stochastic_universal_sampling

pybrops.core.random.sampling.stochastic_universal_sampling(a, p, size=None, rng=None)

Perform stochastic universal sampling.

Parameters:

• a (numpy.ndarray) – Array of shape (n,) containing element from which to sample.

• p (numpy.ndarray) –

  Contribution probability array of shape (n,).

  Where:

  • n is the number of elements.

  Restrictions:

  • Values are restricted to \([0,\infty]\).

  • Sum of values in vector must be \(>0.0\).

• size (Integral, Tuple[Integral,...]) – Target output size.

• rng (Generator, RandomState, None) – Random number source. If None, use default rng.

Returns:

out – Sampling of items from a.

Return type:

numpy.ndarray