ele888 lab1 run - graphs, C (gcc)

ele888 lab1 run - graphs

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% LAB 1, Bayesian Decision Theory
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

% Attribute Information for IRIS data:
%    1. sepal length in cm
%    2. sepal width in cm
%    3. petal length in cm
%    4. petal width in cm

%    class label/numeric label: 
%       -- Iris Setosa / 1 
%       -- Iris Versicolour / 2
%       -- Iris Virginica / 3

%% this script will run lab1 experiments..
clear
load irisdata.mat

%% extract unique labels (class names)
labels = unique(irisdata_labels);

%% generate numeric labels
numericLabels = zeros(size(irisdata_features,1),1);
for i = 1:size(labels,1)
    numericLabels(find(strcmp(labels{i},irisdata_labels)),:)= i;
end

%% feature distribution of x1 for two classes
%  figure
%      
%  subplot(1,2,1), hist(irisdata_features(find(numericLabels(:)==1),2),100), title('Iris Setosa, sepal width (cm)');
%  subplot(1,2,2), hist(irisdata_features(find(numericLabels(:)==2),2),100); title('Iris Veriscolour, sepal width (cm)');
%  
%  figure
%  
%  subplot(1,2,1), hist(irisdata_features(find(numericLabels(:)==1),1),100), title('Iris Setosa, sepal length (cm)');
%  subplot(1,2,2), hist(irisdata_features(find(numericLabels(:)==2),1),100); title('Iris Veriscolour, sepal length (cm)');
%      
%  figure  %for feature Sepal Width
%  
%  histogram(irisdata_features(find(numericLabels(:)==1),2),100, 'BinWidth', 0.1, 'FaceColor', 'r');  
%  hold on;       
%  histogram(irisdata_features(find(numericLabels(:)==2),2),100, 'BinWidth', 0.1, 'FaceColor', 'b');
%  
%  line([3.1 3.1], [0 10], 'LineStyle', '-.', 'Color', 'k') %Line to indicate threshold boundary (Q4)
%  line([3.3 3.3], [0 10], 'LineStyle', '--', 'Color', 'k') %Line to indicate threshold boundary (Q5)
%  hold off;
%  axis ([1.8 4.5 0 10]);
%  
%  legend ("Iris Setosa","Iris Versicolour");
%  xlabel("Sepal Width (cm)");
%  ylabel("Number of Training Data")
%  title("Graph 1: Iris Sepal Width with threshold boundary line");
%  figure %for feature Sepal Length
%  
%  histogram(irisdata_features(find(numericLabels(:)==1),1),100, 'BinWidth', 0.1, 'FaceColor', 'r');  
%  hold on;       
%  histogram(irisdata_features(find(numericLabels(:)==2),1),100, 'BinWidth', 0.1, 'FaceColor', 'b');
%  
%  
%  line([5.5 5.5], [0 16.5], 'LineStyle', '--', 'Color', 'k') %Line to indicate threshold boundary (Q6)
%  hold off;
%  axis ([4.3 7.2 0 16.5]);
%  
%  legend ("Iris Setosa","Iris Versicolour");
%  xlabel("Sepal Length (cm)");
%  ylabel("Number of Training Data")
%  title("Graph 2: Iris Sepal Length with threshold boundary");
%  
%  
%  figure
%  
%  plot(irisdata_features(find(numericLabels(:)==1),1),irisdata_features(find(numericLabels(:)==1),2),'rs'); title('x_1 vs x_2');
%  hold on;
%  plot(irisdata_features(find(numericLabels(:)==2),1),irisdata_features(find(numericLabels(:)==2),2),'k.');
%  axis([4 7 1 5]);
%  legend("Versicolour","Setosa")
%  ylabel("Number of Training Samples")
%  xlabel("Sepal Width (cm)")
%  title("Training Data for Setosa and Versicolour Irises with Feature: Sepal Length")
%

%% build training data set for two class comparison
% merge feature samples with numeric labels for two class comparison (Iris
% Setosa vs. Iris Veriscolour
trainingSet = [irisdata_features(1:100,:) numericLabels(1:100,1) ];

%% Lab1 experiments (include here)
for x = [3.3, 4.4, 5.0, 5.7, 6.3]                    %x = [2 6; 4.4 3; 5 3.5; 5.3 2; 5.5 2.5; 6.6 3.5; 4.5 6.1]
    lab1(x,trainingSet)
end


run \| edit \| history \| help	0