number of elements
Optionallow: number
lower bound (default: -1)
Optionalhigh: number
upper bound (default: 1)
Optionalseed: number
omit for genuine randomness (Math.random, the default);
pass any number for a deterministic, reproducible sequence (mulberry32) —
useful for regression tests that need "random-looking" data without
flaking between runs
Default range [-1, 1):
import { randomFloat32Array } from "wgblas";
const x = randomFloat32Array(4);
console.log(x); // Float32Array [ -0.42, 0.81, -0.07, 0.55 ] (values differ each run)
Custom range [0, 10):
import { randomFloat32Array } from "wgblas";
const x = randomFloat32Array(4, 0, 10);
console.log(x);
Seeded (deterministic):
import { randomFloat32Array } from "wgblas";
// Same seed -> identical output every run, useful for reproducible
// regression tests that need "random-looking" data without real randomness.
const a = randomFloat32Array(4, -1, 1, 42);
const b = randomFloat32Array(4, -1, 1, 42);
console.log(a);
console.log(b);
console.log(
"identical:",
a.every((v, i) => v === b[i]),
);
Returns a Float32Array of n random values uniformly sampled from [low, high).