How to find the mean of adjacent rows in a matrix?

Question

I am trying to find the mean of each adjacent rows in a matrix:

x=
[
1 2 3 4;
2 3 1 0;
1 0 1 0;
3 1 0 1;
]

Simply put,I need a resulting matrix which calculates the mean of row 1 and 2, and then 3 and 4, which is like the following:

y=
[
2;
0.875
] 

How do I go about this?

Answer 1

You can calculate the mean of rows and than the average of two consecutive rows

   z=mean(x,2)
z =

   2.50000
   1.50000
   0.50000
   1.25000

octave:22> y=((z(1:2:end)+z(2:2:end))/2)
y =

   2.00000
   0.87500

Answer 2

You can reshape your array and put each other row onto the same row:

octave:6> [n,m] = size(x);
octave:7> mean(reshape(x.', m*2, n/2))
ans =

   2.00000   0.87500

Due to the transpose the result is a row vector, you can just transpose it back if you really want to have a column vector.

On new enough MATLAB (that I don't have access to) you should be able to use multiple input arguments for the dimension parameter of mean , allowing you to utilize multidimensional arrays in this case:

[n,m] = size(x);
mean(reshape(x, [2, n/2, m]), [1, 3]) % should leave n/2-sized dimension

For obvious reasons the above isn't tested.