/
my_svm_dual_train.m
62 lines (45 loc) · 2.3 KB
/
my_svm_dual_train.m
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
function model = my_svm_dual_train(features, classes, options, qqprog)
warning all on
options = check_svm_options(options,size(features,2));
%% svm options
%% train data size
n = size(features,1);
% calculates the number of existent classes
K = options.nclasses;
alpha = [];
supportVector = [];
supportVectorAlphaClasses = [];
switch options.method
% --------------------------------------------------------------------------------------
case 'binary'
%% f is multiplied with (-1) because we are minimizing
%% whereas svm dual form maximizes..
f = -ones(n,1);
%% aplies a kernel (linear, polynomial, and so on)
Kernel = my_svm_kernelfunction( features, features, options );
classes_matrix = repmat( classes, 1, n );
H = Kernel.*classes_matrix.*classes_matrix';
Aeq = classes';
beq = 0;
% ---------------------------------------------------------------------------------------------------------
case 'std_multiclass_basic_architecture'
[zz, bias] = std_multiclass_basic(classesDiffidx,n,K,features,classes,options,qqprog);
case 'std_multiclass_basic_ext_architecture'
[zz, bias] = std_multiclass_basic_ext(n,K,features,classes,options,qqprog);
case 'std_multiclass_sophisticated_architecture'
[zz, bias] = std_multiclass_sophisticated( n,K,features,classes,options,qqprog);
case 'unimodal_basic_architecture'
[zz, bias] = unimodal_basic( n, K, features, classes, options, qqprog );
% case 'unimodal_basic_architecture_special'
% [zz, bias] = unimodal_basic_special( n,K,features,classes,options,qqprog );
% case 'unimodal_basic_architecture_other'
% [zz, bias] = unimodal_basic_other( classesDiffidx,n,K,features,classes,options,qqprog );
case 'unimodal_sophisticated_architecture'
[zz, bias] = unimodal_sophisticated( n,K,features,classes,options,qqprog );
end
options.nclasses = K;
model = struct('zz',zz,'alpha',alpha, 'features', features, 'nclasses', K, ...
'supportVector', supportVector, 'supportVectorAlphaClasses', ...
supportVectorAlphaClasses, ...
'bias', bias, 'options',options);
return;