matlab/code.m at master · SSabev/matlab · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
clear all;
clc()

load data_90.mat

[B, mean1, mean2] = init_values(data_90);

data_size=size(data_90(:,1));

means = [mean1; mean2];
[clusters2, centres2, sum_squared_func2] = mykmeans(data_90, means);

% figure
% scatter3(data_90(:,1),data_90(:,2),data_90(:,3), 100, clusters2, 'filled')
% xlabel SL, ylabel SW, zlabel PL

[mean3] = NM(data_90,data_size, means);
means = [means; mean3];
[clusters3, centres3, sum_squared_func3] = mykmeans(data_90, means);


load true_90.mat;

true = 0;
false = 0;

for i=1:data_size(1)
    if clusters3(i) == 2
        clusters3(i) = 3;
    elseif clusters3(i) == 3
        clusters3(i) = 2;
    end
end

for i=1:data_size(1)
    if clusters3(i) ~= true_90(i)
        false = false+1;
    else
        true = true+1;
    end
end

true
false


mycov1 = zeros(3,3);
mycov2 = zeros(3,3);
mycov3 = zeros(3,3);

sum1 = sum(clusters3 == 1)
sum2 = sum(clusters3 == 2)
sum3 = sum(clusters3 == 3)

for i=1:data_size
    if clusters3(i) == 1
        mycov1 = mycov1 + (data_90(i,:) - centres3(1,:))'*(data_90(i,:) - centres3(1,:));
    elseif clusters3(i) == 2
        mycov2 = mycov2 + (data_90(i,:) - centres3(3,:))'*(data_90(i,:) - centres3(3,:));
    elseif clusters3(i) == 3
        mycov3 = mycov3 + (data_90(i,:) - centres3(2,:))'*(data_90(i,:) - centres3(2,:));
    end
end


mycov1 = mycov1./sum1
mycov2 = mycov2./sum3
mycov3 = mycov3./sum2

% figure
% scatter3(data_90(:,1),data_90(:,2),data_90(:,3), 100, true_90, 'filled')
% xlabel SL, ylabel SW, zlabel PL
%
% figure
% scatter3(data_90(:,1),data_90(:,2),data_90(:,3), 100, clusters3, 'filled')
% xlabel SL, ylabel SW, zlabel PL

[mean4] = NM(data_90, data_size, means);
means = [means; mean4];
[clusters4, centres4, sum_squared_func4] = mykmeans(data_90, means);

[mean5] = NM(data_90, data_size, means);
means = [means; mean5];
[clusters5, centres5, sum_squared_func5] = mykmeans(data_90, means);


% figure
% subplot(2,2,1); plot(sum_squared_func2)
% title('Sum-squared error function for 2 clusters')
% subplot(2,2,2); plot(sum_squared_func3)
% title('Sum-squared error function for 3 clusters')
% subplot(2,2,3); plot(sum_squared_func4)
% title('Sum-squared error function for 4 clusters')
% subplot(2,2,4); plot(sum_squared_func5)
% title('Sum-squared error function for 5 clusters')

load data_900.mat;

[Distances, mean1, mean2] = init_values(data_900);

data_size=size(data_900(:,1));

gclusters = zeros(data_size(1),1);

for i=1:data_size(1)
    prob1 = (1/(2*pi)^1.5)*(det(mycov1)^1.5)*exp((-0.5)*(data_900(i,:)-centres3(1,:))*mycov1^(-1)*(data_900(i,:)-centres3(1,:))');
    prob2 = (1/(2*pi)^1.5)*(det(mycov3)^1.5)*exp((-0.5)*(data_900(i,:)-centres3(3,:))*mycov3^(-1)*(data_900(i,:)-centres3(3,:))');
    prob3 = (1/(2*pi)^1.5)*(det(mycov2)^1.5)*exp((-0.5)*(data_900(i,:)-centres3(2,:))*mycov2^(-1)*(data_900(i,:)-centres3(2,:))');

    if prob1 == max([prob1, prob2, prob3])
        gclusters(i) = 1;
    elseif prob2 == max([prob1, prob2, prob3])
        gclusters(i) = 2;
    elseif prob3 == max([prob1, prob2, prob3])
        gclusters(i) = 3;
    end
end


confusion_matrix_gaussian = zeros(3,3);
load true_900.mat;

gclusters

for i=1:data_size(1)
    if true_900(i) == gclusters(i) && true_900(i)==1
        confusion_matrix_gaussian(1,1) = confusion_matrix_gaussian(1,1) + 1;
    elseif true_900(i) == gclusters(i) && true_900(i)==2;
        confusion_matrix_gaussian(2,2) = confusion_matrix_gaussian(2,2) + 1;
    elseif true_900(i) == gclusters(i) && true_900(i)==3;
        confusion_matrix_gaussian(3,3) = confusion_matrix_gaussian(3,3) + 1;
    elseif true_900(i)~=gclusters(i) && true_900(i) == 1 && gclusters(i) == 2
        confusion_matrix_gaussian(2,1) = confusion_matrix_gaussian(2,1) +1;
    elseif true_900(i)~=gclusters(i) && true_900(i) == 1 && gclusters(i) == 3
        confusion_matrix_gaussian(3,1) = confusion_matrix_gaussian(3,1) +1;
    elseif true_900(i)~=gclusters(i) && true_900(i) == 2 && gclusters(i) == 1
        confusion_matrix_gaussian(1,2) = confusion_matrix_gaussian(1,2) +1;
    elseif true_900(i)~=gclusters(i) && true_900(i) == 2 && gclusters(i) == 3
        confusion_matrix_gaussian(3,2) = confusion_matrix_gaussian(3,2) +1;
    elseif true_900(i)~=gclusters(i) && true_900(i) == 3 && gclusters(i) == 1
        confusion_matrix_gaussian(1,3) = confusion_matrix_gaussian(3,1) +1;
    elseif true_900(i)~=gclusters(i) && true_900(i) == 3 && gclusters(i) == 2
        confusion_matrix_gaussian(2,3) = confusion_matrix_gaussian(2,3) +1;
    end
end

confusion_matrix_gaussian

means = [mean1; mean2];
[mean3] = NM(data_900,data_size, means);
means = [means; mean3];

[clusters3, centres3, sum_squared_func3_2] = mykmeans(data_900, means);


for i=1:data_size(1)
    if true_900(i) == gclusters(i) && true_900(i)==1
        confusion_matrix_gaussian(1,1) = confusion_matrix_gaussian(1,1) + 1;
    elseif true_900(i) == gclusters(i) && true_900(i)==2;
        confusion_matrix_gaussian(2,2) = confusion_matrix_gaussian(2,2) + 1;
    elseif true_900(i) == gclusters(i) && true_900(i)==3;
        confusion_matrix_gaussian(3,3) = confusion_matrix_gaussian(3,3) + 1;
    elseif true_900(i)~=gclusters(i) && true_900(i) == 1 && gclusters(i) == 2
        confusion_matrix_gaussian(2,1) = confusion_matrix_gaussian(2,1) +1;
    elseif true_900(i)~=gclusters(i) && true_900(i) == 1 && gclusters(i) == 3
        confusion_matrix_gaussian(3,1) = confusion_matrix_gaussian(3,1) +1;
    elseif true_900(i)~=gclusters(i) && true_900(i) == 2 && gclusters(i) == 1
        confusion_matrix_gaussian(1,2) = confusion_matrix_gaussian(1,2) +1;
    elseif true_900(i)~=gclusters(i) && true_900(i) == 2 && gclusters(i) == 3
        confusion_matrix_gaussian(3,2) = confusion_matrix_gaussian(3,2) +1;
    elseif true_900(i)~=gclusters(i) && true_900(i) == 3 && gclusters(i) == 1
        confusion_matrix_gaussian(1,3) = confusion_matrix_gaussian(1,3) +1;
    elseif true_900(i)~=gclusters(i) && true_900(i) == 3 && gclusters(i) == 2
        confusion_matrix_gaussian(2,3) = confusion_matrix_gaussian(2,3) +1;
    end
end


for i=1:data_size(1)
    if clusters3(i) == 2
        clusters3(i) = 3;
    elseif clusters3(i) == 3
        clusters3(i) = 2;
    end
end

confusion_matrix_kmeans = zeros(3,3);
for i=1:data_size(1)
    if true_900(i) == clusters3(i) && true_900(i)==1
        confusion_matrix_kmeans(1,1) = confusion_matrix_kmeans(1,1) + 1;
    elseif true_900(i) == clusters3(i) && true_900(i)==2;
        confusion_matrix_kmeans(2,2) = confusion_matrix_kmeans(2,2) + 1;
    elseif true_900(i) == clusters3(i) && true_900(i)==3;
        confusion_matrix_kmeans(3,3) = confusion_matrix_kmeans(3,3) + 1;
    elseif true_900(i)~=clusters3(i) && true_900(i) == 1 && clusters3(i) == 2
        confusion_matrix_kmeans(1,2) = confusion_matrix_kmeans(1,2) +1;
    elseif true_900(i)~=clusters3(i) && true_900(i) == 1 && clusters3(i) == 3
        confusion_matrix_kmeans(1,3) = confusion_matrix_kmeans(1,3) +1;
    elseif true_900(i)~=clusters3(i) && true_900(i) == 2 && clusters3(i) == 1
        confusion_matrix_kmeans(2,1) = confusion_matrix_kmeans(2,1) +1;
    elseif true_900(i)~=clusters3(i) && true_900(i) == 2 && clusters3(i) == 3
        confusion_matrix_kmeans(2,3) = confusion_matrix_kmeans(2,3) +1;
    elseif true_900(i)~=clusters3(i) && true_900(i) == 3 && clusters3(i) == 1
        confusion_matrix_kmeans(3,1) = confusion_matrix_kmeans(3,1) +1;
    elseif true_900(i)~=clusters3(i) && true_900(i) == 3 && clusters3(i) == 2
        confusion_matrix_kmeans(3,2) = confusion_matrix_kmeans(3,2) +1;
    end
end

confusion_matrix_kmeans

% figure
% scatter3(data_900(:,1),data_900(:,2),data_900(:,3), 100, true_900, 'filled')
% xlabel SL, ylabel SW, zlabel PL
%
% figure
% scatter3(data_900(:,1),data_900(:,2),data_900(:,3), 100, clusters3, 'filled')
% xlabel SL, ylabel SW, zlabel PL


Cov = [1 0.5 0.3
       0.5 2 0
       0.3 0 3];
mu = [1 2 3]';

[U,L] = eig(Cov);
% L: eigenvalue diagonal matrix
% U: eigen vector matrix, each column is an eigenvector

% For N standard deviations spread of data, the radii of the eliipsoid will
% be given by N*SQRT(eigenvalues).

N = 1; % choose your own N
radii = N*sqrt(diag(L));

% generate data for "unrotated" ellipsoid
[xc,yc,zc] = ellipsoid(0,0,0,radii(1),radii(2),radii(3));

% rotate data with orientation matrix U and center mu
a = kron(U(:,1),xc);
b = kron(U(:,2),yc);
c = kron(U(:,3),zc);

data = a+b+c; n = size(data,2);

x = data(1:n,:)+mu(1);
y = data(n+1:2*n,:)+mu(2);
z = data(2*n+1:end,:)+mu(3);

% now plot the rotated ellipse
% sc = surf(x,y,z); shading interp; colormap copper
h = surfl(x, y, z); colormap copper
title('actual ellipsoid represented by mu and Cov')
axis equal
alpha(0.7)