pandemia.random_tools

Wrapper around python’s PRNG to ease the process of performing deterministic re-runs

Module Contents

Classes

Random

Wraps the python random classes as an abstraction layer over them,

Attributes

log
Probability
T

pandemia.random_tools.log

pandemia.random_tools.Probability

pandemia.random_tools.T

class pandemia.random_tools.Random(seed=None)

Wraps the python random classes as an abstraction layer over them, and offers a number of convenience methods

gammavariate(alpha: float, beta: float) → Probability

Sample gamma distributed random variable with pdf given by

x ** (alpha - 1) * math.exp(-x / beta)

pdf(x) = ————————————–.

math.gamma(alpha) * beta ** alpha

That is, a gamma random variable with shape parameter alpha and scale parameter beta.

expovariate(lambd: float) → Probability

Sample exponentially distributed random variable with mean 1 / lambd.

random_randrange(stop: int) → int

Random randrange function

binomial(size: int, prob: float) → int

Random binomial function

randrange_interval(start: int, stop: int) → int

Random randrange function

random_choice(sequence: Sequence[T]) → T

Random choice function

random_choices(population: Sequence[T], weights: Sequence[int], sample_size: int) → list[T]

Random choices function

random_sample(population: Sequence[T], k: int) → list[T]

Select k items from the population given.

random_shuffle(x: MutableSequence[Any]) → None

Random shuffle function

random_float(x: Probability) → float

Return random number between 0 and x

multinoulli(problist: Sequence[Probability]) → int

Sample at random from a list of n options with given probabilities.

Identical to ‘roulette wheel’ random selection.

problist: a list of n items, each of which is a weight.

Returns: The index number of the item chosen

multinoulli_dict(problist_dict: dict[T, Probability]) → T

Sample from a key:value dict and return a key according to the weights in the values, i.e.:

{‘a’: 4, ‘b’: 6} has a 60% chance of returning ‘b’ and a 40% chance of returning ‘a’.

boolean(probability_true: Probability) → bool

Return true with the probability given.