For this coursework you will implement two types of Perceptrons in MATLAB and you will apply them to three datasets. You will create a function for training a Perceptron and another function for testing it on a set of test examples. You are given the following three datasets (to be downloaded from the course home page):

• Breast Cancer Wisconsin (breast-cancer.train.txt and breast-cancer.test.txt) – a dataset consisting of 683 examples of described by 9 attributes based on a digitized image of a fine needle aspirate (FNA) of a breast mass. The aim is to learn to diagnose an image as benign (0) or malignant (1).
• Iris (iris.train.txt and iris.test.txt) – a dataset consisting of 150 examples of iris plants described by their sepal length, sepal width, petal length and petal width. The aim is to learn to classify plants into the three types of iris plants: Iris Setosa (1), Iris Versicolour (2) and Iris Virginica (3).
• Wine (wine.train.txt and wine.test.txt) – a dataset consisting of 178 examples of wines described by 13 attributes based on their chemical analysis. The wines were grown in the same region in Italy but derived from three different cultivars. The aim is to learn to determine the origin of a given wine (from the three different possibilities).

For each of the three datasets you will apply the suitable Perceptron and explore different possible settings: different parameters (learning rates and number of epochs), different (random) initial weights, and batch versus stochastic learning.

Deliverables:

1. Your code in electronic form – a zip file containing all files
2. A short report describing your code and your experiments on each dataset (including the performance of your Perceptrons with the different settings and any conclusions you can make based on your experiments).