Skip to content

WasmGPU.compute.CPUndarray.fromArray

Summary

WasmGPU.compute.CPUndarray.fromArray creates a contiguous CPU ndarray from JavaScript array-like data. It allocates by shape and copies source values into wasm memory. Use this for quick conversion of host arrays/typed arrays into WasmGPU ndarray format. Source length must be at least the ndarray element count.

Syntax

WasmGPU.compute.CPUndarray.fromArray<T extends ArrayLike<number>>(dtype: DType, shape: ReadonlyArray<number>, src: T): CPUndarray
const a = wgpu.compute.CPUndarray.fromArray(dtype, shape, src);

Parameters

Name Type Required Description
dtype DType Yes Destination element data type.
shape ReadonlyArray<number> Yes Output ndarray dimensions.
src ArrayLike<number> Yes Source numeric data copied into the new ndarray.

Returns

CPUndarray - Contiguous CPU ndarray populated from source data.

Example

const canvas = document.querySelector("canvas");
const wgpu = await WasmGPU.create(canvas);

const src = new Float32Array([1, 2, 3, 4, 5, 6]);
const a = wgpu.compute.CPUndarray.fromArray("f32", [2, 3], src);
console.log(a.get(1, 2));

See Also