could anybody help me how to solve the error in the following code

code:

global q;
m=[10 20 30 40 50];
n=[60 70 80 90 100]
for t= 1:length(m)
 for r= 1:length(n)
     A=rand(m(t),n(r));
     n_=ceil(n(r)/m(t))
  C=repmat(diag(1:m(t)),1,n_ )
      unused_rows=1:m;
  t=1;
  while ~isempty(unused_rows)

        n_rows=ceil(sqrt(randi(numel(unused_rows))));
        rows=unused_rows(randsample(length(unused_rows),n_rows));
        [~,idx]=find(ismember(unused_rows,rows));
        unused_rows(idx)=[];
        [C,C_part]=cluster_rows2(A,C,rows);

         CP{q}=C_part %indxing all cells/groups
         q=q+1;
    end
   end
  end

All_CP=[cat(1, CP{:})] %combine all cells/groups into one cell
------------------------------------------------------------------
function [C,C_part]=cluster_rows2(A,B,rows)
global q;

    A_part=A(rows,:);
    B_part=B(rows,:);
    %sum along the the vertical axis to get indices for the non-0 columns
    non_0=sum(B_part);
    %Repeat the vector back to the same size as the partial matrix. It will be
    %converted to a logical later on.
    non_0=repmat(non_0,length(rows),1);
    %create a copy of B, as that is the basis for C
     C=B;
     C_part=B_part;
    %for all non-zero columns, replace the 0 positions with the values from A
     C_part(non_0 & C_part==0)=A_part(non_0 & C_part==0)
    %paste the edited partial matrix back
    C(rows,:)=C_part;

    CP{q}=C_part;
  end

[Torsten]

Error message ?

[Prabha Kumaresan]

Error in (line 21) CP{q}=C_part %indxing all cells/groups The left hand side is initialized and has an empty range of indices. However, the right hand side returned one or more results.

[Stephen23]

"...how to solve the error in the following code"

Simple: do NOT use global variables.

[Prabha Kumaresan]

Ok.I stop using global variables.could you help me to concatenate vertically E_part together with respect to m and n.

Answer 1

[Jan]

The old problem of global variables.

global q;
...
CP{q} = C_part;

Now it seems like your q is not a scalar. It is impossible to find the reason for this, because by declaring it as global, the value can be set from anywhere.

It is impossible to fix this code for you, because you did not show the part, where q is defined. I recommend to omit the global and use a standard variable:

q = 1;

If you need it in cluster_rows2(), provide it as input argument.

It looks strange that

CP{q}=C_part;

appears in the subfunction and the mainfunction also. Defining CP inside cluster_rows2() is meaningless, because this variable is not returned to the caller.

[Prabha Kumaresan]

I have made the changes in the code 
global q;
m=[10 20 30 40 50  ];
n=[60 70 80 90 100 ];
iwant = cell(length(m),1);
for t= 1:length(m)
 for r= 1:length(n)
     A=rand(m(t),n(r));
     n_=ceil(n(r)/m(t))
  B=repmat(diag(1:m(t)),1,n_ )
iwant{t}=B(:,1:n(r));%crop back to only needed cols
 C = iwant{t};  
      unused_rows=1:m;
  q =1;
  while ~isempty(unused_rows)

        n_rows=ceil(sqrt(randi(numel(unused_rows))));
        rows=unused_rows(randsample(length(unused_rows),n_rows));
        [~,idx]=find(ismember(unused_rows,rows));
        unused_rows(idx)=[];
        [C,C_part]=cluster_rows2(A,B,rows);
        CP{q}=C_part %indxing all cells/groups
        q=q+1;
    end
   end
  end

All_CP=[cat(1, CP{:})] %combine all cells/groups into one cell
-------------------------
------------------------------
function [C,C_part]=cluster_rows2(A,B,rows)
global q;

    A_part=A(rows,:);
    B_part=B(rows,:);
    %sum along the the vertical axis to get indices for the non-0 columns
    non_0=sum(B_part);
    %Repeat the vector back to the same size as the partial matrix. It will be
    %converted to a logical later on.
    non_0=repmat(non_0,length(rows),1);
    %create a copy of B, as that is the basis for C
     C=B;
     C_part=B_part;
    %for all non-zero columns, replace the 0 positions with the values from A
     C_part(non_0 & C_part==0)=A_part(non_0 & C_part==0)
    %paste the edited partial matrix back
    C(rows,:)=C_part;

    CP{q}=C_part;
  end

but still i am getting Index exceeds matrix dimensions.

Error in cluster_rows2 (line 15) C_part(non_0 & C_part==0)=A_part(non_0 & C_part==0)

Error in(line 22) [C,C_part]=cluster_rows2(A,B,rows); Could you please help me to solve it.

[Jan]

Did you read my answer? Avoid the globals, because they are a shot in your knee. Omit unneeded code, because it is just confusing and time consuming clutter: "CP{q}=C_part" is not needed inside "cluster_rows2", because it is not replied as output. Therefore you do not need q inside this function and then there is no need for a global at all.

For the indexing problem: Use the debugger to find out, why the index exceeds the array dimension.

[Prabha Kumaresan]

If i am avoiding 
global q
q=1
EP{q}=E_part
    PP{q}=P_part
    UP{q}=U_part
    VP{q}=V_part
%      %indxing all cells/groups
        q=q+1
All_EP=[cat(1, EP{:})]
All_PP=[cat(1, PP{:})]
All_UP=[cat(1, UP{:})]
All_VP=[cat(1, VP{:})]

The coding executes without error. but I want to vertically concatenate all E_part with respect to t and r.Could you tell me how it can be done.

[Jan]

What is the purpose of the square brackets in this code:

All_EP=[cat(1, EP{:})]

Where are the "E_part"s coming from? I do not find this in your code.

? [] concatenates the included elements to a matrix, but here you provide one element only. So omit the useless brackets:

All_EP = cat(1, EP{:});

Now you ask for "vertically concatenate all E_part with respect to t and r". I do not know what "respect to t and r" means here. But "cat(1, EP{:})" looks good for a vertical concatenation. What is the problem with this command?

[Prabha Kumaresan]

sorry it was my mistake its not E_part its C_part I want to vertically concatenate with respect to m=[10 20 30 40 50] n=[60 70 80 90 100] which means all C_part of (10,60) together (10,70),(10,80),(10,90),(10,100) followed by (20,60),(20,70),(20,80),(20,90),(20,100),.......

Answer 2

[Torsten]

Initialize q to a value (e.g. 1) at the beginning of your code.

Best wishes

Torsten