TensorFlow is an open source software library from Google, which is extensively used for numerical computation. It is one of the most popularly used libraries for building deep learning models. It is highly scalable and runs on multiple platforms, such as Windows, Linux, macOS, and Android. It was originally developed by the researchers and engineers of the Google Brain team.

TensorFlow supports execution on everything, including CPUs, GPUs, and TPUs, which are tensor processing units, and on mobile and embedded platforms. Due to its flexible architecture and ease of deployment, it has become a popular choice of library among many researchers and scientists for building deep learning models.

In TensorFlow, every computation is represented by a data flow graph, also known as a computational graph, where a node represents operations, such as addition or multiplication, and an edge represents tensors. Data flow graphs can also be shared and executed on many different platforms. TensorFlow provides a visualization tool, called TensorBoard, for visualizing data flow graphs.

A tensor is just a multidimensional array. So, when we say TensorFlow, it is literally a flow of multidimensional arrays (tensors) in a computation graph.

You can install TensorFlow easily through pip by just typing the following command in your Terminal. We will install TensorFlow 1.13.1:

pip install tensorflow==1.13.1

We can check the successful installation of TensorFlow by running the following simple Hello TensorFlow! program:

import tensorflow as tf

hello = tf.constant("Hello TensorFlow!")
sess = tf.Session()
print(sess.run(hello))

The preceding program should print Hello TensorFlow!. If you get any errors, then you probably have not installed TensorFlow correctly.