Reinforcement Learning with TensorFlow

Basic computations in TensorFlow

The base of TensorFlow is the computational graph, which we discussed earlier in this chapter, and tensors. A tensor is an n-dimensional vector. Thus, a scalar and a matrix variable is also a tensor. Here, we will try some of the basic computations to start with TensorFlow. Please try to implement this section in a python IDE such as Jupyter Notebook.

For the TensorFlow installation and dependencies please refer to the following link:

https://www.tensorflow.org/install/

Import tensorflow by the following command:

import tensorflow as tf

tf.zeros() and tf.ones() are some of the functions that instantiate basic tensors. The tf.zeros() takes a tensor shape (that is, a tuple) and returns a tensor of that shape ...