Google has announced TensorFlow 1.0 at part of the first annual TensorFlow Developer Summit. The summit held in Mountain View, with the launch. It is used by users in over 6000 open-source repositories online.
Here are some of the features,
- Experimental APIs now provided for Java and Go.
- It also comes with a high-level API, with tf.layers, tf.metrics, and tf.losses modules. A new tf.keras module has been included and is compatible with another popular high-level neural networks library, Keras.
- The new version is amazingly fast.
- It’s debugger (tfdbg) introduced, which is a command-line interface and API. It is used for debugging it’s live programs.
- Python 3 docker images added, in the new version.
Google states in its official Google Developers Blog,
The TensorFlow ecosystem continues to grow with new techniques like Fold for dynamic batching and tools like the Embedding Projector along with updates to our existing tools like TensorFlow Serving. We’re incredibly grateful to the community of contributors, educators, and researchers who have made advances in deep learning available to everyone