stormm::random::RandomNumberMill< T > Class Template Reference

An series of "Xorshiro128+" generators, with state vectors for all of them and the means for seeding the series based on long jumps from a single state vector. More...

#include <random.h>

Public Member Functions
size_t	getGeneratorCount () const
	Get the number of (forward-jumped) generators.
size_t	getDepth () const
	Get the depth of the bank for each generator.
size_t	getRefreshStride () const
	Get the number of generators for which to refesh all banked values at one time.
T	getBankValue (size_t generator_index, size_t layer_index) const
	Get a random number out of the bank.
void	uniformRandomNumbers (size_t first_gen, size_t last_gen)
	Populate a portion of this object's bank with random numbers from each of the respective generators.
void	gaussianRandomNumbers (size_t first_gen, size_t last_gen)
	Populate a portion of this object's bank with normally distributed random numbers from each of the respective generators.
	RandomNumberMill (const ullint2 state_in, size_t generators_in=1LLU, size_t depth_in=2LLU, RandomNumberKind init_kind=RandomNumberKind::GAUSSIAN, size_t bank_limit=constants::giga_zu)
	The constructor can work off of a simple random initialization seed integer or a specific state vector for the first generator in the series.
	RandomNumberMill (const ullint4 state_in, size_t generators_in=1LLU, size_t depth_in=2LLU, RandomNumberKind init_kind=RandomNumberKind::GAUSSIAN, size_t bank_limit=constants::giga_zu)
	RandomNumberMill (size_t generators_in=1LLU, size_t depth_in=2LLU, RandomAlgorithm style_in=RandomAlgorithm::XOSHIRO_256PP, RandomNumberKind init_kind=RandomNumberKind::GAUSSIAN, int igseed_in=default_random_seed, int niter=default_xoshiro256pp_scrub, size_t bank_limit=constants::giga_zu)

Detailed Description

template<typename T>
class stormm::random::RandomNumberMill< T >

An series of "Xorshiro128+" generators, with state vectors for all of them and the means for seeding the series based on long jumps from a single state vector.

Constructor & Destructor Documentation

RandomNumberMill()

template<typename T>

stormm::random::RandomNumberMill< T >::RandomNumberMill	(	const ullint2	state_in,
		size_t	generators_in = 1LLU,
		size_t	depth_in = 2LLU,
		RandomNumberKind	init_kind = RandomNumberKind::GAUSSIAN,
		size_t	bank_limit = constants::giga_zu )

The constructor can work off of a simple random initialization seed integer or a specific state vector for the first generator in the series.

Parameters

generators_in	The count of generators in the series
depth_in	Quantity of random numbers from each generator to store in the bank
init_kind	Style in which to initialize the random numbers in the bank, i.e. from a uniform or a normal distribution
igseed_in	The seed for the first generator in the series
niter	The number of iterations to use in initializing each generator
bank_limit	The maximum length of the random number cache
state_in	The state to apply to generator zero, thus determining the initial states of all other generators

Member Function Documentation

gaussianRandomNumbers()

template<typename T>

void stormm::random::RandomNumberMill< T >::gaussianRandomNumbers	(	size_t	first_gen,
		size_t	last_gen )

Populate a portion of this object's bank with normally distributed random numbers from each of the respective generators.

Parameters

first_gen	Index of the first generator to draw random numbers from
last_gen	Index of the generator before which to stop drawing new random numbers

getBankValue()

template<typename T>

T stormm::random::RandomNumberMill< T >::getBankValue	(	size_t	generator_index,
		size_t	layer_index ) const

Get a random number out of the bank.

Parameters

generator_index	The generator series from which to obtain the value
layer_index	Layer from which to obtain the value

uniformRandomNumbers()

template<typename T>

void stormm::random::RandomNumberMill< T >::uniformRandomNumbers	(	size_t	first_gen,
		size_t	last_gen )

Populate a portion of this object's bank with random numbers from each of the respective generators.

Parameters

first_gen	Index of the first generator to draw random numbers from
last_gen	Index of the generator before which to stop drawing new random numbers

---

The documentation for this class was generated from the following file:

• src/Random/random.h