Google has introduced a preview version of TensorFlow 2.0. Earlier this year, Google indicated that 2.0 was on its way. This week, at the TensorFlow Dev Summit 2019, the preview version was made public.
"TensorFlow 2.0 focuses on simplicity and ease of use, with updates like eager execution, intuitive higher-level APIs, and flexible model building on any platform," the TensorFlow team commented in its introduction to 2.0.
To help beginners, Keras will act as the high-level API for building and training deep learning models. Further, Google has deprecated redundant APIs and added new APIs for new features. The out of the box performance of TensorFlow 2.0 is better which makes its toolsets more accessible for researchers and new users.
The beginner notebook tutorial was drafted specifically for developers new to machine learning. Those accustomed with TensorFlow can check out all of the changes and new features in the Effective TensorFlow 2.0 guide. Links to videos, blog posts, and helpful docs can all be found at the alpha preview site.