benchmarking.utilities namespace

Submodules

benchmarking.utilities.apply_noise module

benchmarking.utilities.apply_noise.apply_noise(value: float, sigma: float, mean: float = 0, variance: float = 1) float[source]

Apply noise to the given function values.

Applies noise of the form \(\sigma * N\) to the function value as

\[f_bar(x) = f(x) + \sigma * N\]

where N is a normally distributed random variable with (by default) mean = 0 and variance = 1. \(\sigma\) refers to the degree of perturbation of the value with \(\sigma = 0\) representing the unperturbed problem.

benchmarking.utilities.generate_box module