RandSimplexGridUniform

class poisson_approval.RandSimplexGridUniform(cls, denominator, keys, d_key_fixed_share=None, **kwargs)[source]

A random factory of an object defined by shares on a discrete simplex.

Parameters:
  • cls (class) – The class of object we want to create. It must accept as parameter a dictionary of the form key: share, where share is a number.
  • denominator (int) – The grain of the grid.
  • keys (iterable) – These keys will have a variable share.
  • d_key_fixed_share (dict) – A dictionary. For each entry key: fixed_share, this key will have at least this fixed share. The total must be lower or equal to 1.
  • kwargs – Additional parameters are passed to cls when creating the object.

Examples

Basic usage:

>>> initialize_random_seeds()
>>> rand_dict = RandSimplexGridUniform(cls=DictPrintingInOrder, denominator=7, keys=['a', 'b'])
>>> rand_dict()
{'a': Fraction(6, 7), 'b': Fraction(1, 7)}

If d_key_fixed_share is given, then these shares are fixed, and the remaining share is split between keys:

>>> from fractions import Fraction
>>> initialize_random_seeds()
>>> rand_dict = RandSimplexGridUniform(cls=DictPrintingInOrder, denominator=3, keys=['a', 'b'],
...                                    d_key_fixed_share={'c': Fraction(4, 7)})
>>> rand_dict()
{'a': Fraction(2, 7), 'b': Fraction(1, 7), 'c': Fraction(4, 7)}

The keys in d_fixed_share may overlap with keys:

>>> initialize_random_seeds()
>>> rand_dict = RandSimplexGridUniform(cls=DictPrintingInOrder, denominator=3, keys=['a', 'b'],
...                                    d_key_fixed_share={'b': Fraction(4, 7)})
>>> rand_dict()
{'a': Fraction(2, 7), 'b': Fraction(5, 7)}

If you want the created object to meet a particular condition, use RandConditional.