Build your Machine Learning Model on Edge with React Native
summary
Shows a live React Native demo loading ONNX.js with YOLOv5 to draw bounding boxes for image recognition and builds a TensorFlow.js sentiment classifier end to end. Explains concrete optimizations: quantization, pruning, knowledge distillation, and runtime tuning; recommends mobile runtimes like TensorFlow Lite and PyTorch Lite. Cites cloud economics — $490B invested in 2022 with ~30% (~$197B) wasted — and lists bottlenecks, measurements, and the exact code artifacts used for deployment.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 0:58Machine learning on edge devices
- 4:06Machine learning basics
- 8:22Neural networks & image classification
- 10:28Building ML models on device
- 17:58Live image recognition demo
- 18:19Sentiment analysis model
- 21:24ML edge applications
- 22:06ML edge deployment challenges
- 23:31Optimizing ML on edge
- 25:14Resources & closing
related talks
10mMachine Learning And React Native
07mMachine Learning on the Edge
30mBuilding Machine Learning Models on Edge with React
15mSimplified Machine Learning on mobile, browsers, and edge devices with on-device ML
26mSmart Apps: Machine Learning in React Native
09mMachine Learning on the Edge
23mChallenges in Building React Native Applications with On-Device AI
39mMachine Learning on Android Demystified