npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2024 – Pkg Stats / Ryan Hefner

@webonnx/wonnx-wasm

v0.5.1

Published

Wonnx is an ONNX runtime based on wgpu aimed at being a universal GPU runtime, written in Rust.

Downloads

46

Readme

WONNX WebAssembly package

This crate allows using WONNX on the Web through WebAssembly (WASM).

How it works

The wonnx create is compiled to a WASM module which a browser can run at near-native speed. The module also comes with a thin binding layer in JavaScript, which allows JS code to call public functions in the WASM module, and allows the WASM module to call certain functions in the JS code. To perform inference, the JS code calls into the WASM module to load an ONNX model and inference inputs into memory. WONNX will then compile the model into shader code, just as it would when runing natively. However, in this case WONNX will use the WebGPU API in the browser (which it has access to through the binding layer) to perform the heavy lifting.

Usage

The wonnx-wasm package provides a simple interface using which you can perform inference using ONNX models, mostly resembling the API of the native version. The API looks like this:

import init, { Session, Input } from "@webonnx/wonnx-wasm";

async function run() {
	await init();

	try {
		const session = await Session.fromBytes(modelBytes /* Uint8Array containing the ONNX file */);
		const input = new Input();
		input.insert("x", [13.0, -37.0]);
		const result = await session.run(input); // This will be an object where the keys are the names of the model outputs and the values are arrays of numbers.
		session.free();
		input.free();
	}
	catch(e) {
		console.error(e.toString()); // The error will be of type SessionError
	}
}

run();

For a basic example, see index.html, or see squeeze.html for an example of how to use images.

Building

From the root of the repository, run the following commands:

# Install the wasm-pack build tool
cargo install wasm-pack

# Build release version. The RUSTFLAGS are needed because WebGPU APIs are still unstable in web_sys
RUSTFLAGS=--cfg=web_sys_unstable_apis wasm-pack build --target web -d `pwd`/target/pkg --out-name wonnx --scope webonnx ./wonnx-wasm

# Add --dev if you want a debug build
RUSTFLAGS=--cfg=web_sys_unstable_apis wasm-pack build --target web -d `pwd`/target/pkg --out-name wonnx --scope webonnx ./wonnx-wasm --dev

To test the freshly built package, run python3 -m http.server 8080 and open http://localhost:8080/wonnx-wasm/ in a web browser.

Note that you will have to use a browser that supports WebGPU (and has the feature enabled). For Chrome, install Chrome Canary and enable the Unsafe WebGPU flag by navigating to chrome://flags/#enable-unsafe-webgpu. For Firefox, enable the dom.webgpu.enabled setting by navigating to about:config.