Random number generator

class stats_arrays.RandomNumberGenerator(uncertainty_type, params, size=1, maximum_iterations=100, seed=None, **kwargs)

__init__(uncertainty_type, params, size=1, maximum_iterations=100, seed=None, **kwargs)

Create a random number generator from a Parameter array and an uncertainty distribution.

Upon instantiation, the class checks that:

• The minimum and maximum bounds, if any, are reasonable
• The given uncertainty type can be used

uncertainty_type is not required to be a subclass of UncertaintyBase, but needs to have the method bounded_random_variables.

The returned class instance can be called directly:

>>> from stats_arrays import RandomNumberGenerator, TriangularUncertainty
>>> params = TriangularUncertainty.from_dicts(
...     {'loc': 5, 'minimum': 3, 'maximum': 10},
...     {'loc': 1, 'minimum': 0.7, 'maximum': 4.4}
...     )
>>> rng = RandomNumberGenerator(TriangularUncertainty, params)
>>> rng.generate_random_numbers()
array([[ 8.00843856],
   [ 1.54968237]])

but can also be used as an iterator:

>>> zip(range(2), rng)
[(0, array([[ 5.34298156],
   [ 1.02447677]])),
 (1, array([[ 5.45360508],
   [ 1.99372889]]))]

Args:

• uncertainty_type (object): An uncertainty type class (subclass of stats_arrays.distributions.UncertaintyBase)
• params (array): The Parameter array
• size (int, optional): The number of samples to draw from each parameter. Default is 1.
• maximum_iterations (int, optional): The number of times to draw samples that fit within the given bounds, if any, before raising stats_arrays.MaximumIterationsError. Default is 100.
• seed (int, optional): Seed value for the random number generator. Default is None.

Returns:

A class instance

verify_params(params=None, uncertainty_type=None)

Verify that parameters are within bounds. Mean is not restricted to bounds, unless the distribution requires it (e.g. triangular).

verify_uncertainty_type(uncertainty_type=None)

Make sure the given uncertainty type provides the method bounded_random_variables.