minimum of different sized cell arrays using cell2mat error

How do i find the overall minimum of a cell array, cell2mat gives error.

Y= {[100 200] [50 100] [20] [30 140];
      [10 130] [40]  [60  200]  [30]};

min(cell2mat(Y))

Error using cat
Dimensions of matrices being concatenated are not consistent.

Error in cell2mat (line 78)
        m = cat(1,m{:});

Answer 1

[Jos (10584)]

cell2mat fails because not all cells of Y are the same size. But you do not need this step to find the minimum:

Y= {[100 200] [50 100] [20] [30 140];
      [10 130] [40]  [60  200]  [30]};
minC = cellfun(@(c) min([c(:) ; Inf]), Y) % get the minimum of each cell, added Inf to deal with empty cells
minY = min(minC(:)) % get the global minimum

[Joseph Lee]

Thanks, that helped remove this additional step i used

Y = Y(~cellfun(@isempty, Y))

Answer 2

[Stephen23]

>> min([Y{:}])
ans =  10

[Jos (10584)]

Nice. I thought of that too, but it will fail when some cells of Y are not row-vectors.

Answer 3

[Jan]

The fast C-Mex function FEX: Cell2Vec converts the contents of the cell to a vector independent from the shapes of the cell elements:

Y = {[100 200] [50 100] [20] [30 140]; ...
     [10 130] [40]  [60  200]  [30]};
C = Cell2Vec(Y);
min(C)

With FEX: MinMaxElem you can search the minimal and maximal values also:

[Min, Max, MinIndex, MaxIndex, MinArg, MaxArg] = MinMaxElem(Y{:})

This searches the max and the indices also. NOTE: This is not faster than min and max of modern Matlab versions anymore. But perhaps more convenient for your case.