Machine learning tools from Google
summary
TensorFlow updates focus on concrete tools and numbers: TF 2.4 adds MultiWorkerMirroredStrategy and mixed precision for faster GPU and TPU training, TF 2.6 ships Keras preprocessing layers and splits Keras into a separate pip package, and the project celebrates six years. TF Hub hosts hundreds of ready models (spam detector, bird classifier) with a zero‑install browser tester, and TensorFlow.js runs and trains models everywhere. New MoveNet multi-person detection tracks multiple people and preserves identity across swaps, enabling client‑side deployment and privacy preserving applications.
AI-written summary · about
chapters (7)
- 0:00Start
- 0:25Machine Learning at Google
- 1:13Tensorflow
- 1:40Applied Ml
- 4:15Tensorflow Hub
- 5:11Tensorflow Js
- 6:11Movenet Single Person Detection
related talks
33mWhat's new in Machine Learning
26mMachine Learning with TensorFlow
13mTaking Machine Learning to the Next Level
54mAdventures in Javascript Machine Learning
15mMachine learning for next gen web apps with TensorFlow.js
28mMachine Learning on the web with TensorFlow.js
09mInterview with Martin Görner
20mTensorFlow.js Bringing Machine Learning to the Web and Beyond