MLP
MLP is a very complex algorithm with many details, and going over its functioning and different parts abstractly will be difficult to follow. So, let's dive in with an example. We will continue using the number of MSU applications in this section.
While linear regression uses an equation, MLP uses a network of neurons to connect the independent attributes to the dependent attribute. An example of such a network is shown in the following screenshot:
Figure 6.8 – An MLP network example for the number of MSU applications problem
Every MLP network has six distinct parts. Let's go through these parts using Figure 6.8, as follows:
- Neurons: Each of the circles in Figure 6.8 is called a neuron. A neuron could be in the input layer, output layer, and hidden layers. We will cover three tree types of layers in the following section.
- Input layer: A layer of neurons from which values are inputted to the network. In a prediction task,...
