Hidden layers are the heart of a neural network. The actual decision process happens there. The design of the hidden layers is based on hitting a level beyond which a neural network cannot be optimized further. This level can be defined as the optimal number of hidden layers that produce optimal results.

Hidden layers are the place where the neural network transforms the inputs into a different format that the output layer can consume and use to make predictions. In this recipe, we will design hidden layers for a neural network.