svetlanna.phase_retrieval_problem package
Submodules
svetlanna.phase_retrieval_problem.algorithms module
- svetlanna.phase_retrieval_problem.algorithms.gerchberg_saxton_algorithm(target_intensity: Tensor, source_intensity: Tensor, forward: Callable, reverse: Callable, initial_approximation: Tensor, tol: float, maxiter: int, target_phase: Tensor | None = None, target_region: Tensor | None = None) PhaseRetrievalResult
Gerchberg-Saxton algorithm(GS) for solving the phase retrieval problem
Parameters
- target_intensitytorch.Tensor
Intensity profile in the Fourier plane
- source_intensitytorch.Tensor
Intensity distribution before the optical setup(in the image plane)
- forwardCallable
Function which describes forward propagation through the optical system
- reverseCallable
Function which describes reverse propagation through the optical system
- initial_approximationtorch.Tensor
Initial approximation for the phase profile
- tolfloat
Accuracy for the algorithm
- maxiterint
Maximum number of iterations
- target_phasetorch.Tensor | None, optional
Phase profile on the Fourier plane(optional for the generating target intensity profile problem) for reconstructing phase profile problem, by default None
- target_regiontorch.Tensor | None, optional
Region to preserve phase and amplitude profiles in the Fourier plane for reconstructing phase profile problem(optional for the generating target intensity profile problem), by default None
Returns
- prr.PhaseRetrievalResult
Exemplar of class PhaseRetrievalResult which presents result of optimization
- svetlanna.phase_retrieval_problem.algorithms.hybrid_input_output(target_intensity: Tensor, source_intensity: Tensor, forward: Callable, reverse: Callable, initial_approximation: Tensor, tol: float, maxiter: int, target_phase: Tensor | None = None, target_region: Tensor | None = None, constant_factor: float = 0.9) PhaseRetrievalResult
Hybrid Input-Output(HIO) algorithm for for solving the phase retrieval problem
Parameters
- target_intensitytorch.Tensor
Intensity profile in the Fourier plane
- source_intensitytorch.Tensor
Intensity distribution before the optical setup(in the image plane)
- forwardCallable
Function which describes forward propagation through the optical system
- reverseCallable
Function which describes reverse propagation through the optical system
- initial_approximationtorch.Tensor
Initial approximation for the phase profile
- tolfloat
Accuracy for the algorithm
- maxiterint
Maximum number of iterations
- target_phasetorch.Tensor | None, optional
Phase profile on the Fourier plane(optional for the generating target intensity profile problem) for reconstructing phase profile problem, by default None
- target_regiontorch.Tensor | None, optional
Region to preserve phase and amplitude profiles in the Fourier plane for reconstructing phase profile problem(optional for the generating target intensity profile problem), by default None
- constant_factor: float
Learning rate value for the HIO algorithm, by default 0.9
Returns
- prr.PhaseRetrievalResult
Exemplar of class PhaseRetrievalResult which presents result of optimization
svetlanna.phase_retrieval_problem.phase_retrieval module
- class svetlanna.phase_retrieval_problem.phase_retrieval.AlgorithmOptions
Bases:
TypedDict- constant_factor: float
- disp: bool
- maxiter: int
- tol: float
- class svetlanna.phase_retrieval_problem.phase_retrieval.SetupLike(*args, **kwargs)
Bases:
ProtocolA class for phase_retrieval_problem with personal realizations of forward and reverse methods instead of methods in svetlanna.setup.LinearOpticalSetup
Parameters
- Protocol_type_
_description_
- forward(input_field: Tensor) Tensor
- reverse(transmission_field: Tensor) Tensor
- svetlanna.phase_retrieval_problem.phase_retrieval.retrieve_phase(source_intensity: Tensor, optical_setup: LinearOpticalSetup | SetupLike, target_intensity: Tensor, *, initial_phase: Tensor | None = None, method: Literal['GS', 'HIO'] = 'GS', options: AlgorithmOptions | None = None) PhaseRetrievalResult
- svetlanna.phase_retrieval_problem.phase_retrieval.retrieve_phase(source_intensity: Tensor, optical_setup: LinearOpticalSetup | SetupLike, target_intensity: Tensor, target_phase: Tensor, target_region: Tensor, *, initial_phase: Tensor | None = None, method: Literal['GS', 'HIO'] = 'GS', options: AlgorithmOptions | None = None) PhaseRetrievalResult
Function for solving phase retrieval problem: generating target intensity profile or reconstructing the phase profile of the field
Parameters
- source_intensitytorch.Tensor
Intensity distribution before the optical setup
- optical_setupLinearOpticalSetup | SetupLike
Optical system through which the beam is propagated
- target_intensitytorch.Tensor
Intensity profile in the Fourier plane
- target_phasetorch.Tensor | None, optional
Phase profile on the Fourier plane(optional for the generating target intensity profile problem)
- target_regiontorch.Tensor, optional
Region to preserve phase and amplitude profiles in the Fourier plane( optional for the generating target intensity profile problem)
- initial_phasetorch.Tensor, optional
Initial approximation for the phase profile, by default None
- methodLiteral['GS', 'HIO'], optional
Algorithms for phase retrieval problem, by default ‘GS’
- optionsdict, optional
Dictionary with optimization parameters, by default { ‘tol’: 1e-16, # criteria for stop optimization ‘maxiter’: 100, # maximum number of iterations ‘constant_factor’: 0.9, # convergence parameter for HIO ‘disp’: False # show result of optimization }
Returns
- prr.PhaseRetrievalResult
Exemplar of class PhaseRetrievalResult which presents result of optimization
Raises
- ValueError
Unknown optimization method
svetlanna.phase_retrieval_problem.phase_retrieval_result module
- class svetlanna.phase_retrieval_problem.phase_retrieval_result.PhaseRetrievalResult(solution: Tensor, cost_func: float, cost_func_evolution: list[float], number_of_iterations: int)
Bases:
objectRepresents the phase retrieval result
- cost_func: float
- cost_func_evolution: list[float]
- number_of_iterations: int
- solution: Tensor
Module contents
- class svetlanna.phase_retrieval_problem.PhaseRetrievalResult(solution: Tensor, cost_func: float, cost_func_evolution: list[float], number_of_iterations: int)
Bases:
objectRepresents the phase retrieval result
- cost_func: float
- cost_func_evolution: list[float]
- number_of_iterations: int
- solution: Tensor
- class svetlanna.phase_retrieval_problem.SetupLike(*args, **kwargs)
Bases:
ProtocolA class for phase_retrieval_problem with personal realizations of forward and reverse methods instead of methods in svetlanna.setup.LinearOpticalSetup
Parameters
- Protocol_type_
_description_
- forward(input_field: Tensor) Tensor
- reverse(transmission_field: Tensor) Tensor
- svetlanna.phase_retrieval_problem.gerchberg_saxton_algorithm(target_intensity: Tensor, source_intensity: Tensor, forward: Callable, reverse: Callable, initial_approximation: Tensor, tol: float, maxiter: int, target_phase: Tensor | None = None, target_region: Tensor | None = None) PhaseRetrievalResult
Gerchberg-Saxton algorithm(GS) for solving the phase retrieval problem
Parameters
- target_intensitytorch.Tensor
Intensity profile in the Fourier plane
- source_intensitytorch.Tensor
Intensity distribution before the optical setup(in the image plane)
- forwardCallable
Function which describes forward propagation through the optical system
- reverseCallable
Function which describes reverse propagation through the optical system
- initial_approximationtorch.Tensor
Initial approximation for the phase profile
- tolfloat
Accuracy for the algorithm
- maxiterint
Maximum number of iterations
- target_phasetorch.Tensor | None, optional
Phase profile on the Fourier plane(optional for the generating target intensity profile problem) for reconstructing phase profile problem, by default None
- target_regiontorch.Tensor | None, optional
Region to preserve phase and amplitude profiles in the Fourier plane for reconstructing phase profile problem(optional for the generating target intensity profile problem), by default None
Returns
- prr.PhaseRetrievalResult
Exemplar of class PhaseRetrievalResult which presents result of optimization
- svetlanna.phase_retrieval_problem.hybrid_input_output(target_intensity: Tensor, source_intensity: Tensor, forward: Callable, reverse: Callable, initial_approximation: Tensor, tol: float, maxiter: int, target_phase: Tensor | None = None, target_region: Tensor | None = None, constant_factor: float = 0.9) PhaseRetrievalResult
Hybrid Input-Output(HIO) algorithm for for solving the phase retrieval problem
Parameters
- target_intensitytorch.Tensor
Intensity profile in the Fourier plane
- source_intensitytorch.Tensor
Intensity distribution before the optical setup(in the image plane)
- forwardCallable
Function which describes forward propagation through the optical system
- reverseCallable
Function which describes reverse propagation through the optical system
- initial_approximationtorch.Tensor
Initial approximation for the phase profile
- tolfloat
Accuracy for the algorithm
- maxiterint
Maximum number of iterations
- target_phasetorch.Tensor | None, optional
Phase profile on the Fourier plane(optional for the generating target intensity profile problem) for reconstructing phase profile problem, by default None
- target_regiontorch.Tensor | None, optional
Region to preserve phase and amplitude profiles in the Fourier plane for reconstructing phase profile problem(optional for the generating target intensity profile problem), by default None
- constant_factor: float
Learning rate value for the HIO algorithm, by default 0.9
Returns
- prr.PhaseRetrievalResult
Exemplar of class PhaseRetrievalResult which presents result of optimization
- svetlanna.phase_retrieval_problem.retrieve_phase(source_intensity: Tensor, optical_setup: LinearOpticalSetup | SetupLike, target_intensity: Tensor, target_phase: Tensor | None = None, target_region: Tensor | None = None, *, initial_phase: Tensor | None = None, method: Literal['GS', 'HIO'] = 'GS', options: AlgorithmOptions | None = None) PhaseRetrievalResult
Function for solving phase retrieval problem: generating target intensity profile or reconstructing the phase profile of the field
Parameters
- source_intensitytorch.Tensor
Intensity distribution before the optical setup
- optical_setupLinearOpticalSetup | SetupLike
Optical system through which the beam is propagated
- target_intensitytorch.Tensor
Intensity profile in the Fourier plane
- target_phasetorch.Tensor | None, optional
Phase profile on the Fourier plane(optional for the generating target intensity profile problem)
- target_regiontorch.Tensor, optional
Region to preserve phase and amplitude profiles in the Fourier plane( optional for the generating target intensity profile problem)
- initial_phasetorch.Tensor, optional
Initial approximation for the phase profile, by default None
- methodLiteral['GS', 'HIO'], optional
Algorithms for phase retrieval problem, by default ‘GS’
- optionsdict, optional
Dictionary with optimization parameters, by default { ‘tol’: 1e-16, # criteria for stop optimization ‘maxiter’: 100, # maximum number of iterations ‘constant_factor’: 0.9, # convergence parameter for HIO ‘disp’: False # show result of optimization }
Returns
- prr.PhaseRetrievalResult
Exemplar of class PhaseRetrievalResult which presents result of optimization
Raises
- ValueError
Unknown optimization method