while executing the following matlab code, I have an error like, "Index exceeds matrix dimensions" and "Error in classification (line 34) if test_targetsl(i,1)==1 && result(i,1)==1". plz help me to find the solution to this error.

Question

pramod chaudhary 2017-11-21

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/368284-while-executing-the-following-matlab-code-i-have-an-error-like-index-exceeds-matrix-dimensions

回答： Walter Roberson 2017-11-24

clear;
Data='clusteredData.csv';
CData=csvread(Data);
%Output evaluation variables.
TP=0;
TN=0;
FP=0;
FN=0;
%take out data for mining.
train_data=CData(1:1545,1:2)'
train_targets=CData(1:1545,3)'
%take out data for testing.
test_data=CData(1546:4417,1:2)'
test_targets=CData(1546:4417,3)'
%call C5.0 classification method to create the tree.
[tree,discrete_dim]=C5_0(train_data,train_targets,1)
disp('classify test samples using the tree')
%use the tree built using training data to examine test data. input:test
%data, number of records in test data, tree, discrete_dim indicates number
%of classes i.e. 2, 1 or 2.
result=use_tree(test_data,1:size(test_data,2),tree,discrete_dim,unique(train_targets));
%transpose the matrix
result'
%restore the result to its original state.
resultl=result'
test_data=test_data'
test_targetsl=test_targets'
[S,dim]=size(test_targetsl);
desResult=zeros(S,1);
figure(1)
for i=1:S
    if test_targetsl(i,1)==1 && result(i,1)==1
       TP=TP+1;
       plot(test_data(i,1),test_data(i,2),'*g')
       hold on
    elseif test_targetsl(i,1)==2 && resultl(i,1)==2
        TN=TN+1;
        plot(test_data(i,1),test_data(i,2),'om')
        hold on
    elseif test_targetsl(i,1)==1 && result1(i,1)==2
        FN=FN+1;
        plot(test_data(i,1),test_data(i,2),'xr')
        hold on
    elseif test_targetsl(i,1)==2 && result1(i,1)==1
        FP=FP+1;
        plot(test_data(i,1),test_data(i,2),'+y')
        hold on
    end
end
title('Test data after classification');
grid on
axis equal
for i=1:S
    if result1(i,1)==1
        desResult(i,1)='A';
    else
        desResult(i,1)='N';
    end
end
classifiedReslut=[test_data test_targets' result'];
dlmwrite('classifiedReslut.txt',classifiedReslut,'delimiter','\t');
dlmwrite('NormalabnormalclassifiedReslutReslut.txt',string(desResult),'delimiter','\t');
disp('True Negative')
TN
disp('True Positive')
TP
disp('False Negative')
FN
disp('False Positive')
FP
disp('Detection Rate')
DetectionRate=(TP)/(TP+FP)
disp('False Alarm')
FalseAlarm=(FP)/(FP+TN)
disp('Accuracy')
Accuracy=(TP+TN)/(TP+TN+FP+FN)