MULTIPLE LAYER PERCEPTRONs (MLPs)

Multiple Layer Perceptrons (MLPs), MLPs are a great way to get started learning about deep learning technology. MLPs are a type of feedforward neural network that has multiple layers of perceptrons with activation functions. MLPs are made up of a fully connected input and output layer. They have the same number of input and output layers but may have multiple hidden layers and can be used to create speech recognition, image recognition, and machine translation software.

In this section, i will make more several exercise for multi-layer perceptron in deep learning. But, actually i have done experiment on the course of neural network using keras and tensorflow. You can check both of them.

MLPs work by following several steps: First, MLPs deliver data to the network's input layer. The neurons are connected in a graph so that the signal only travels in one direction. Second, MLPs compute input based on the weights that exist between the input and hidden layers. Third, to determine which nodes to fire, MLPs use activation functions. ReLUs, sigmoid functions, and tanh are examples of activation functions. Forth, MLPs use a training data set to train the model to understand the correlation and learn the dependencies between the independent and target variables.

An MLP is illustrated below. To classify images of cats and dogs, the diagram computes weights and bias and employs appropriate activation functions.

Introduction about Iris Flower

The Iris Flower Dataset, also known as Fisher's Iris, is a dataset created by Ronald Fisher, a British statistician and biologist who has made numerous scientific contributions. Ronald Fisher is well known throughout the world as an example of linear discriminant analysis in his paper The use of multiple measurements in taxonomic problems. Ronald Fisher introduced the Iris flower dataset in this paper. The iris database contains 50 samples from three different species of iris. Each of these samples has distinct characteristics that allow them to be divided into three groups: Iris Setosa, Iris Virginica, and Iris versicolor. In this tutorial, we will separate and classify iris samples using a multilayer perceptron. The data set includes 50 samples from each of three Iris species (Iris setosa, Iris virginica, and Iris versicolor). Each sample had four features measured in centimeters: the length and width of the sepals and petals. Fisher created a linear discriminant model to distinguish the species based on the combination of these four features. In this example, I'll build a multilayer perceptron without using any Python libraries. However, we will use the numpy, matplotlib, seaborn, and scikit-learn libraries to help us format and manipulate the iris data set.

Full Code Multilayer Perceptron from Scratch

See the full code below to do more experiment, and download the IRIS.CSV file here.