Samplers

By default Everest generates perturbations to estimate gradients using sampling code from SciPy. It provides sampling from common distributions such as Gaussian, Uniform and Bernoulli, and some additional methods such as Sobol and Latin hypercube sampling.

The sampling method and options are specified in the sampler subsection of the controls configuration settings. The sampling method is selected using the method keyword. If the method keyword is missing, the backend will select a default method, which samples from a normal distribution.

The sampling methods in the SciPy backend support several options that can be passed using the options keyword. Please consult the online SciPy manual for details for the options to those sampler that can be passed.

Everest supports the possibility to install additional samplers via a plugin mechanism. The correct plugin to use will generally be inferred from the method name. If several plugins have a method named A, pick a specific backend B by putting B/A in the method field.

Example

optimization:
    sampler:
        method: sobol
        options:
            scramble: False