LSVT-classification/accuracynb.m at master · Merna-Atef/LSVT-classification · 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
function [acc sen spec ret] = accuracynb(train,trainresult,test,testresult)
ind2 = find(trainresult==2);
ind1 = find(trainresult==1);

class1 = train(ind1,:);
class2 = train(ind2,:);

mu1=mean(class1);
sigma1=std(class1);
ret =[];
mu2=mean(class2);
sigma2=std(class2);

pclass1 = size(class1,1)/size(train,1);
pclass2 = size(class2,1)/size(train,1);

for i=1:size(test,1)
prob1 = normpdf(test(i,:),mu1,sigma1);
prob2 = normpdf(test(i,:),mu2,sigma2);

prob1 = prod(prob1)*pclass1;
prob2 = prod(prob2)*pclass2;

reta = prob1<=prob2;
reta=reta+1;
ret=[ret; reta];
end
P = find(ret==1);
P = size(P,1);
N = find(ret==2);
N = size(N,1);

TP = (ret==testresult.* (ret==1));
TN = (ret==testresult.* (ret==2));
TP = find(TP==1);
TP = size(TP,1);
TN = find(TN==1);
TN = size(TN,1);
sen = TP/P;
spec =TN/N;
acc = ret==testresult;
acc= find(acc==1);
acc = size(acc,1)/size(test,1);
end