Unique Function based on 2 columns [Instead of rows]

Suppose, I have a cell array, a, with contents as such:

a{1}=[1 3 4 5;
      3 3 4 5;
      5 5 4 5
      2 4 2 6;
      6 5 2 6
      7 2 3 1;]

How can I apply the 'unique' function on 2 columns [column 3 and 4] such that they will return the value '3' and '2'. [Since there are 3 duplicates for the pair 4,5 and 2 duplicates for the pair 2,6.]

Any hint is greatly appreciated.

[Cedric]

See the edit in my answer.

[Cedric]

See 2nd edit.

Answer 1

[Cedric]

NEW ANSWER AFTER 2ND EDIT

You could do it as follows (you might want to fine tune it):

 diff(find([true; any(diff(a{1}(:,3:4)),2); true]))

To understand this solution, evaluate it by part:

 >> a{1}(:,3:4)                   % Index col 3 and 4.
 ans =
     4     5
     4     5
     4     5
     2     6
     2     6
     3     1

 >> diff(a{1}(:,3:4))             % Difference between rows in col 3 and 4.
 ans =
     0     0
     0     0
    -2     1
     0     0
     1    -5

 >> any(diff(a{1}(:,3:4)),2)      % Flag rows were there is a diff in either col.
 ans =
     0
     0
     1
     0
     1

 >> [true; any(diff(a{1}(:,3:4)),2); true]        % Add flags for boundaries.
 ans =
     1
     0
     0
     1
     0
     1
     1

 >> find([true; any(diff(a{1}(:,3:4)),2); true])  % Get positions of all flags.
 ans =
     1
     4
     6
     7

Finally, compute positions differences.

 >> diff(find([true; any(diff(a{1}(:,3:4)),2); true]))
 ans =
     3
     2
     1

which give you the size (in terms of number of rows) of each block of similar cols 3 and 4.

FORMER ANSWER

 >> [u, ia] = unique(a{1}(:,3:4), 'rows', 'stable')
 u =
     4     5
 ia =
     1

The flag 'rows' tests rows uniqueness (taking all columns into account). So here it seems that you want to pass columns 3 and 4 of array a{1} to UNIQUE, using the 'rows' flag. The stable flag makes unique return the first occurrence instead of the last in this case.

If you wanted to operate on columns (for all rows), you could pass a subset of the transpose of a{1} to UNIQUE.

EDIT (after you edited the question): remove the 'stable' flag if you want 4.

[RDG]

Thank you Cedric, but it seems that my question was a poorly structured.

[RDG]

Thanks again Cedric for including in the explanation. Been having a hard time translating my ideas into code.

Answer 2

[Azzi Abdelmalek]

b=a{1};
out=all(~(any(diff(b(:,3:4)))))

[RDG]

What if I want the value to return '4'? [Since there are 4 duplicates]