Easily deploy TF Lite models to the web
summary
Claims that TensorFlow Lite models run in-browser with no conversion by loading the TensorFlow Lite task library via TensorFlow.js and WebAssembly. It shows using script tags to load the tfjs tflite wasm APIs, then loading model URLs for image classification, segmentation, object detection, and text classification. Examples load a movie-review sentiment model and a cartoon game model, call classify/predict, process output tensors and draw to canvas — demos run live in the browser.
AI-written summary · about
related talks
19mML at Client Side
13mLiteRT.js, Google’s high performance WebAI runtime
20mTensorFlow.js Bringing Machine Learning to the Web and Beyond
48mTensorFlow.js
24mThe Future of Machine Learning & Javascript
46mMachine learning in the browser using TensorFlow.js
31mIntroduction to Machine Learning in Tensorflow.js
37mTensorFlow.js - Bringing ML and Linear Algebra to Node.js