How to Create Multidimensional table in for loop
5 Ansichten (letzte 30 Tage)
Ältere Kommentare anzeigen
Karthi Ramachandran
am 3 Apr. 2019
Kommentiert: Karthi Ramachandran
am 9 Apr. 2019
Trajectory =1:15;
%TrajDetail = zeros(328,4,15);
for i = 1:length(Trajectory)
TSPANtf = [ 0 t(i) ] ;
%TSPANtf2 = [ 0 t_corrected_2 ] ;
RelTol = 3.e-14 ; AbsTol = 1.e-16;
Options = odeset('RelTol',3.e-14,'AbsTol',1.e-16);
[tt,xx] = ode113(@(t,x) CRes3BP_xy(t,x,mu),TSPANtf,x0_corrected(i,1:4)',Options);
plot(xx(:,1),xx(:,2),'k',xx(:,1),-xx(:,2),'k')
pause(0.01) ;
hold on
axis equal
pos_x = xx(:,1);
pos_y = xx(:,2);
vel_x = xx(:,3);
vel_y = xx(:,4);
TrajDetail (i)= table(pos_x,pos_y,vel_x,vel_y); %here the comes as (see below the italised stuff)
end
% I have not taken the tt values out but still i need
what I wanted to do is get the values of xx and tt for each iteration and store in a table named TrajDetail 1,TrajDetail 2 ...TrajDetail 15.
Note that xx changes the row size according to "tt"
eg: Iteration 1 may cause xx to have 160 x 4 ; Iteration two may have 200 x 4 etc
table should become like Table1 (ieTrajDetail 1) containg
tt;pos_x;pos_y:vel_x;vel_y and has to be done for each iteration
Error for above
Subscripting a table using linear indexing (one subscript) or multidimensional indexing (three or more subscripts) is not supported. Use a row subscript and a variable
subscript.
1 Kommentar
Rik
am 3 Apr. 2019
Tables are 2D in Matlab, containing numbered rows and named columns. If you want something else, you can't use a table. You could use a double array to create a single large multidimensional array, or encapsulate it in a cell array.
Akzeptierte Antwort
Rik
am 3 Apr. 2019
Bearbeitet: Rik
am 3 Apr. 2019
A bold guess based on your commented code:
Trajectory =1:15;
TrajDetail = zeros(1,4,15);
for i = 1:length(Trajectory)
TSPANtf = [ 0 t(i) ] ;
%TSPANtf2 = [ 0 t_corrected_2 ] ;
RelTol = 3.e-14 ; AbsTol = 1.e-16;
Options = odeset('RelTol',3.e-14,'AbsTol',1.e-16);
[tt,xx] = ode113(@(t,x) CRes3BP_xy(t,x,mu),TSPANtf,x0_corrected(i,1:4)',Options);
plot(xx(:,1),xx(:,2),'k',xx(:,1),-xx(:,2),'k')
pause(0.01) ;
hold on
axis equal
TrajDetail (1:size(xx,1),:,i)=xx(:,1:4);
end
8 Kommentare
Peter Perkins
am 9 Apr. 2019
Rik is correct but perhaps this needs some more explanation.
Tables contain "variables", which need not be column vectors (that's why they are referred to as "variables" and not as "columns"). So if your data are 3D, NxMxP say, one way to store them is in an NxM table, each variable of which is itself NxP. Or as an NxP table with NxM variables.
Another way to store NxMxP data in a table is by flattening to 2-D as an (N*M)xP table, and then add an extra "indicator variable" to keep track of what 2nd dim slice each row corresponds to.
And finally, if all your data are numeric, just store as an NxMxP numeric matrix.
Weitere Antworten (1)
Guillaume
am 3 Apr. 2019
Bearbeitet: Guillaume
am 3 Apr. 2019
You could store your 15 tables into a cell array,
Trajectory = 1:15;
TrajDetail = cell(size(Trajectory));
for i = 1:numel(Trajectory) %I recommend using numel instead of length
%... your code as is
TrajDetail{i} = table(pos_x,pos_y,vel_x,vel_y);
end
However, you may be better of using just one table with an extra column indicating which iteration the trajectory came from:
Trajectory = 1:15
TrajDetail = [];
for i = 1:numel(Trajectory)
%... your code as is
trajindex = repmat(i, size(pos_x, 1), 1);
TrajDetail = [TrajDetail; table(trajindex, pos_x,pos_y,vel_x,vel_y)]
end
By the way, instead of
pos_x = xx(:,1);
pos_y = xx(:,2);
vel_x = xx(:,3);
vel_y = xx(:,4);
something = table(pos_x,pos_y,vel_x,vel_y);
You could just write:
something = array2table(xx, 'VariableNames', {'pos_x', 'pos_y', 'vel_x', 'vel_y'});
2 Kommentare
Guillaume
am 3 Apr. 2019
I assume that error comes from my 2nd example, where I store everything in just one table. I've fixed a typo in the repmat of that code.
You certainly can't get that error from my 1st example.
Siehe auch
Kategorien
Mehr zu MATLAB Compiler finden Sie in Help Center und File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!