Select a row and append it at the end of a matrix | Move a selected row to end of matrix

Hello People,

Assume i have a matrix H =

[1 0 0 1

0 1 0 0

1 1 1 0

1 1 0 1

0 1 0 1

0 0 1 1]

Now I want to move row number 2 and 4 to the end of the matrix. NOT SWAP.

Like removing row 2 & row 4 and appending them after row 6. The remaining rows cascade upward in the same order.

I need pointers on how to achieve this. I unknowingly used swap and realised that it was the reason my algorithm failed. Any help would be appreciated.

Thank You

[KALYAN ACHARJYA]

data=H([1,3,5,6,2,4],:)

Answer 1

[Matt J]

 
H=[1    0    0    1
 0    1    0    0
 1    1    1    0
 1    1    0    1
 0    1    0    1
 0    0    1    1];

rows2move=[2,4];
idx=[setdiff(1:size(H,1), rows2move), rows2move];

H=H(idx,:)

H = 6×4

1 0 0 1 1 1 1 0 0 1 0 1 0 0 1 1 0 1 0 0 1 1 0 1

●

[Rishi Balasubramanian]

So I am trying to adapt this to my code, but it says dimensions are not consistent. I am attaching the actual matrix (93x155) and the code which I am trying to generalise. Could you help me on it?

[m, n] = size(H);

for i = 1:n
    a = sum(H);
end

amax = max(a);
amin = min(a);

if (amax == amin)   %select random col if there is no column with min ones
    c1 = n;
else (amax > amin)  %select the col with the min ones
    [r,c] = find(a==min(a));
    c1 = c;
end

%identify rows containing 1's in selected column.
 rows = find(H(:,c1)==1)
 
 %I HAVE FOUND THE ROWS WITH ONES IN IT, I WANT TO MOVE THOSE ROWS TO BOTTOM, HOW DO I DO IT?
 
 %%%%%%%%% UPDATE %%%%%%%%%%
 %SO I tried incorporating it into the code using loops. I feel like I almost got it, but it says index exceeds dimensions
 for i = 1:size(rows)
idx=[setdiff(1:size(H,1), rows(i)), rows(i)];
H = H(idx(i), :);
 end
 
 %COuld you say if this is right?

So the 'rows' will consist of the indexes of the rows that i want to move to after the end of the matrix, i.e. those rows must come after row 155, and the rest of the rows mustn't change except moving up to fill the void. Thanks in advance for your help.

[Matt J]

That's even easier. The whole operation should just be 2 lines. No loops:

  rows = H(:,c1)==1;

  H = [H(~rows, :); H(rows,:)];

[Rishi Balasubramanian]

I'm getting a matrix thats just 3x155.

Now I need these 3 rows at the end of the matrix. How do I do that?

I mean, this is really close, it exactly got the 3 selected rows, now it just needs to be appended to the end of the matrix with the remaining rows.

[Matt J]

I'm getting a matrix thats just 3x155.

Maybe you're working with a partially edited version. I get a 93x155 result:

load(websave('PCMatrix','https://www.mathworks.com/matlabcentral/answers/uploaded_files/465005/PCMatrix.mat'));

[m, n] = size(H);

for i = 1:n
    a = sum(H);
end

amax = max(a);
amin = min(a);

if (amax == amin)   %select random col if there is no column with min ones
    c1 = n;
else (amax > amin)  %select the col with the min ones
    [r,c] = find(a==min(a));
    c1 = c;
end

  rows = H(:,c1)==1;

  H = [H(~rows, :); H(rows,:)];
 
whos H

  Name       Size              Bytes  Class     Attributes

  H         93x155            115320  double              

[Rishi Balasubramanian]

Yup, got the answer. Thanks a ton man.

[Rishi Balasubramanian]

But what I don't understand is this. In my version, i used FIND function to find the indexes of the rows that have 1s in them.

This is your version

rows = H(:,c1)==1;

I don't understand how it works. It works. But i can't figure out how and why. Could you explain it, as I will be using many versions of this across my algorithm. It'll be really helpful.