TensorFlow Lite - Solution for Running ML On-Device

Explore TensorFlow Lite, Google's lightweight cross-platform solution for mobile and embedded devices, in this 34-minute conference talk from TF World '19. Discover how to enable on-device machine learning inference with low latency, high performance, and a small binary size. Learn about Mobile Bert, TensorFlow Lite Models, Support Library, performance techniques, Op Coverage, Selective Registration, microcontrollers, interpreter, Arduino applications, and person detection. Presented by Pete Warden and Nupur Garg, this talk covers the standard solution at Google and the primary inference framework for all on-device use-cases.

Intro
Slow Motion
Mobile Bert
TensorFlow Lite Models
Support Library
Performance
Techniques
Op Coverage
Selective Registration
microcontrollers
interpreter
Arduino
Applications
Person Detection
Conclusion