Help with the zeros function

I am well aware that we can easily delete arrays of a vector using the [ ] function.

I however need to replace certain rows/ columns of my vector with zeros.

So how do I go on about doing that? Its really confusing because 'zeros' is mainly used to make a matrix full of zeros.

Also. Once we have that row/ column replaced with zero, how would we get rid of it without using [ ]. I know [ ] would be ideal and more efficient but i am trying to work with the zeros function specifically.

for example:

A = [1 2 3 4
     9 1 2 2
     3 4 2 2]

B = (length(zeros(:,3)))<-- I know this is wrong but something like this should work. 

>>>B = 

>>>1 2 3 4

   0 0 0 0

   3 4 2 2

C = 1 2 3 4

    3 4 2 2

Then we remove the row we replaced. So we basically need to delete the row we have saved in variable B.

How would we do it?

Thanks

[Matt J]

Duplicate thread,

http://www.mathworks.com/matlabcentral/answers/243877-how-to-delete-a-submatrix-from-a-matrix-using-the-zeros-command-instead-of-square-brackets

Answer 1

[Walter Roberson]

B(2,:) = zeros(1,size(B,2));

Or just

B(2,:) = 0;

To get rid of rows that are all zero (even if not specifically written over but just happen to have that value):

B = B(any(B,2),:);

Answer 2

[Image Analyst]

Try this:

   B = A; % Initialize
   B(2,:)=0 % Set all elements of row 2 to 0

   % Make C
   C=B([1,3],:)

[Tab for Lab]

Thank You so much. Just another question. Is the zeros function the same as making rows eqivalent to 0 like you did?

[Walter Roberson]

B(2,:) = zeros(1,size(B,2)); is the same as B(2,:)=0 for all practical purposes, but is slightly slower than B(2,:)=0 . You would typically use zeros() when you are allocating a block of memory at execution time that has to be all 0. For example,

C = zeros(3,5);

allocates a 3 x 5 block of double precision 0's and assigns it to C, either defining that variable for the first time or overwriting its value completely. On the other hand

C = 0;

would only assign a single double precision 0 to C, either creating C for the first time or overwriting its value completely. And

C(:) = 0;

would write a 0 of C's current class to all the elements of C, retaining C's current shape and using the same block of memory that it already has, with it being an error if C does not already exist.

[Tab for Lab]

Thanks buddy.

cheers.