How does ANN work?
In this section, we will see which are the elements taking part in an ANN-MLP. First, we will represent a regular ANN-MLP shape with one layer each of input, output, and hidden, and how the information flows across them:
An MLP network is formed by at least three layers:
- Input layer: Every MLP always has one of these layers. It is a passive layer, which means that it does not modify the data. It receives information from the outside world and sends it out to the network. The number of nodes (neurons) in this layer will depend on the amount of features or descriptive information we want to extract from the images. For example, in case of using feature vectors, there will be one node for each of the columns within the vector.
- Hidden layers: This layer is where all the groundwork happens. It transforms the inputs into something that the output layer or another hidden layer can use (there can be more than one). This layer works as a black box, sensing patterns within received...
