generate a digraph using a cell array and an array

I'm trying to generate a digraph that shows the mapping between the array [1:10] to this cell array:

1×10 cell array

Columns 1 through 6

{[1 5 7]} {[2 3 4 5 7]} {[1 2 3 5 6 7 8]} {[2 4 5 7 9]} {[1 3 6 7 9]} {[2 4 6 8 9 10]}

Columns 7 through 10

{[1 7 9 10]} {[1 3 4 5 6 9 10]} {[1 3 4 5 8 10]} {[5 9 10]}

So 1 maps to 1, 5 and 7; 2 maps to 2, 3, 4, 5, and 7 and so on. Is there a way to do that?

Thank you.

Answer 1

[Cris LaPierre]

I think you want to use this syntas: G = digraph(s,t)

This example is helpful: https://www.mathworks.com/help/matlab/ref/digraph.html#mw_26035adf-ff90-4a33-a8f8-42048d7e39a6

You need an element for each edge in s and t. You can achieve that with a little creativity

e = [{[1 5 7]}    {[2 3 4 5 7]}    {[1 2 3 5 6 7 8]}    {[2 4 5 7 9]}    {[1 3 6 7 9]}    {[2 4 6 8 9 10]} ...
    {[1 7 9 10]}    {[1 3 4 5 6 9 10]}    {[1 3 4 5 8 10]}    {[5 9 10]}]

e = 1×10 cell array
    {[1 5 7]}    {[2 3 4 5 7]}    {[1 2 3 5 6 7 8]}    {[2 4 5 7 9]}    {[1 3 6 7 9]}    {[2 4 6 8 9 10]}    {[1 7 9 10]}    {[1 3 4 5 6 9 10]}    {[1 3 4 5 8 10]}    {[5 9 10]}

% Determine how many elements are in each cell
n = cellfun(@numel,e)

n = 1×10

3 5 7 5 5 6 4 7 6 3

% Extract all values from the cell array to a target vector, t
t = [e{:}]

t = 1×51

1 5 7 2 3 4 5 7 1 2 3 5 6 7 8 2 4 5 7 9 1 3 6 7 9 2 4 6 8 9

% use cumsum to find indices of first value from each cell 
idx = cumsum([1,n(1:end-1)])

idx = 1×10

1 4 9 16 21 26 32 36 43 49

% Create s as a source vector of zeros the same size as t
s=zeros(size(t))

s = 1×51

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

% Set index corresponding to fist cell value to 1
s(idx) = 1

s = 1×51

1 0 0 1 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0

% Use cumsum to create a vector of sources that aligns with t
s = cumsum(s)

s = 1×51

1 1 1 2 2 2 2 2 3 3 3 3 3 3 3 4 4 4 4 4 5 5 5 5 5 6 6 6 6 6

% Create digraph
G = digraph(s,t);
plot(G,'Layout','force')

[Akira Agata]

+1

Another possible solution:

t ={...
  [1 5 7], [2 3 4 5 7], ...
  [1 2 3 5 6 7 8], [2 4 5 7 9], ...
  [1 3 6 7 9], [2 4 6 8 9 10], ...
  [1 7 9 10], [1 3 4 5 6 9 10], ...
  [1 3 4 5 8 10], [5 9 10]};

% Create adjacency matrix
A = zeros(10);
for kk = 1:10
  A(kk, t{kk}) = 1;
end

% Create graph object
G = digraph(A);

% Visualize the result
figure
plot(G, "Layout", "circle")

[Steven Lord]

Another solution uses repelem:

e = [{[1 5 7]}    {[2 3 4 5 7]}    {[1 2 3 5 6 7 8]}    {[2 4 5 7 9]}    {[1 3 6 7 9]}    {[2 4 6 8 9 10]} ...
    {[1 7 9 10]}    {[1 3 4 5 6 9 10]}    {[1 3 4 5 8 10]}    {[5 9 10]}];
% Determine how many elements are in each cell
n = cellfun(@numel,e);
% Extract all values from the cell array to a target vector, t
t = [e{:}];

% The new part
s = repelem(1:numel(e), n)

s = 1×51

1 1 1 2 2 2 2 2 3 3 3 3 3 3 3 4 4 4 4 4 5 5 5 5 5 6 6 6 6 6

G = digraph(s,t);
plot(G,'Layout','force')

[Cris LaPierre]

+1 That's much cleaner.