Accessing data from cells inside of cells

So I have the code below, which solves an ODE 1000 times with random variables at 10 different intervals (tspans). The result I end up with a 1000x1 cell array of 10x1 cell arrays. Each entry of the 10x1 cell arrays represents a 41x2 numeric array, which is the solution for x1 and x2 in each of the ten tspans. What I would like to end up with is a 410x1000 numeric array, where each column represents ten column 2 entries (x2 solutions) from the 41x2 numeric arrays per each 10x1 cell. Is this possible? Thanks in advance.

n = 1000;
result = cell(n,1);
for k=1:n
tspan1=[0 1];A01=rand;P01=rand;g1=rand;p1=rand;B1=rand;
[t1,x1] = ode45(@(t1,x1) [-g1*x1(1) + p1*x1(1); -x1(1)*x1(2)+ B1*x1(2)], tspan1, [A01 P01]);

tspan2=[1 2];A02=rand;P02=rand;g2=rand;p2=rand;B2=rand;
[t2,x2] = ode45(@(t2,x2) [-g2*x2(1) + p2*x2(1); -x2(1)*x2(2)+ B2*x2(2)], tspan2, [A02 P02]);

tspan3=[2 3];A03=rand;P03=rand;g3=rand;p3=rand;B3=rand;
[t3,x3] = ode45(@(t3,x3) [-g3*x3(1) + p3*x3(1); -x3(1)*x3(2)+ B3*x3(2)], tspan3, [A03 P03]);

tspan4=[3 4];A04=rand;P04=rand;g4=rand;p4=rand;B4=rand;
[t4,x4] = ode45(@(t4,x4) [-g4*x4(1) + p4*x4(1); -x4(1)*x4(2)+ B4*x4(2)], tspan4, [A04 P04]);

tspan5=[4 5];A05=rand;P05=rand;g5=rand;p5=rand;B5=rand;
[t5,x5] = ode45(@(t5,x5) [-g5*x5(1) + p5*x5(1); -x5(1)*x5(2)+ B5*x5(2)], tspan5, [A05 P05]);

tspan6=[5 6];A06=rand;P06=rand;g6=rand;p6=rand;B6=rand;
[t6,x6] = ode45(@(t6,x6) [-g6*x6(1) + p6*x6(1); -x6(1)*x6(2)+ B6*x6(2)], tspan6, [A06 P06]);

tspan7=[6 7];A07=rand;P07=rand;g7=rand;p7=rand;B7=rand;
[t7,x7] = ode45(@(t7,x7) [-g7*x7(1) + p7*x7(1); -x7(1)*x7(2)+ B7*x7(2)], tspan7, [A07 P07]);

tspan8=[7 8];A08=rand;P08=rand;g8=rand;p8=rand;B8=rand;
[t8,x8] = ode45(@(t8,x8) [-g8*x8(1) + p8*x8(1); -x8(1)*x8(2)+ B8*x8(2)], tspan8, [A08 P08]);

tspan9=[8 9];A09=rand;P09=rand;g9=rand;p9=rand;B9=rand;
[t9,x9] = ode45(@(t9,x9) [-g9*x9(1) + p9*x9(1); -x9(1)*x9(2)+ B9*x9(2)], tspan9, [A09 P09]);

tspan10=[9 10];A10=rand;P10=rand;g10=rand;p10=rand;B10=rand;
[t10,x10] = ode45(@(t10,x10) [-g10*x10(1) + p10*x10(1); -x10(1)*x10(2)+ B10*x10(2)], tspan10, [A10 P10]);
x={x1;x2;x3;x4;x5;x6;x7;x8;x9;x10};    
result{k} = x;
end

Answer 1

[Star Strider]

There is no guarantee that the row lengths of each ‘x’ are going to be the same using a two-element ‘tspan’ vector, so what you outlined may not be possible.

However if you fix the number of elements in ‘tspan’, what you want to do is straightforward:

n = 1000;
tsv = linspace(0, 1, 50);
xtm = zeros(numel(tsv)*10,2,n);
for k1 = 1:n
    xim = zeros(9*numel(tsv), 2);
    for k2 = 1:10
        tspan=tsv+(k2-1);
        A=rand; P=rand; g=rand; p=rand; B=rand;
        [t,x] = ode45(@(t,x) [-g*x(1) + p*x(1); -x(1)*x(2)+ B*x(2)], tspan, [A P]);
        xim((1:numel(tsv))+numel(tsv)*(k2-1),:) = x;
    end
    xtm(:,:,k1) = xim;
end

The ‘xtm’ matrix will be (500x2x1000), representing the concatenated rows, the two columns, and the 1000 trials. It is also easy to index into.

[Thomas Veith]

Thank you so much! Very helpful!

[Star Strider]

As always, my pleasure!