A fast random number generator

Algorithm Library .NET Random Statistics

Go to file

Thorsten Sommer 0d1d48db3b Spelling		2024-05-29 11:32:25 +02:00
.idea/.idea.FastRng/.idea	Update	2023-07-06 09:49:29 +02:00
documents	Added distributions	2020-11-01 12:30:30 +01:00
FastRng	Increment version	2023-11-14 15:43:11 +01:00
FastRngTests	Added unit test to cover issue #8	2023-11-14 13:57:37 +01:00
images	Added distributions	2020-11-01 12:25:41 +01:00
.gitattributes	Enabled Git LFS	2020-11-01 11:40:26 +01:00
.gitignore	Initial commit	2020-09-23 19:59:56 +02:00
FastRng.sln	Added unit test project	2020-09-24 20:20:46 +02:00
LICENSE	Initial commit	2020-09-23 19:59:56 +02:00
README.md	Spelling	2024-05-29 11:32:25 +02:00

README.md

FastRng

FastRng is a multithreaded pseudo-random number generator. Besides the generation of uniformly distributed random numbers, there are several other distributions to choose from. For performance reasons, the parameters of the distributions are not user-definable. For some distributions, therefore, different parameter variations are available. If a different combination is desired, a separate class can be created.

Please note, that Math.NET's (https://www.mathdotnet.com/) random number generator is in some situations faster. Unlike Math.NET, MultiThreadedRng is multithreaded. Consumers can use a token to cancel an operation.

FastRng (class MultiThreadedRng) using a shape fitter (a rejection sampler) to enforce arbitrary shapes of probabilities for desired distributions. By using the shape fitter, it is even easier to define discontinuous, arbitrary functions as shapes. Any consumer can define and use own distributions.

The class MultiThreadedRng uses the George Marsaglia's MWC algorithm. The algorithm's implementation based loosely on John D. Cook's (johndcook.com) implementation. Thanks, John, for the inspiration.

Usage

Example code:

using FastRng.Float;
using FastRng.Float.Distributions;

[...]

using var rng = new MultiThreadedRng<float>();
var dist = new ChiSquareK1<float>(rng);

var value1 = dist.NextNumber();
var value2 = dist.NextNumber(rangeStart: -1.0f, rangeEnd: 1.0f);
if(dist.HasDecisionBeenMade(above: 0.8f, below: 0.9f))
{
    // Decision has been made
}

Notes:

MultiThreadedRng and all distributions are using generic math types: you might use float, double, or Half.
MultiThreadedRng is IDisposable. It is important to call Dispose, when the generator is not needed anymore. Otherwise, the supporting background threads are still running.
MultiThreadedRng fills some buffers after creation. Thus, create and reuse it as long as needed. Avoid useless re-creation.
Distributions need some time in creation to calculate probabilities. Thus, create a distribution once and use reuse it. Avoid useless re-creation.

Available Distributions

README.md

FastRng

Usage

Available Distributions

Normal Distribution (std. dev.=0.2, mean=0.5)

Beta Distribution (alpha=2, beta=2)

Beta Distribution (alpha=2, beta=5)

Beta Distribution (alpha=5, beta=2)

Cauchy / Lorentz Distribution (x0=0)

Cauchy / Lorentz Distribution (x0=1)

Chi-Square Distribution (k=1)

Chi-Square Distribution (k=4)

Chi-Square Distribution (k=10)

Exponential Distribution (lambda=5)

Exponential Distribution (lambda=10)

Inverse Exponential Distribution (lambda=5)

Inverse Exponential Distribution (lambda=10)

Gamma Distribution (alpha=5, beta=15)

Inverse Gamma Distribution (alpha=3, beta=0.5)

Laplace Distribution (b=0.1, mu=0)

Laplace Distribution (b=0.1, mu=0.5)

Log-Normal Distribution (sigma=1, mu=0)

StudentT Distribution (nu=1)

Weibull Distribution (k=0.5, lambda=1)