UQTestFun

Module containing the concrete implementation of UQ test functions.

This module provides the UQTestFun class, which encapsulates an evaluation function, a probabilistic input model, and optional parameters into a single callable object for uncertainty quantification test functions.

class uqtestfuns.core.uqtestfun.UQTestFun(evaluate: Callable, prob_input: ProbInput, parameters: Parameters | None = None, name: str | None = None, description: str | None = None, output_dimension: int = 1)

A concrete class for UQ test functions.

Encapsulates an evaluation function, a probabilistic input model, optional parameters, and metadata into a single callable object.

Parameters:

• evaluate (Callable) – The evaluation function. Must accept a 2D numpy array of shape (n_samples, input_dimension) and return an array of shape (n_samples,) or (n_samples, output_dimension). If parameters are provided, they are passed as keyword arguments.

• prob_input (ProbInput) – The probabilistic input model defining the input space and the distribution of the input variables.

• parameters (Parameters, optional) – The parameter set of the UQ test function. If provided, passed as keyword arguments to evaluate. Default is None.

• name (str, optional) – The name of the UQ test function. Default is None.

• description (str, optional) – A human-readable description of the UQ test function. Default is None.

• output_dimension (int, optional) – The number of output values produced per input point. Default is 1.

Raises:

TypeError – If any argument has an incorrect type.

__init__(evaluate: Callable, prob_input: ProbInput, parameters: Parameters | None = None, name: str | None = None, description: str | None = None, output_dimension: int = 1)

property description: str | None

Get the description of the UQ test function.

Returns:

The description of the test function, or None if not set.

Return type:

Optional[str]

get_sample(sample_size: int = 1, rng: Generator | int | None = None) → ndarray

Generate a sample of output values from the UQ test function.

Generates random input points from the probabilistic input model and evaluates the test function at those points. Unlike prob_input.get_sample(), which returns input samples, this method returns the corresponding function output values.

Parameters:

• sample_size (int, optional) – The number of sample points to generate. Default is 1.

• rng (Union[np.random.Generator, int, None], optional) – Random number generator or seed for reproducibility. Accepts a np.random.Generator, an integer seed, or None for the default generator. Default is None.

Returns:

Output array of shape (sample_size,) if output_dimension is 1, or (sample_size, output_dimension) otherwise.

Return type:

np.ndarray

property input_dimension: int

Get the input dimension of the UQ test function.

Returns:

The dimension of the input space, determined by the probabilistic input model.

Return type:

int

property name: str | None

Get the name of the UQ test function.

Returns:

The name of the test function, or None if not set.

Return type:

Optional[str]

property output_dimension: int

Get the output dimension of the UQ test function.

Returns:

The dimension of the output space, representing the number of output values produced by the test function for each input.

Return type:

int

property parameters: Parameters | None

Get the parameters of the UQ test function.

Returns:

The parameters of the test function, or None if not set.

Return type:

Optional[Parameters]

property prob_input: ProbInput

Get the probabilistic input model of the UQ test function.

Returns:

The probabilistic input model that defines the input space of the test function.

Return type:

ProbInput