Real-world Machine Learning with TensorFlow and Cloud AI
summary
Shows concrete pipelines using TensorFlow and Cloud AutoML/AutoML Vision: upload ~50,000 labeled images, train in ~12 hours for ≈$300, export a model to run on Raspberry Pi or mobile GPUs, and accelerate inference with an HTTP accelerator giving up to 42× or 13.8× speedups; one retailer gets 95% accuracy distinguishing 45 ramen branches, an insurer sees 20–30% better fraud detection on a $680M spend, and mobile inference reaches ~15 ms per image.
AI-written summary · about
chapters (7)
- 0:00Start
- 1:00Introduction
- 9:10Neural Network Algorithm
- 26:04Automate Tables
- 31:56Smart Mirror
- 33:20Use Cases
- 36:30Use Case of the Tensorflow by a Dry Cleaning Shop Owner
related talks
51mCloud AI: How to Get Started Injecting AI Into Your Applications
47mUnlocking Data-Driven Apps with TensorFlow
13mTaking Machine Learning to the Next Level
30mTensorFlow for Mobile Heroes
59mML and the IoT: Living on the Edge
42mWhat's New with TensorFlow?
27mMachine Learning on the edge using TensorFlow Lite
57mBuilding smarter solutions with Machine Learning, from magic to reality