Installing dependencies

First, let's install TensorFlow. It is an open source framework for building Artificial Neural Network (ANN):

pip install tensorflow

You should see the following output, which indicates a successful installation:

We can also manually check the installation. Type python to open the Python prompt on the command line. Once inside the Python prompt, type import tensorflow and hit Enter. You should see the following output:

Type exit() to return to the default command line, keep in mind that we are still inside the ai-projects conda environment.

Next, we are going to install Keras, a wrapper over TensorFlow that makes designing deep neural networks much more intuitive. We continue to use the pip command:

pip install keras

On successful installation, we should see the following output:

To manually check the installation, type python to open the Python prompt on the command line. Once inside the Python prompt, type import keras and hit Enter. You should see the following output, with no errors. Observe that the output mentions that Keras is using TensorFlow as its backend:

Great! We have now installed the main dependencies required to create our very own neural networks. Let's go ahead and build an ANN to predict real estate prices.