Extract elemnts from a matrix and put them in a new one

Hello everyone,

I'm looking for a command or an algorithm able to solve the following problem:

I have a matrix M in which there are many zeros with decimal values and few elements which are higher than 0.99 such as this matrix:

M=[ 0.45 5.45 0.64 0.78 0.32

3.65 0.79 0.56 0.23 0.11

0.21 6.78 0.55 3.45 0.57

0.67 9.54 5.44 9.00 3.23]

I would like to make a new matrix N by extracting the elements of matrix M which are larger than 1 with a certain arrangement like this:

N=[ 3.65 5.45 5.44 3.45 3.23

NaN 6.78 NaN 9.00 NaN

NaN 9.54 NaN NaN NaN]

Thanks for your attention :)

[Matt J]

Are you going to explain to us what the pattern of NaNs is supposed to be in the output N?

Answer 1

[Image Analyst]

Try this:

M=[ 0.45  5.45  0.64  0.78  0.32
	3.65  0.79  0.56  0.23  0.11
	0.21  6.78  0.55  3.45  0.57
	0.67  9.54  5.44  9.00  3.23]
[rows, columns] = size(M)
MOut = nan(size(M));
for col = 1 : columns
	thisCol = M(:, col)
	moreThan1 = thisCol(thisCol >= 1)
	for row = 1 : length(moreThan1)
		MOut(row, col) = moreThan1(row);
	end
end
% Get rid of any all non rows
rowsToDelete = all(isnan(MOut), 2);
MOut(rowsToDelete, :) = []

You get

MOut =

3.65 5.45 5.44 3.45 3.23

NaN 6.78 NaN 9 NaN

NaN 9.54 NaN NaN NaN

If it's not your homework assignment, you can use this code.

Answer 2

[Matt J]

Using the attached file sortlidx.m

M=[    0.45  5.45  0.64  0.78  0.32
       3.65  0.79  0.56  0.23  0.11
       0.21  6.78  0.55  3.45  0.57
       0.67  9.54  5.44  9.00  3.23];

N=M; 
N(N<1)=nan;

[~,is]=sortlidx(isnan(N));

N=N(is);
N(all(isnan(N),2),:)=[]

N = 3×5

3.6500 5.4500 5.4400 3.4500 3.2300 NaN 6.7800 NaN 9.0000 NaN NaN 9.5400 NaN NaN NaN