Return and call multiple value from function - matlab

Question

I am trying to return all values of for loop from my function and also I need to call them outside of the function.But i only get last row of the for loop, I also need previous result of for loop.
Here is my code
i=input('Enter a start row: ');
j=input('Enter a end row: ');
c=input('Enter classifier variable column number:')
search= importfiledataset('search-queries-features.csv',i,j);
[n,p]=size(search);
if j>n
disp('Please enter a smaller number!');
end
[D1,Eps1]=findD(i,j,c,search);
disp(D1);
disp(n1);
function [D1,Eps1] = findD(i,j,c,search) %Find D vlaues with for loop for each classification value
for numOfClassifier = 1 : 100
a = search(search(:,c)==numOfClassifier,:) ;
q1 = a(all(~isnan(a),2),:); % for nan - rows
D1 = a(:,all(~isnan(a))) % for nan - columns WE FIND D1
n1=size(D1,1) %number of record belongs to classification
sampleSpace = size(search,1) %sample space of the priop probability(#of c1 + #of c2 .... cn)
pc1 = n1/sampleSpace %prior probability of the n
mu1 = mean(D1) %mean of D1
Z1 = D1 - mu1 % centered data of D1
Eps1 = (1/n1)*(transpose(Z1)*Z1) %covariance matrix if Z1
numOfClassifier = numOfClassifier + 1;
if search(:,c) ~= numOfClassifier
break
end
end
end
I need to return
D1
Eps1
I want to return the entire values of for loop but I only end up with values from the last row.

Answer 1

You can store the iteration results in a cell array, and return the cell array:
Function returning two cell arrays (I named them allD1 and allEps1):
function [allD1, allEps1] = findD(i,j,c,search)
Initialize the cell arrays to empty cells before the loop:
allD1 = {};
allEps1 = {};
Add D1 and Eps1 to the end of the cell arrays (place it before the line with the break statement):
allD1{end + 1} = D1;
allEps1{end + 1} = Eps1;
Here is the modified code:
i=input('Enter a start row: ');
j=input('Enter a end row: ');
c=input('Enter classifier variable column number:')
search= importfiledataset('search-queries-features.csv',i,j);
[n,p]=size(search);
if j>n
disp('Please enter a smaller number!');
end
[D1,Eps1]=findD(i,j,c,search);
disp(D1);
disp(n1);
function [allD1, allEps1] = findD(i,j,c,search) %Find D vlaues with for loop for each classification value
%Initialize empty cell arrays
allD1 = {};
allEps1 = {};
for numOfClassifier = 1 : 100
a = search(search(:,c)==numOfClassifier,:) ;
q1 = a(all(~isnan(a),2),:); % for nan - rows
D1 = a(:,all(~isnan(a))) % for nan - columns WE FIND D1
n1=size(D1,1) %number of record belongs to classification
sampleSpace = size(search,1) %sample space of the priop probability(#of c1 + #of c2 .... cn)
pc1 = n1/sampleSpace %prior probability of the n
mu1 = mean(D1) %mean of D1
Z1 = D1 - mu1 % centered data of D1
Eps1 = (1/n1)*(transpose(Z1)*Z1) %covariance matrix if Z1
%Add D1 (and Eps1) to the end of the cell array
allD1{end + 1} = D1;
allEps1{end + 1} = Eps1;
numOfClassifier = numOfClassifier + 1;
if search(:,c) ~= numOfClassifier
break
end
end
end
Cell arrays are more general then arrays - cells can store values of different types and shapes, opposed to arrays where all elements must be of the same type.
In your case, multi-dimensional arrays might fit, but it's more confusing.
With cell arrays, each cell holds the result of the matching iteration.
For example: allD1{3} holds the value of D1 in the third iteration.