quara.loss_function.probability_based_loss_function module

class ProbabilityBasedLossFunction(num_var=None, func_prob_dists=None, func_gradient_prob_dists=None, func_hessian_prob_dists=None, prob_dists_q=None)[source]

Bases: quara.loss_function.loss_function.LossFunction

Constructor

Subclasses have a responsibility to set on_value, on_gradient, on_hessian.

Parameters

num_var (int, optional) – number of variables, by default None
func_prob_dists (List[Callable[[np.ndarray], np.ndarray]], optional) – functions map variables to a probability distribution.
func_gradient_prob_dists (List[Callable[[int, np.ndarray], np.ndarray]], optional) – functions map variables and an index of variables to gradient of probability distributions.
func_hessian_prob_dists (List[Callable[[int, int, np.ndarray], np.ndarray]], optional) – functions map variables and indices of variables to Hessian of probability distributions.
prob_dists_q (List[np.ndarray], optional) – vectors of q, by default None.

property func_gradient_prob_dists: List[Callable[[int, numpy.ndarray], numpy.ndarray]]

returns functions map variables and an index of variables to gradient of probability distributions.

Returns: functions map variables and an index of variables to gradient of probability distributions.
Return type: List[Callable[[int, np.ndarray], np.ndarray]]

property func_hessian_prob_dists: List[Callable[[int, int, numpy.ndarray], numpy.ndarray]]

returns functions map variables and indices of variables to Hessian of probability distributions.

Returns: functions map variables and indices of variables to Hessian of probability distributions.
Return type: List[Callable[[int, int, np.ndarray], np.ndarray]]

property func_prob_dists: List[Callable[[numpy.ndarray], numpy.ndarray]]

returns functions map variables to a probability distribution.

Returns: functions map variables to a probability distribution.
Return type: List[Callable[[np.ndarray], np.ndarray]]

property on_func_gradient_prob_dists: bool

returns whether or not to support func_gradient_dists.

Returns: whether or not to support func_gradient_dists.
Return type: bool

property on_func_hessian_prob_dists: bool

returns whether or not to support func_hessian_dists.

Returns: whether or not to support func_hessian_dists.
Return type: bool

property on_func_prob_dists: bool

returns whether or not to support func_prob_dists.

Returns: whether or not to support func_prob_dists.
Return type: bool

property on_prob_dists_q: bool

returns whether or not to support prob_dists_q.

Returns: whether or not to support prob_dists_q.
Return type: bool

property prob_dists_q: List[numpy.ndarray]

returns vectors of q, by default None.

Returns: vectors of q, by default None.
Return type: List[np.ndarray]

set_from_standard_qtomography_option_data(qtomography, option, data, is_gradient_required, is_hessian_required)[source]

sets settings of loss function.

Parameters

qtomography (StandardQTomography) – StandardQTomography for settings of loss function.
option (LossFunctionOption) – ProbabilityBasedLossFunctionOption for settings of loss function.
data (List[Tuple[int, np.ndarray]]) – empirical distributions for settings of loss function.
is_gradient_required (bool) – whether or not to require gradient.
is_hessian_required (bool) – whether or not to require Hessian.

Return type

None

set_func_gradient_prob_dists(func_gradient_prob_dists)[source]

sets functions map variables and an index of variables to gradient of probability distributions.

Parameters: func_gradient_prob_dists (List[Callable[[int, np.ndarray], np.ndarray]]) – functions map variables and an index of variables to gradient of probability distributions.
Return type: None

set_func_gradient_prob_dists_from_standard_qt(qt)[source]

sets the gradient of probability distributions from StandardQTomography.

Parameters: qt (StandardQTomography) – StandardQTomography to set the gradient of probability distributions.
Return type: None

set_func_hessian_prob_dists(func_hessian_prob_dists)[source]

sets functions map variables and indices of variables to Hessian of probability distributions.

Parameters: func_hessian_prob_dists (List[Callable[[int, int, np.ndarray], np.ndarray]]) – functions map variables and indices of variables to Hessian of probability distributions.
Return type: None

set_func_hessian_prob_dists_from_standard_qt(qt)[source]

sets the Hessian of probability distributions from StandardQTomography.

Parameters: qt (StandardQTomography) – StandardQTomography to set the Hessian of probability distributions.
Return type: None

set_func_prob_dists(func_prob_dists)[source]

sets functions map variables to a probability distribution.

Parameters: func_prob_dists (List[Callable[[np.ndarray], np.ndarray]]) – functions map variables to a probability distribution.
Return type: None

set_func_prob_dists_from_standard_qt(qt)[source]

sets the function of probability distributions from StandardQTomography.

Parameters: qt (StandardQTomography) – StandardQTomography to set the function of probability distributions.
Return type: None

set_prob_dists_q(prob_dists_q)[source]

sets vectors of q, by default None.

Parameters: prob_dists_q (List[np.ndarray]) – vectors of q, by default None.
Return type: None

size_prob_dists()[source]

returns size of func_prob_dists.

Returns: size of func_prob_dists.
Return type: int

class ProbabilityBasedLossFunctionOption(mode_weight, weights=None, weight_name=None)[source]

Bases: quara.loss_function.loss_function.LossFunctionOption

Constructor

Parameters

mode_weight (str, optional) – mode weight, by default None
weights (List, optional) – list of weight, by default None
weight_name (str, optional) – weight name for reporting, by default None