Working with cell arrays

10 views (last 30 days)
L'O.G. on 5 May 2023
Commented: Stephen23 on 8 May 2023
I would like to separate columns in an array based on the unique values of the first column. I think the way to do this is a cell array since the vectors would be of different length. Is there a way to vectorize this? For example, the following array
example = [
5 2.3 1.7
5 0.2 9.1
99 3.3 3.6
99 4.1 3.2
99 5.1 2.2
103 1.1 6.8
];
should give
A{1} = [2.3,0.2];
A{2} = [3.3,4.1,5.1];
A{3} = [1.1];
B{1} = [1.7,9.1];
B{2} = [3.6,3.2,2.2];
B{3} = [6.8];
(since there are three unique values in column 1 of the example)

Paul on 5 May 2023
example = [
5 2.3 1.7
5 0.2 9.1
99 3.3 3.6
99 4.1 3.2
99 5.1 2.2
103 1.1 6.8
];
G = findgroups(example(:,1));
A = splitapply(@(x) mat2cell(x,numel(x)),example(:,2),G);
B = splitapply(@(x) mat2cell(x,numel(x)),example(:,3),G);
A,B
A = 3×1 cell array
{2×1 double} {3×1 double} {[ 1.1000]}
B = 3×1 cell array
{2×1 double} {3×1 double} {[ 6.8000]}
A{1}
ans = 2×1
2.3000 0.2000
B{1}
ans = 2×1
1.7000 9.1000
Stephen23 on 8 May 2023
Also without the MAT2CELL call:
X = [
5 2.3 1.7
5 0.2 9.1
99 3.3 3.6
99 4.1 3.2
99 5.1 2.2
103 1.1 6.8
];
G = findgroups(X(:,1));
A = accumarray(G,X(:,2),[],@(a){a});
B = accumarray(G,X(:,3),[],@(a){a});
A,B
A = 3×1 cell array
{2×1 double} {3×1 double} {[ 1.1000]}
B = 3×1 cell array
{2×1 double} {3×1 double} {[ 6.8000]}
A{1}
ans = 2×1
2.3000 0.2000
B{1}
ans = 2×1
1.7000 9.1000

Categories

Find more on Multidimensional Arrays in Help Center and File Exchange

R2021b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!