Skip to main content
Using Deno in production at your company? Earn free Deno merch.
Give us feedback
Module

x/netsaur/src/core/engine.ts>TensorBackend

Powerful machine learning, accelerated by WebGPU
denosaurs/netsaur
Go to Latest
interface TensorBackend
import { type TensorBackend } from "https://deno.land/x/netsaur@0.2.4/src/core/engine.ts";

The Tensor Backend is responsible for creating and managing Tensors.

Methods

zeroes<R extends Rank, B extends BackendType>(shape: Shape[R]): Tensor<R>
from<R extends Rank, B extends BackendType>(values: Float32Array, shape: Shape[R]): Tensor<R>
get(tensor: Tensor<Rank>): Promise<Float32Array>
set(tensor: Tensor<Rank>, values: Float32Array): void