I need to know what's the problem with this code

Hello I need to know what's the problem with this code

% tedata contain  20 cells 
% trdata  contain  60 cells 

for j=1:length(tedata)
  for i=1:length(trdata)  
    y=cell2mat(tedata(j));
    x=cell2mat(trdata(i));
    similarity  = sqrt(sum((x - y) .^ 2));  
    if similarity>180 
      similarity=similarity-360;
    elseif similarity<-180
      similarity=similarity+360;
    end
    kk(i)=similarity;
  end
 [result indxofmin] =min(kk);

but it return the value of similarity at first step [similarity = sqrt(sum((x - y) .^ 2)); ] not after the condition

[dpb]

Some observations...

for j=1:length(tedata)
  y=cell2mat(tedata(j));  % move out of _i_ loop; invariant w/ i
  for i=1:length(trdata)  
    x=cell2mat(trdata(i));
    similarity=norm(x-y);      % use builtin function
    if similarity > 180 
      similarity=similarity-360;
    elseif similarity < -180
      similarity=similarity+360;
    end
    kk(i)=similarity;
  end
 [result indxofmin] =min(kk);

Your code is missing closing end for j loop; presume is at end but you're not saving the result kk for each j; instead overwriting each loop.

similarity = sqrt(sum((x-y).^ 2));

or, more succinctly,

similarity = norm(x-y);

...
elseif similarity < -180
  similarity=similarity+360;

similarity can never be <0; the square operation ensures is always positive, so that condition is totally superfluous.

As for the other, depending on x,y values, it's quite possible that norm(x-y) is never >180 so the condition is never met.

[wisam kh]

#dpb

Thank you very much, things are clear now.

[Image Analyst]

Another major problem with it is that there are not enough comments in it. That would not be acceptable in our company.

[wisam kh]

Ok thank you.