Is there a way to do a by row comparison of two data sets with the same number of columns, different number of rows, and different parameters for each column?

I have two data sets and I want to compare them each row in one set to every row in the other with different comparison parameters for each column. Example: If the comparison is that 1a=1b if 1b is in the range of 1a+/-0.0001 and 2a=2b if 2b is the range of 2a +/- 1. Where 1 or 2 is the column number and a or b notates the data set.

&nbsp

So the entries in row 1 are equal in both data sets and row 2 in data set A is equal to row 3 in data set B.

[per isakson]

Did you try Lia = ismember(A,B,'rows')?

[JChemPhD]

It doesn't allow me to set a specific tolerance.

Answer 1

[Walter Roberson]

http://www.mathworks.com/matlabcentral/answers/223612#answer_182595

[JChemPhD]

That works for the comparison by row but it still doesn't allow me to set a different tolerance for each column.

Answer 2

[Thorsten]

A = [1.0001 2.0002;
     8.1234 5.0];
B = [1.0002 3.0;
     8.1235 3.0
     8.1233 6.0];

tol = [0.001 1];

nB = size(B, 1);
R = nan(size(A,1), nB);

Tol = repmat(tol, [nB 1]); 
for i = 1:size(A,1)
   R(i,:) = all(abs(B - repmat(A(i,:), [nB 1])) <= Tol, 2);
end