Getting Started with TensorFlow 2.0

on YouTube ↗32m

" summary

TensorFlow 2.0 alpha focuses on ease of use and reproducible examples that run in Colab runnable notebooks with one-click installs; many demos are under 50 lines (eg. an English→Spanish sequence‑to‑sequence translator and image captioning). Keras is the default API with Keras Sequential, Functional, Subclassing styles and eager execution by default; automatic graph acceleration via AutoGraph is optional. Migration tools include tf_upgrade_v2 and tf.compat.v1; SavedModel exports target TF Serving, Lite, JS and mobile. Reported gains: ~1.8× training speed on NVIDIA Teslas and 2 ms inference on Edge TPU for quantized models.

AI-written summary · about

" chapters (7)

1. 0:00Intro
2. 1:15Deep Learning
3. 7:10User Experience
4. 9:58Karos API
5. 15:23Documentation
6. 16:15TensorFlow Closure
7. 23:48What is TensorFlow

" related talks

• 28mTensorFlow 1.0: open source machine learningAndrew Selle · FOSSASIA Summit 2017
• 19mWhat's new in TensorFlow for RDaniel Falbel · rstudio::conf 2019
• 02h 41mHello TensorFlow: 3 Workshops to Get Started with TensorFlow 2.0Alexia Audevart, Philippe Antoine · Devoxx France 2021
• 23mDeep Learning with RPaige Bailey · rstudio::conf 2019
• 47mSwifty TensorFlowSiamak Ashrafi · 360|iDev 2019
• 41mTensorFlow ever higherYoann Benoit · DevFest Paris 2019
• 41mLearning TensorFlowRobert Layton · PyCon AU 2016 Science and Data Miniconf
• 33mWhat's new in Machine LearningKemal El Moujahid, Sarah Sirajuddin, Craig Wiley · Google I/O 2021