TensorFlow.js is a library for machine learning in JavaScript
Develop ML models in JavaScript, and use ML directly in the browser or in Node.js.
How it works
Run existing models
Use off-the-shelf JavaScript models or convert Python TensorFlow models to run in the browser or under Node.js.
Retrain existing models
Retrain pre-existing ML models using your own data.
Develop ML with JavaScript
Build and train models directly in JavaScript using flexible and intuitive APIs.
Demos
Unify your mobile and web ML deployments by reusing optimized TF Lite models and running in the browser via WebAssembly, no JavaScript rewrite required. Our new TF.js task APIs support a variety of models and backends.
Improve website user experience by training a custom machine learning model with site navigation data to predict next pages, and use an Angular app to prefetch the content and improve site speed.
Get a high level overview of what TensorFlow.js is, how it's currently being used, what's new this year, plans for the future, and how you can get involved with our newly formed special interest and working groups.
MoveNet is a human pose detection architecture designed to detect difficult poses and fast body motions. The model can run in the browser with very little latency, opening the door for a new class of applications and interactive experiences.
