Deep Learning with Python: A Hands-on Introduction

© Nikhil Ketkar 2017

Nikhil Ketkar, Deep Learning with Python, https://doi.org/10.1007/978-1-4842-2766-4_11

11. Introduction to Tensorflow

Nikhil Ketkar¹ 

(1)Bangalore, Karnataka, India

In this chapter we will cover Tensorflow which allows users to define mathematical functions via computational graphs and to compute their gradients. Tensorflow is conceptually similar to Theano, and Keras uses both of them as back ends.

Tensorflow supports training of large models in a multiple Graphical Processing Unit (GPU) , distributed setting and while the project is not as mature as Theano , it has strong backing from industry leaders such as Google and hence should be in the toolkit of anyone working on deep learning. Another important characteristic of Tensorflow ...