A Primer on TensorFlow
TensorFlow is one of the most popular deep learning libraries. In upcoming chapters, we will use TensorFlow to build deep reinforcement models. So, in this chapter, we will get ourselves familiar with TensorFlow and its functionalities.
We will learn about what computational graphs are and how TensorFlow uses them. We will also explore TensorBoard, which is a visualization tool provided by TensorFlow used for visualizing models. Going forward, we will understand how to build a neural network with TensorFlow to perform handwritten digit classification.
Moving on, we will learn about TensorFlow 2.0, which is the latest version of TensorFlow. We will understand how TensorFlow 2.0 differs from its previous versions and how it uses Keras as its high-level API.
In this chapter, we will learn about the following:
- Computational graphs and sessions
- Variables, constants, and placeholders