Slice a large 2D matrix into a 3D one -- error

Hi, I'm trying to extract a series of 32x32 matrices of a large original 6601x32 matrix, and save them into a 3D (32x32x200) one. Each matrix is a sample I've logged from a scanning device (200 in total). I'm having this error but I couldn't find out why it happens or how to fix it.

    % Index exceeds matrix dimensions.

This is the part of my program that throws the error:

 A = cell2mat(raw);
[m,n] = size(A);
% Create 200 Frames 32x32
framesize = 32 % Number of Rows and Columns of Each Frame
Frame3d = zeros(32,32,200); % Prealocate Frames 
% % Now scan A, getting each frame and putting it as a slice of a 3D array.
framenumber=1;
for row = 1 : 33 : m
        row1 = row + 1; % framesizer = 32, and row1 = row +1 to avoid first row 
        row2 = row1 + framesize - 1;
        Frame3d(:,:,framenumber) = A(row1:row2,:); % Assign the submatrix to the slice --> Here throws the error
        framenumber = framenumber + 1; 
 end

Basically I cannot figure it out.

Kind regards.

Answer 1

[Image Analyst]

They are not the same size:

>> 6601*32
ans =
      211232
>> 32*32*200
ans =
      204800

Why do you think the reshaped matrix should contain all of the input data?

Try this:

% Make up sample data.
frame2d = randi(255, 6601,32, 'uint8');
% Preallocate output data.
frame3d = randi(255, 32,32,200, 'uint8');
% Transfer 32x32 chunks to output array.
slice = 1;
for row = 1 : 32 : size(frame2d, 1) - 31
  frame3d(:, :, slice) = frame2d(row:row+31, :);
  slice = slice + 1;
end

[rookie]

Thanks a lot! I've should have explained my problem a little better.. I did not say that had to remove one row from each frame which had some metadata like the time..etc. But thanks to your comment I realized that when I imported the data matrix it was adding an end row with NAN in the fields, so I removed it.

Then the 2D row size was 6600 and not 6601, and removing the metadata rows I was able to create the 3D matrix with all the slices.

Cheers