Difficulty: beginner
Estimated Time: 10 minutes

TensorFlow is the platform enabling building deep Neural Network architectures and performing Deep Learning. This tutorial introduces the basics needed to create more complex structures.

In this scenario, you will learn how to use TensorFlow Core concepts. The content is based on the official TensorFlow tutorial.

To take the most of this course, you should know how to program in Python or other languages that would allow you to understand Python syntax.


You've completed TensorFlow Core scenario.

You've learned how to:

  • Create computational graph
  • Use constants, placeholders and variables
  • Run the session

TensorFlow Core

Step 1 of 6

Importing TensorFlow

To start working with Python shell run the following command:


You can use quit() to leave it.

To start working TensorFlow you should first import it to give Python access to all its assets:

import tensorflow as tf

Later in a code whenever you want to use TensorFlow classes, methods or symbols you should just refer to tf variable.

The core concept of the library (hence the name) are tensors. A tensor is an array of any number of dimensions. This data shape is very convenient for a lot of neural networks computations. We will see the examples later in this and future scenarios.