] = None,
epsRs: Optional[Sequence[float]] = None,
engine: str = "scipy",
show: bool = False,
**kwargs_solver,
) -> Tuple[NDArray, int, int, float, float]:
r"""Run entire solver
Parameters
----------
y : :obj:`np.ndarray`
Data of size :math:`[N \times 1]`
Regs : :obj:`list`
Regularization operators (``None`` to avoid adding regularization)
x0 : :obj:`numpy.ndarray`, optional
Initial guess
Weight : :obj:`pylops.LinearOperator`, optional
Weight operator
dataregs : :obj:`list`, optional
Regularization data (must have the same number of elements
as ``Regs``)
epsRs : :obj:`list`, optional
Regularization dampings (must have the same number of elements
as ``Regs``)
engine : :obj:`str`, optional
Solver to use (``scipy`` or ``pylops``)
show : :obj:`bool`, optional
Display log
**kwargs_solver
Arbitrary keyword arguments for chosen solver
(:py:func:`scipy.sparse.linalg.lsqr` and
:py:func:`pylops.optimization.solver.cgls` are used for engine ``scipy``
and ``pylops``, respectively)
Returns
-------
xinv : :obj:`numpy.ndarray`
Inverted model.
istop : :obj:`int`
Gives the reason for termination
``1`` means :math:`\mathbf{x}` is an approximate solution to
:math:`\mathbf{y} = \mathbf{Op}\,\mathbf{x}`
``2`` means :math:`\mathbf{x}` approximately solves the least-squares
problem
itn : :obj:`int`
Iteration number upon termination
r1norm : :obj:`float`
:math:`||\mathbf{r}||_2^2`, where
:math:`\mathbf{r} = \mathbf{y} - \mathbf{Op}\,\mathbf{x}`
r2norm : :obj:`float`
:math:`\sqrt{\mathbf{r}^T\mathbf{r} +
\epsilon^2 \mathbf{x}^T\mathbf{x}}`.
Equal to ``r1norm`` if :math:`\epsilon=0`
"""
self.setup(
y=y, Regs=Regs, Weight=Weight, dataregs=dataregs, epsRs=epsRs, show=show
)
x, istop, itn, r1norm, r2norm = self.run(
x0, engine=engine, show=show, **kwargs_solver
)
self.finalize(show)
return x, istop, itn, r1norm, r2norm
[docs]class PreconditionedInversion(Solver):
r"""Preconditioned inversion.
Solve a system of preconditioned equations given the operator
``Op`` and a preconditioner ``P``.
Parameters
----------
Op : :obj:`pylops.LinearOperator`
Operator to invert of size :math:`[N \times M]`.
See Also
--------
RegularizedInversion: Regularized inversion
NormalEquationsInversion: Normal equations inversion
Notes
-----
Solve the following system of preconditioned equations given the operator
:math:`\mathbf{Op}`, a preconditioner :math:`\mathbf{P}`,
the data :math:`\mathbf{y}`
.. math::
\mathbf{y} = \mathbf{Op}\,\mathbf{P} \mathbf{p}
where :math:`\mathbf{p}` is the solution in the preconditioned space
and :math:`\mathbf{x} = \mathbf{P}\mathbf{p}` is the solution in the
original space.
"""
def _print_setup(self) -> None:
self._print_solver(nbar=65)
strprec = f"Prec={self.P}"
print(strprec)
print("-" * 65)
def _print_finalize(self) -> None:
print(f"\nTotal time (s) = {self.telapsed:.2f}")
print("-" * 65 + "\n")
def setup(
self,
y: NDArray,
P: "LinearOperator",
show: bool = False,
) -> None:
r"""Setup solver
Parameters
----------
y : :obj:`np.ndarray`
Data of size :math:`[N \times 1]`
P : :obj:`pylops.LinearOperator`
Preconditioner
show : :obj:`bool`, optional
Display setup log
"""
self.y = y
self.P = P
self.ncp = get_array_module(y)
# preconditioned operator
self.POp = self.Op @ P
# print setup
if show:
self._print_setup()
def step(self) -> None:
raise NotImplementedError(
"PreconditionedInversion uses as default the"
" scipy.sparse.linalg.lsqr solver, therefore the "
"step method is not implemented. Use directly run or solve."
)
def run(
self,
x: NDArray,
engine: str = "scipy",
show: bool = False,
**kwargs_solver,
) -> Tuple[NDArray, int, int, float, float]:
r"""Run solver
Parameters
----------
x : :obj:`np.ndarray`
Current model vector to be updated by multiple steps of the solver.
If ``None``, x is assumed to be a zero vector
engine : :obj:`str`, optional
Solver to use (``scipy`` or ``pylops``)
show : :obj:`bool`, optional
Display iterations log
**kwargs_solver
Arbitrary keyword arguments for chosen solver
(:py:func:`scipy.sparse.linalg.lsqr` and
:py:func:`pylops.optimization.solver.cgls` are used for engine ``scipy``
and ``pylops``, respectively)
Returns
-------
xinv : :obj:`numpy.ndarray`
Inverted model.
istop : :obj:`int`
Gives the reason for termination
``1`` means :math:`\mathbf{x}` is an approximate solution to
:math:`\mathbf{y} = \mathbf{Op}\,\mathbf{x}`
``2`` means :math:`\mathbf{x}` approximately solves the least-squares
problem
itn : :obj:`int`
Iteration number upon termination
r1norm : :obj:`float`
:math:`||\mathbf{r}||_2^2`, where
:math:`\mathbf{r} = \mathbf{y} - \mathbf{Op}\,\mathbf{x}`
r2norm : :obj:`float`
:math:`\sqrt{\mathbf{r}^T\mathbf{r} +
\epsilon^2 \mathbf{x}^T\mathbf{x}}`.
Equal to ``r1norm`` if :math:`\epsilon=0`
"""
if x is not None:
self.y = self.y - self.Op.matvec(x)
if engine == "scipy" and self.ncp == np:
if show:
kwargs_solver["show"] = 1
pinv, istop, itn, r1norm, r2norm = lsqr(
self.POp,
self.y,
**kwargs_solver,
)[0:5]
elif engine == "pylops" or self.ncp != np:
if show:
kwargs_solver["show"] = True
pinv, istop, itn, r1norm, r2norm = cgls(
self.POp,
self.y,
self.ncp.zeros(self.POp.shape[1], dtype=self.POp.dtype),
**kwargs_solver,
)[0:5]
# force it 1d as we decorate this method with disable_ndarray_multiplication
pinv = pinv.ravel()
else:
raise NotImplementedError("Engine must be scipy or pylops")
xinv = self.P.matvec(pinv)
if x is not None:
xinv = x + xinv
return xinv, istop, itn, r1norm, r2norm
def solve(
self,
y: NDArray,
P: "LinearOperator",
x0: Optional[NDArray] = None,
engine: str = "scipy",
show: bool = False,
**kwargs_solver,
) -> Tuple[NDArray, int, int, float, float]:
r"""Run entire solver
Parameters
----------
y : :obj:`np.ndarray`
Data of size :math:`[N \times 1]`
P : :obj:`pylops.LinearOperator`
Preconditioner
x0 : :obj:`np.ndarray`, optional
Initial guess of size :math:`[M \times 1]`. If ``None``, initialize
internally as zero vector
engine : :obj:`str`, optional
Solver to use (``scipy`` or ``pylops``)
show : :obj:`bool`, optional
Display log
**kwargs_solver
Arbitrary keyword arguments for chosen solver
(:py:func:`scipy.sparse.linalg.lsqr` and
:py:func:`pylops.optimization.solver.cgls` are used for engine ``scipy``
and ``pylops``, respectively)
Returns
-------
x : :obj:`numpy.ndarray`
Inverted model.
istop : :obj:`int`
Gives the reason for termination
``1`` means :math:`\mathbf{x}` is an approximate solution to
:math:`\mathbf{y} = \mathbf{Op}\,\mathbf{x}`
``2`` means :math:`\mathbf{x}` approximately solves the least-squares
problem
itn : :obj:`int`
Iteration number upon termination
r1norm : :obj:`float`
:math:`||\mathbf{r}||_2^2`, where
:math:`\mathbf{r} = \mathbf{y} - \mathbf{Op}\,\mathbf{x}`
r2norm : :obj:`float`
:math:`\sqrt{\mathbf{r}^T\mathbf{r} +
\epsilon^2 \mathbf{x}^T\mathbf{x}}`.
Equal to ``r1norm`` if :math:`\epsilon=0`
"""
self.setup(y=y, P=P, show=show)
x, istop, itn, r1norm, r2norm = self.run(
x0, engine=engine, show=show, **kwargs_solver
)
self.finalize(show)
return x, istop, itn, r1norm, r2norm
