Machine Learning on the web with TensorFlow.js
summary
A live AMA that explains what TensorFlow.js actually does and how it is used: runs models in-browser and on Node.js, converts Keras/TensorFlow and TF Lite models, and adds TF Lite + Task API support; demonstrates real builds like MoveNet for physiotherapy and ModiFace virtual try‑on that work in production; reports concrete speedups (many ops ~2×, Android WebGL ~100% faster) and discusses backends WebGL and upcoming WebGPU, plus quick training via Teachable Machine.
AI-written summary · about
chapters (14)
- 0:00Start
- 2:37How Easy It Is It To Use and Do I Need To Learn Python
- 3:34Are There Ready To Use Pre-Trained Models and Where Can I Find Them
- 4:05Pre-Trained Models
- 7:19How Will the Machine Hardware Impact the Working and Training of Models in the Browser
- 9:43Is the Python Model Really Compulsory in Tensorflow Js
- 10:46Tensorflow Light Support for Tensorflow Js Does It Allow You To Run Models with Ops Previously Not Supported by Tensorflow Js Backends
- 12:29Possible To Extract Metadata from the Tf Lite Models Using the Tfgs Task Api
- 15:24I Am Planning To Develop a Web App Using Flutter Web but Wish To Use Tensorflow with It Too Is There an Interface That Allows Tensorflow Js and Dart To Talk to each Other
- 16:19How Does the Tensorflow Team Improve Debuggability on the Webassembly
- 21:13Optimization
- 22:28Resources
- 25:48How Can I Contribute to or Join the Tensorflow Js Community
- 26:26Special Interest Group for Tensorflow Js
related talks
31mIntroduction to Machine Learning in Tensorflow.js
20mTensorFlow.js Bringing Machine Learning to the Web and Beyond
37mTensorFlow.js - Bringing ML and Linear Algebra to Node.js
41mTensorFlow.js 101: ML in the Browser and Beyond
21mA Journey into the World of Machine Learning with TensorFlow.js
20mKeynote: Building Great Web Experiences with AMP and TensorFlow.js
46mMachine learning in the browser using TensorFlow.js
36mML at the Heart of JS