Folders and files

Name		Name	Last commit message	Last commit date
Latest commit History 6,112 Commits
.github		.github
.tslint		.tslint
.vscode		.vscode
dockers		dockers
docs		docs
e2e		e2e
link-package		link-package
remote-execution		remote-execution
scripts		scripts
tfjs-automl		tfjs-automl
tfjs-backend-cpu		tfjs-backend-cpu
tfjs-backend-nodegl		tfjs-backend-nodegl
tfjs-backend-wasm		tfjs-backend-wasm
tfjs-backend-webgl		tfjs-backend-webgl
tfjs-backend-webgpu		tfjs-backend-webgpu
tfjs-converter		tfjs-converter
tfjs-core		tfjs-core
tfjs-data		tfjs-data
tfjs-inference		tfjs-inference
tfjs-layers		tfjs-layers
tfjs-node-gpu		tfjs-node-gpu
tfjs-node		tfjs-node
tfjs-react-native		tfjs-react-native
tfjs-tfdf		tfjs-tfdf
tfjs-tflite		tfjs-tflite
tfjs-vis		tfjs-vis
tfjs		tfjs
tools		tools
.bazelignore		.bazelignore
.bazelrc		.bazelrc
.bazelversion		.bazelversion
.gitignore		.gitignore
BAZEL_MIGRATION.md		BAZEL_MIGRATION.md
BUILD.bazel		BUILD.bazel
CONTRIBUTING.md		CONTRIBUTING.md
CONTRIBUTING_MISSING_OP.md		CONTRIBUTING_MISSING_OP.md
DEVELOPMENT.md		DEVELOPMENT.md
GALLERY.md		GALLERY.md
LICENSE		LICENSE
README.md		README.md
WORKSPACE		WORKSPACE
cloudbuild-release.yml		cloudbuild-release.yml
cloudbuild-verdaccio.yml		cloudbuild-verdaccio.yml
cloudbuild.yml		cloudbuild.yml
package.json		package.json
pull_request_template.md		pull_request_template.md
rollup.config.helpers.js		rollup.config.helpers.js
tfjs.code-workspace		tfjs.code-workspace
tsconfig.json		tsconfig.json
tsconfig.test.json		tsconfig.test.json
tsconfig_base.json		tsconfig_base.json
tsconfig_ts_library.json		tsconfig_ts_library.json
tsconfig_tslint.json		tsconfig_tslint.json
tslint.json		tslint.json
yarn.lock		yarn.lock

Repository files navigation

TensorFlow.js

TensorFlow.js is an open-source hardware-accelerated JavaScript library fortraining and deploying machine learning models.

Develop ML in the Browser
Use flexible and intuitive APIs to build models from scratch using the low-levelJavaScript linear algebra library or the high-level layers API.

Develop ML in Node.js
Execute native TensorFlow with the same TensorFlow.js API under the Node.jsruntime.

Run Existing models
Use TensorFlow.js model converters to run pre-existing TensorFlow models rightin the browser.

Retrain Existing models
Retrain pre-existing ML models using sensor data connected to the browser orother client-side data.

About this repo

This repository contains the logic and scripts that combineseveral packages.

APIs:

TensorFlow.js Core,a flexible low-level API for neural networks and numerical computation.
TensorFlow.js Layers,a high-level API which implements functionality similar toKeras.
TensorFlow.js Data,a simple API to load and prepare data analogous totf.data.
TensorFlow.js Converter,tools to import a TensorFlow SavedModel to TensorFlow.js
TensorFlow.js Vis,in-browser visualization for TensorFlow.js models
TensorFlow.js AutoML,Set of APIs to load and run models produced byAutoML Edge.

Backends/Platforms:

TensorFlow.js CPU Backend, pure-JS backend for Node.js and the browser.
TensorFlow.js WebGL Backend, WebGL backend for the browser.
TensorFlow.js WASM Backend, WebAssembly backend for the browser.
TensorFlow.js WebGPU, WebGPU backend for the browser.
TensorFlow.js Node, Node.js platform via TensorFlow C++ adapter.
TensorFlow.js React Native, React Native platform via expo-gl adapter.

If you care about bundle size, you can import those packages individually.

If you are looking for Node.js support, check out theTensorFlow.js Node directory.

Examples

Check out ourexamples repositoryand ourtutorials.

Gallery

Be sure to check outthe gallery of all projects related to TensorFlow.js.

Pre-trained models

Be sure to also check out ourmodels repository where we host pre-trained modelson NPM.

Benchmarks

Local benchmark tool. Use this webpage tool to collect the performance related metrics (speed, memory, etc) of TensorFlow.js models and kernelson your local device with CPU, WebGL or WASM backends. You can benchmark custom models by following thisguide.
Multi-device benchmark tool. Use this tool to collect the same performance related metricson a collection of remote devices.

Getting started

There are two main ways to get TensorFlow.js in your JavaScript project:viascript tagsor by installing it fromNPMand using a build tool likeParcel,WebPack, orRollup.

via Script Tag

Add the following code to an HTML file:

<html><head><!-- Load TensorFlow.js --><scriptsrc="https://cdn.jsdelivr.net/npm/@tensorflow/tfjs/dist/tf.min.js"></script><!-- Place your code in the script tag below. You can also use an external .js file --><script>// Notice there is no 'import' statement. 'tf' is available on the index-page// because of the script tag above.// Define a model for linear regression.constmodel=tf.sequential();model.add(tf.layers.dense({units:1,inputShape:[1]}));// Prepare the model for training: Specify the loss and the optimizer.model.compile({loss:'meanSquaredError',optimizer:'sgd'});// Generate some synthetic data for training.constxs=tf.tensor2d([1,2,3,4],[4,1]);constys=tf.tensor2d([1,3,5,7],[4,1]);// Train the model using the data.model.fit(xs,ys).then(()=>{// Use the model to do inference on a data point the model hasn't seen before:// Open the browser devtools to see the outputmodel.predict(tf.tensor2d([5],[1,1])).print();});</script></head><body></body></html>

Open up that HTML file in your browser, and the code should run!

via NPM

Add TensorFlow.js to your project usingyarnornpm.Note: Becausewe use ES2017 syntax (such asimport), this workflow assumes you are using a modern browser or a bundler/transpilerto convert your code to something older browsers understand. See ourexamplesto see how we useParcel to buildour code. However, you are free to use any build tool that you prefer.

import*astffrom'@tensorflow/tfjs';// Define a model for linear regression.constmodel=tf.sequential();model.add(tf.layers.dense({units:1,inputShape:[1]}));// Prepare the model for training: Specify the loss and the optimizer.model.compile({loss:'meanSquaredError',optimizer:'sgd'});// Generate some synthetic data for training.constxs=tf.tensor2d([1,2,3,4],[4,1]);constys=tf.tensor2d([1,3,5,7],[4,1]);// Train the model using the data.model.fit(xs,ys).then(()=>{// Use the model to do inference on a data point the model hasn't seen before:model.predict(tf.tensor2d([5],[1,1])).print();});