Google announced Wednesday more information on TensorFlow, the open source library that has already downloaded more than 41 millions of times. To make it even more accessible, Google released the alpha version of TensorFlow 2.0, which Google says is simpler and more intuitive to use.
With TensorFlow 2.0, Google makes APIs better integrated with tf.keras, the high-end API for most users. This will help developers move more easily from data acquisition to conversion, building new models, training, and ultimately developing.
Meanwhile, Google adds also new add-ons, such as TensorFlow Federated. The open source framework allows experimentation with mechanical learning and other calculations in decentralized data, or where the data is generated.
For those who do not know, TensorFlow Privacy is an open source library that enables developers to train their engineering learning models using techniques based on the privacy theory of the different.
This provides a guarantee that learning engine models developed will not learn or remember specific details about specific users.