Getting the mean from each cell row in a cell array

I have a cell array called data. This is the first row.

A = {5×1 double}, B = {4×1 double}, C = {7×1 double}

I want to get the mean of each row from all cells. For example, if we cell2mat each cell we get

A = 1           B = 2          C = 3    
    2               4              6
    3               6              9
    4               8              1
    5                              5
                                   6
                                   8

I want this output, D:

D = (1 + 2 + 3)/3 = 2
    (2 + 4 + 6)/3 = 4
    (3 + 6 + 9)/3 = 6
    (4 + 8 + 1)/3 = 4.3
    (5 + 5)/2 =     5
                    6
                    8

I tried converting the cells into matrices and then combining them.

[rw,cs] = size(data);
matrix = [];
for x = 1:rw
    for y = 1:cs
        matrix = cell2mat(data(x, y));
        matrix(n) = data;
    end
end

This returns an error that: Unable to perform assignment because the indices on the left side are not compatible with the size of the right side.

Answer 1

[DGM]

I'm sure there are more elegant ways than this, but eh.

celery = {(1:5).' (2:2:8).' [3 6 9 1 5 6 8].'}

celery = 1×3 cell array
    {5×1 double}    {4×1 double}    {7×1 double}

n = cellfun(@numel,celery);
D = nan(max(n),numel(celery));
for k = 1:numel(celery)
    D(1:n(k),k) = celery{k};
end
Dm = mean(D,2,'omitnan')

Dm = 7×1

2.0000 4.0000 6.0000 4.3333 5.0000 6.0000 8.0000

[Sheung Man Keung]

Thanks for your response. However, when I tried to extend your code to a cell array with more than 1 row, it runs into a problem. It takes the max of the longest cell array instead of the max in each row.

celery = {(1:5).' (2:2:8).' [3 6 9 1 5 6 8].'};
carrot = {(1:10).' (3:3:24).' [1 3 9 1 8 1].'};
celery = [celery; carrot]         

celery = 2×3 cell array
    { 5×1 double}    {4×1 double}    {7×1 double}
    {10×1 double}    {8×1 double}    {6×1 double}

Dm = {};
n = cellfun(@numel,celery);
[rows, cols] = size(n);
for i = 1:rows
    max_value = max(n(i, :));
    test_matrix = nan(max_value, numel(celery));
    for k = 1:numel(celery)
        test_matrix(1:n(k), k) = celery{k};   
    end
    Dm{i} = mean(test_matrix, 2, 'omitnan');
end
Dm{1}, Dm{2}

ans = 10×1

1.8333 3.8333 6.5000 5.0000 7.6000 7.7500 12.0000 5.3333 1.5000 1.6667

ans = 10×1

1.8333 3.8333 6.5000 5.0000 7.6000 7.7500 12.0000 16.0000 9.0000 10.0000

[DGM]

Maybe this is closer

celery = {(1:5).' (2:2:8).' [3 6 9 1 5 6 8].'};
carrot = {(1:10).' (3:3:24).' [1 3 9 1 8 1].'};
celery = [celery; carrot]         

celery = 2×3 cell array
    { 5×1 double}    {4×1 double}    {7×1 double}
    {10×1 double}    {8×1 double}    {6×1 double}

allveclen = cellfun(@numel,celery);
longestvector = max(allveclen,[],2);

Dm = cell(size(celery,1),1);
for i = 1:size(celery,1)
    test_matrix = nan(longestvector(i),size(celery,2));
    for k = 1:size(celery,2)
        test_matrix(1:allveclen(i,k), k) = celery{i,k};   
    end
    Dm{i} = mean(test_matrix, 2, 'omitnan');
end

Dm{1}, Dm{2}

ans = 7×1

2.0000 4.0000 6.0000 4.3333 5.0000 6.0000 8.0000

ans = 10×1

1.6667 3.6667 7.0000 5.6667 9.3333 8.3333 14.0000 16.0000 9.0000 10.0000

[Sheung Man Keung]

Thanks for your expertise!