I wanna use kNN algortihm without using kNNclassfy in matlab.But in my codes there is a big fault that i cant see now.Please contact with me immediately.Coz i have to finish it till tuesday.
1 view (last 30 days)
Show older comments
figure axis([0,10,0,10])
[x,y] = ginput(20); scatter(x,y,'r+') A=[x,y];
hold
[z,t] =ginput(20); scatter(z,t,'b*')
B=[z,t];
[m,l]=ginput(10); scatter(m,l,'go') C=[m,l];
pause
k= input( ' k değerini giriniz= ')
%[row,col]=size(A);
%[row1,col1]=size(B);
%dist=zeros(row,row1);
%nnarray = zeros(row,row1);
%nnarray1 = zeros(row,m);
%for i=1:row
%for j=1:row1
% dist(i,j)=sqrt(sum((A(i,:)-C(j,:)).^2));
%end
%[y,index]=sort(dist(i,:));
%nnarray(i,:)=index';
% q=index';
%end
%for i=1:row
%for j=1:row1
% dist(i,j)=sqrt(sum((B(i,:)-C(j,:)).^2));
%end
%[y1,index1]=sort(dist(i,:));
% nnarray(i,:)=index1';
% w=index1';
%end
%pause(2)
%for i=1:10 %for j=1:k % q1(j,i)=q(j,i); % q1((j+1),k)=w(j,i) % end %end
%q2= sort(q1,1,'ascend');
%for f=1:10
%m=0
%n=0
%for i=1:k
%for j=1:10
% if q2(i,f) ==q(j,f)
% m=m+1
% else if q2(i,f) == w(j,f)
% n=n+1
% end
% end
% end
% end
%if m<n;
% e(f)=C(f)
% plot(e,'g+')
%else
% f(f)=C(f)
% plot (f,'r*')
%end
%end
In my question i want to create 2 clusters and classify the data via kNN.Especially each cluster(composed of 20 points) totally 40 point in figure frame via mouse we have to select the points with ginput code.Also for testing we have to select 10 points.We have to write down the kNN codes in matlab without using kNNclassify.Please show the cluster of the training data.
3 Comments
Image Analyst
on 26 Jan 2015
All of the code after you ask the user to click points is commented out and I don't know what code you want in there or don't. And you haven't described what "a big fault" is, or where it occurs. After you read the helpful link John gave, I suggest you also read this link: http://blogs.mathworks.com/videos/2012/07/03/debugging-in-matlab/
Answers (0)
See Also
Categories
Find more on Parallel and Cloud in Help Center and File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!