Hands-On Python Deep Learning for the Web

Project architecture

We will be using Node.js in this project to create a server. This is done so that we get the benefits of faster computational performance of TF.js when used through the Node.js backend. We will create a very basic frontend that will be able to issue a command to perform the training of the neural network built using TF.js and another button to issue a command to predict the class of a hypothetical feature vector of an Iris flower based on input provided by the user.

The following diagram shows the components of the project, along with their interactions:

Now that we know about the architecture, let's start with the project. ...