@thi.ng/random

What is @thi.ng/random?

@thi.ng/random is a versatile library for generating random numbers and sequences. It provides various random number generators (RNGs) and utilities for creating random values, sequences, and distributions.

What are @thi.ng/random's main functionalities?

Basic Random Number Generation

This feature allows you to generate basic random numbers, both floating-point and integer values, using the default system RNG.

const { SYSTEM } = require('@thi.ng/random');
const rng = SYSTEM;
console.log(rng.float()); // Generates a random float between 0 and 1
console.log(rng.int()); // Generates a random integer

Seeded Random Number Generation

This feature allows you to create a seeded random number generator, which produces reproducible sequences of random numbers.

const { XsAdd } = require('@thi.ng/random');
const rng = new XsAdd(12345); // Seeded RNG
console.log(rng.float()); // Generates a reproducible random float
console.log(rng.int()); // Generates a reproducible random integer

Random Value Generation from Arrays

This feature allows you to randomly select values from an array, which is useful for shuffling or sampling data.

const { SYSTEM } = require('@thi.ng/random');
const rng = SYSTEM;
const arr = [1, 2, 3, 4, 5];
console.log(rng.choice(arr)); // Randomly selects an element from the array

Random Distributions

This feature allows you to generate random numbers from various statistical distributions, such as normal (Gaussian) distributions.

const { SYSTEM } = require('@thi.ng/random');
const rng = SYSTEM;
console.log(rng.norm(0, 1)); // Generates a random number from a normal distribution with mean 0 and standard deviation 1

Other packages similar to @thi.ng/random

random-js

random-js is a library for generating random numbers and values. It provides a variety of RNGs and utilities for creating random values, similar to @thi.ng/random. However, random-js focuses more on providing a simple API for common randomization tasks.

seedrandom

seedrandom is a library for seeded random number generation. It allows you to create reproducible random sequences by providing a seed. While it offers fewer utilities compared to @thi.ng/random, it is highly focused on the quality and reproducibility of random numbers.

chance

chance is a library for generating random data for various purposes, including numbers, strings, and even entire objects. It offers a broader range of random data generation compared to @thi.ng/random, which is more focused on numerical randomization.

README

This project is part of the @thi.ng/umbrella monorepo.

• About
  • Status
• Installation
• Dependencies
• Usage examples
• API
  • Other utilities
• Authors
• License

About

Pseudo-random number generators w/ unified API.

This package provides the IRandom interface and various (mostly seedable) pseudo-random number generator implementations, incl. a IRandom wrapper for Math.random():

• Smush32
• System
• Xoshiro128
• XorShift128
• XorWow
• XsAdd

Partially ported from C implementations taken from c.thi.ng.

Status

STABLE - used in production

Installation

yarn add @thi.ng/random

// ES module
<script type="module" src="https://unpkg.com/@thi.ng/random?module" crossorigin></script>

// UMD
<script src="https://unpkg.com/@thi.ng/random/lib/index.umd.js" crossorigin></script>

Package sizes (gzipped, pre-treeshake): ESM: 1.23 KB / CJS: 1.33 KB / UMD: 1.37 KB

Dependencies

• @thi.ng/api
• @thi.ng/checks

Usage examples

Several demos in this repo's /examples directory are using this package.

A selection:

Screenshot	Description	Live demo	Source
	Interactive inverse FFT toy synth	Demo	Source
	Evolutionary shader generation using genetic programming	Demo	Source

API

Generated API docs

import { Smush32 } from "@thi.ng/random";

const rnd = new Smush32(0xdecafbad);

// the following methods are available for all generators

// next uint (0 .. 2^32-1)
rnd.int()
// 4022849029

// next float [0.0 .. 1.0)
rnd.float()
// 0.2698542904737066

// next normalized float (w/ opt scale)
// [-scale .. +scale)
rnd.norm(100)
// 57.70723665079737

// next float in given interval [min .. max)
rnd.minmax(10, 20)
// 15.295951807707537

// next gaussian (using iterative CLT approach)
// optional params: num samples, offset, scale
rnd.gaussian()
// 0.10632886109089679

Other utilities

• randomBytes
• randomID

Authors

Karsten Schmidt

License

© 2016 - 2020 Karsten Schmidt // Apache Software License 2.0