quara.minimization_algorithm.minimization_algorithm module

class MinimizationAlgorithm[source]

Bases: object

Constructor.

Subclasses have a responsibility to set the following variables.

_is_gradient_required: whether or not to require gradient.
_is_hessian_required: whether or not to require Hessian.

property is_gradient_required: bool

returns whether or not to require gradient.

Returns: whether or not to require gradient.
Return type: bool

property is_hessian_required: bool

returns whether or not to require Hessian.

Returns: whether or not to require Hessian.
Return type: bool

is_loss_and_option_sufficient()[source]

returns whether the loss and the option are sufficient.

In the default implementation, this function returns True. Override with subclasses as needed.

Returns: whether the loss and the option are sufficient.
Return type: bool

is_loss_sufficient()[source]

returns whether the loss is sufficient.

In the default implementation, this function returns True. Override with subclasses as needed.

Returns: whether the loss is sufficient.
Return type: bool

is_option_sufficient()[source]

returns whether the option is sufficient.

In the default implementation, this function returns True. Override with subclasses as needed.

Returns: whether the option is sufficient.
Return type: bool

property loss: quara.loss_function.loss_function.LossFunction

returns loss function.

Returns: loss function.
Return type: LossFunction

abstract optimize(loss_function, loss_function_option, algorithm_option, on_iteration_history=False)[source]

optimizes using specified parameters.

Parameters

loss_function (LossFunction) – Loss Function
loss_function_option (LossFunctionOption) – Loss Function Option
algorithm_option (MinimizationAlgorithmOption) – Minimization Algorithm Option
on_iteration_history (bool, optional) – whether to return iteration history, by default False

Returns

the result of the optimization.

Return type

MinimizationResult

Raises

NotImplementedError – this function does not be implemented in the subclass.

property option: quara.minimization_algorithm.minimization_algorithm.MinimizationAlgorithmOption

returns algorithm option.

Returns: algorithm option.
Return type: LossFunctionOption

abstract set_constraint_from_standard_qt_and_option(qt, option)[source]

sets constraint from StandardQTomography and Algorithm Option.

Parameters

qt (StandardQTomography) – StandardQTomography to set constraint.
option (MinimizationAlgorithmOption) – Algorithm Option.

Raises

NotImplementedError – this function does not be implemented in the subclass.

Return type

None

set_from_loss(loss)[source]

sets from LossFunction and calls is_loss_sufficient function.

Parameters: loss (MinimizationAlgorithmOption) – loss to set.
Return type: None

set_from_option(option)[source]

sets option from MinimizationAlgorithmOption and calls is_option_sufficient function.

Parameters: option (MinimizationAlgorithmOption) – option to set.
Return type: None

class MinimizationAlgorithmOption(on_algo_eq_constraint=True, on_algo_ineq_constraint=True, var_start=None, max_iteration_optimization=None)[source]

Bases: object

Constructor

Parameters

on_algo_eq_constraint (bool, optional) – whether this algorithm needs on algorithm equality constraint, by default True
on_algo_ineq_constraint (bool, optional) – whether this algorithm needs on algorithm inequality constraint, by default True
var_start (np.ndarray, optional) – initial variable for the algorithm, by default None.
max_iteration_optimization (int, optional) – maximun number of iterations of optimization, by default None.

property max_iteration_optimization: int

returns imaximun number of iterations of optimization.

Returns: maximun number of iterations of optimization.
Return type: int

property on_algo_eq_constraint: bool

returns whether this algorithm needs on algorithm equality constraint.

Returns: whether this algorithm needs on algorithm equality constraint.
Return type: bool

property on_algo_ineq_constraint: bool

returns whether this QOperation is on algorithm inequality constraint.

Returns: whether this QOperation is on algorithm inequality constraint.
Return type: bool

property var_start: numpy.ndarray

returns initial variable for the algorithm.

Returns: initial variable for the algorithm.
Return type: np.ndarray

class MinimizationResult(value, computation_time=None)[source]

Bases: object

Constructor

Parameters

value (np.ndarray) – the result of the minimization.
computation_time (float, optional) – computation time for the minimization, by default None

property computation_time: float

returns computation time for the estimate.

Returns: computation time for the estimate.
Return type: float

property value: numpy.ndarray

returns the result of the minimization.

Returns: the result of the minimization.
Return type: np.ndarray