In 2015, Google made TensorFlow open source, including all of its reference implementation. All of the source code was made available on GitHub under the Apache 2.0 license. Since then, TensorFlow has been widely adopted in academia and industrial research, and the most stable version, 1.6, has recently been released with a unified API.

It is important to note that the APIs in TensorFlow 1.6 (and higher) are not all backward compatible for pre v1.5 code. This means that some programs that worked on pre v1.5 will not necessarily work on TensorFlow 1.6.

Now let us see the new and exciting features that TensorFlow v1.6 has.

You can install and use TensorFlow on a number of platforms such as Linux, macOS, and Windows. Moreover, you can also build and install TensorFlow from the latest GitHub source of TensorFlow. Furthermore, if you have a Windows machine, you can install TensorFlow via native pip or Anacondas. TensorFlow supports Python 3.5.x and 3.6.x on Windows.

In addition, Python 3 comes with the pip3 package manager, which is the program you will use to install TensorFlow. Therefore, you do not need to install pip if you are using this Python version. From our experience, even if you have NVIDIA GPU hardware integrated on your machine, it would be worth installing and trying the CPU-only version first and if you don't experience good performance, you should switch to GPU support then.

The GPU–enabled version of TensorFlow has several requirements such as 64–bit Linux, Python 2.7 (or 3.3+ for Python 3), NVIDIA CUDA® 7.5 or higher (CUDA 8.0 required...

When thinking of executing a TensorFlow program, we should be familiar with the concepts of graph creation and session execution. Basically, the first one is for building the model, and the second one is for feeding the data in and getting the results.

Interestingly, TensorFlow does everything on the C++ engine, which means not even a little multiplication or addition is executed in Python. Python is just a wrapper. Fundamentally, the TensorFlow C++ engine consists of the following two things:

The TensorFlow lib is an extraordinary lib in terms of coding and it is not like conventional Python code (for example, you can write statements and they get executed). TensorFlow code consists of different operations. Even variable initialization is special in TensorFlow. When you are performing a complex operation with TensorFlow...

The TensorFlow programming model signifies how to structure your predictive models. A TensorFlow program is generally divided into four phases when you have imported the TensorFlow library:

These main phases define the programming model in TensorFlow. Consider the following example, in which we want to multiply two numbers:

import tensorflow as tf # Import TensorFlow

x = tf.constant(8) # X op
y = tf.constant(9) # Y op
z = tf.multiply(x, y) # New op Z

sess = tf.Session() # Create TensorFlow session

out_z = sess.run(z) # execute Z op
sess.close() # Close TensorFlow session
print('The multiplication of x and y: %d' % out_z)# print result

The preceding code segment can be represented by the following figure:

Figure...

The data model in TensorFlow is represented by tensors. Without using complex mathematical definitions, we can say that a tensor (in TensorFlow) identifies a multidimensional numerical array. We will see more details on tensors in the next subsection.

Tensor

Let's see the formal definition of tensor from Wikipedia (https://en.wikipedia.org/wiki/Tensor):

"Tensors are geometric objects that describe linear relations between geometric vectors, scalars, and other tensors. Elementary examples of such relations include the dot product, the cross product, and linear maps. Geometric vectors, often used in physics and engineering applications, and scalars themselves are also tensors."

This data structure is characterized by three parameters: rank, shape, and type, as shown in the following figure:

Figure 6: Tensors are nothing but geometric objects with a shape, rank, and type, used to hold a multidimensional array

A tensor can thus be thought of as the generalization...

TensorFlow includes functions that allow you to debug and optimize programs in a visualization tool called TensorBoard. With TensorBoard, you can graphically observe different types of statistics concerning the parameters and details of any part of the graph.

Moreover, while doing predictive modeling using a complex DNN, the graph can be complex and confusing. To make it easier to understand, debug, and optimize TensorFlow programs, you can use TensorBoard to visualize your TensorFlow graph, plot quantitative metrics about the execution of your graph, and show additional data, such as images that pass through it.

Therefore, TensorBoard can be thought of as a framework designed for analyzing and debugging predictive models. TensorBoard uses the so-called summaries to view the parameters of the model: once a TensorFlow code is executed, we can call TensorBoard to view the summaries in a GUI.