ANN exercise session

Size: px

Start display at page:

Download "ANN exercise session"

Kevin Parks
5 years ago
Views:

1 ANN exercise session In this exercise session, you will read an external file with Iris flowers and create an internal database in Java as it was done in previous exercise session. A new file contains list of 150 observations of iris flowers from 3 different species iris-setosa, iris-versicolor and iris-virginica. There are 4 measurements of given flowers: sepal length, sepal width, petal length and petal width, all in the same unit of centimetres. In this session you need to implement an Artificial Neural Network to define specie of iris flower based on dimensional measurements. Exercise instructions: 1) Use the code developed in Java exercise session VI to create an internal database FlowerList (ArrayList of Flower objects) from csv file. 2) Assign a three number array to each flower to denote their type. For example, the flower Set ose can be represented by (1,0,0). 3) Normalize the data. (x norm = (x i min)/(max min)) 4) Develop ANN algorithm to define a specie of an iris flower based on dimensional parameters: - Create classes Layers, Node and Neurons - Create an ANN with three layers. Input, hidden and Output. - Create 4 nodes and 12 neurons in the input layer. - Create 3 nodes and 9 neurons in the hidden layer. - Create 3 nodes in the output layer. - Write feedforward algorithms based on the given weights for the ANN 5) Test the developed ANN algorithm on all the 150 observations and find the accuracy of the algorithm in classifying the flowers. Core structure of the code: Code for main class: package neuralnetwork; import java.io.bufferedreader; import java.io.bufferedwriter; import java.io.filenotfoundexception; import java.io.filereader; import java.io.filewriter; import java.io.ioexception; import java.util.arraylist; import java.util.collections; import java.util.random;

2 public class ANN { public static void main(string[] args) { // TODO Auto-generated method stub int stop =0,count=0,correct=0; String datafile = "iris.csv"; BufferedReader br = null; String line = ""; String SplitBy = ","; double errorref = ; Random rand = new Random(); Layers inputlayer = new Layers(4,12,null); Layers hiddenlayer = new Layers(3,9,inputLayer); Layers outlayer = new Layers(3,0,hiddenLayer); //Assign Random weights to the neurons double[] weightsinput = { , , , , , , , , , , , ; double[] weightshidden = { , , , , , , , , ; inputlayer.setweights(weightsinput); hiddenlayer.setweights(weightshidden); ArrayList<Flower> flowerlist = new ArrayList<Flower>(); ArrayList<Flower> learnlist = new ArrayList<Flower>(); ArrayList<Flower> testlist = new ArrayList<Flower>(); try { // use code from Java exercise IV to create internal database of iris flowers catch (FileNotFoundException e) { e.printstacktrace(); catch (IOException e) { e.printstacktrace(); normalizedata(flowerlist); while(count<learnlist.size()){ count=0; stop++; ArrayList<Double> error = new ArrayList<Double>(); for(int i=0;i<learnlist.size();i++){ feedforward(inputlayer,hiddenlayer,outlayer,learnlist.get(i),i); error.add(calerror(outlayer,learnlist.get(i),i)); if(error.get(i)<errorref){ count++; error.clear();

3 if(stop==3000) break; clearvalues(inputlayer,hiddenlayer,outlayer); clearvalues(inputlayer,hiddenlayer,outlayer); // Test the developed neural network with the test set public static void normalizedata(arraylist<flower> flowerlist){ // write code for normalizing the data public static void clearvalues(layers input, Layers hidden, Layers output){ //clear the values at each node for the next iteration public static double calerror(layers output, Flower fl, int index){ double error=0; //calculate the error between the neural network output and target values return error; public static void feedforward(layers input, Layers hidden, Layers output, Flower fl, int index){ //calculate the value at every node Code for Class Layers: import java.util.arraylist; public class Layers { ArrayList<Node> nodes = new ArrayList<Node>(); ArrayList<Neurons> neuronsf = new ArrayList<Neurons>(); ArrayList<Node> previousnodes = new ArrayList<Node>();

4 int units; public Layers(int numnodes,int numneurons,layers layer){ for(int j=0;j<numneurons;j++){ this.neuronsf.add(new Neurons()); for(int i=0;i<numnodes;i++){ this.nodes.add(new Node()); for(int j=0;j<(numneurons/numnodes);j++){ this.nodes.get(i).neuronsforward.add(this.neuronsf.get(i+j*(numnodes))); if(layer!=null){ for(int i=0;i<this.nodes.size();i++){ for(int j=0;j<layer.nodes.size();j++){ this.previousnodes.add(layer.nodes.get(j)); this.nodes.get(i).neuronsprevious.add(layer.neuronsf.get(j+i*layer.nodes.size())); public void setweights(double[] weight){ for(int i=0;i<this.neuronsf.size();i++){ this.neuronsf.get(i).setweight(weight[i]); Code for class Node:

5 import java.util.arraylist; public class Node { ArrayList<Double> value = new ArrayList<Double>(); ArrayList<Neurons> neuronsprevious = new ArrayList(); ArrayList<Neurons> neuronsforward = new ArrayList(); double delta; Code for class Neurons: public class Neurons { double weight; public void setweight(double weight){ this.weight = weight;

KTH ROYAL INSTITUTE OF TECHNOLOGY. Lecture 14 Machine Learning. K-means, knn

KTH ROYAL INSTITUTE OF TECHNOLOGY Lecture 14 Machine Learning. K-means, knn Contents K-means clustering K-Nearest Neighbour Power Systems Analysis An automated learning approach Understanding states in