pybop.costs.feature_distances#

Classes#

`ExponentialFeatureDistance`	Exponential fit cost function.
`FeatureDistance`	Base for defining cost functions based on comparing fit functions.
`SquareRootFeatureDistance`	Square-root fit cost function.

Functions#

indices_of(values, target)

Module Contents#

class pybop.costs.feature_distances.ExponentialFeatureDistance(dataset: pybop.processing.dataset.Dataset, target: str = None, feature: str = 'inverse_timescale', time_start: float = None, time_end: float = None)[source]#

Bases: FeatureDistance

Exponential fit cost function.

Fits an exponential and compares either its asymptote, its magnitude, or its timescale between model predictions and target data.

Supported features: - “asymptote”: The exponential fit value at infinite time. - “magnitude”: The prefactor of the exponential term. - “timescale”: The denominator in the exponential argument. - “inverse_timescale”: 1 over “timescale”; may perform better.

_feature_selection(fit)[source]#

_fit_guess(t, y)[source]#

_inverse_fit_function(y, b, c, d)[source]#: Logarithm function to transform data for a linear fit.

_supported_features = ['asymptote', 'magnitude', 'timescale', 'inverse_timescale']#

class pybop.costs.feature_distances.FeatureDistance(dataset: pybop.processing.dataset.Dataset, target: str = None, feature: str = None, time_start: float = None, time_end: float = None)[source]#

Bases: pybop.costs.base_cost.BaseCost

Base for defining cost functions based on comparing fit functions.

Parameters:

dataset (pybop.Dataset) – Dataset object containing the target data.
target (str, optional) – The name of the target variable.
feature (str, optional) – The name of the parameter to use for fitting, must be one of the supported features.
time_start (float, optional) – Set the time (in seconds) from which onwards the data shall be fitted, counted from the start of the data. Default is the start.
time_end (float, optional) – Set the time (in seconds) until which the data shall be fitted, counted from the start of the data. Default is the end.

__call__(y: numpy.ndarray, dy: numpy.ndarray | None = None) → float | tuple[float, numpy.ndarray][source]#

_feature_selection(fit)[source]#

_fit(t, y)[source]#: Uses SciPy to fit data. For numerical reasons, the fitting involves applying the fit function to data and comparing to identity.

_fit_guess(t, y)[source]#

_inverse_fit_function(y, *args)[source]#

evaluate(solution: pybop.simulators.solution.Solution, inputs: pybop.parameters.parameter.Inputs | None = None, calculate_sensitivities: bool = False) → pybop.costs.evaluation.Evaluation[source]#: Evaluate the feature distance for the given solution.

_supported_features = []#

end_index#

feature = None#

start_index#

time_end = None#

time_start = None#

class pybop.costs.feature_distances.SquareRootFeatureDistance(dataset: pybop.processing.dataset.Dataset, target: str = None, feature: str = 'inverse_slope', time_start: float = None, time_end: float = None)[source]#

Bases: FeatureDistance

Square-root fit cost function.

Fits a square-root fit function and compares either its offset or its slope between model predictions and target data.

Supported features: - “offset”: The value of the square-root fit at the start. - “slope”: The prefactor of the square-root over time. - “inverse_slope”: 1 over “slope”; may perform better.

_feature_selection(fit)[source]#

_fit_guess(t, y)[source]#

_inverse_fit_function(y, b, c)[source]#: Square function to transform data for a linear fit.

_supported_features = ['offset', 'slope', 'inverse_slope']#

pybop.costs.feature_distances.indices_of(values, target)[source]#