Take values greater than 0 excluding NaN

Hi all, I have a matrix A, say, like this:

NaN	NaN	1
NaN	-2	-3
NaN	NaN	NaN
2	NaN	-3
NaN	-3	-3
1	-1	2

and would like to make a dummy matrix out of A having values 1 for values that are greater than 0 but ignoring missing values. In other words the expected result is:

NaN	NaN	1
NaN	0	0
NaN	NaN	NaN
1	NaN	 0
NaN	0	0
1	0   1

I tried this:

A=A(~isnan(A))>0

but it provides a vector while I would like a matrix of the same dimensions as A.

Thank you

Answer 1

[Steven Lord]

A = [NaN	NaN	1;
NaN	-2	-3;
NaN	NaN	NaN;
2	NaN	-3;
NaN	-3	-3;
1	-1	0] % Adding an explicit 0

A = 6×3

NaN NaN 1 NaN -2 -3 NaN NaN NaN 2 NaN -3 NaN -3 -3 1 -1 0

B = A;
B(B <= 0) = 0;
B(B > 0) = 1

B = 6×3

NaN NaN 1 NaN 0 0 NaN NaN NaN 1 NaN 0 NaN 0 0 1 0 0

Or:

C = discretize(A, [-Inf, eps(0), Inf], [0, 1])

C = 6×3

NaN NaN 1 NaN 0 0 NaN NaN NaN 1 NaN 0 NaN 0 0 1 0 0

For this last approach anything in the interval [-Inf, eps(0)) (which includes 0; the fact that the right edge is not included is why I used eps(0) instead of 0) becomes 0. Anything in the interval [eps(0), Inf] (this last bin includes both left and right edges) becomes 1. Anything in neither of those intervals (aka NaN) becomes NaN.

If you want an explicit 0 in A to be a 1, use 0 instead of eps(0) in the second input.