How can i solve the issue of struct2table

For a while now, i've been trying to implement the struct2table command in my code to like get my simulated variables in form f table but it has been futile despite watching,reading and trying all i have seen, still no postitive change. i need help, probably there is something i am not doing the right way until now.

intlinsol is a 768 x 1 (these are binary integers numbers of 1's and 0's) i got after simulating about 16 variables which has 48 binary enteries (of 1's and 0's) each

so then i wrote this:

tbl=struct2table(intlinsol);
vars= {'xwm1','xwm2','xdw1','xdw2','xvc1','xvc2','xmw','xtt','xsi','xest1','xest2','xov','xtm','xtd1','xtd2','xrf','uwm1','uwm2','udw1','udw2','uvc1','uvc2','umw','utt','usi','uest1','uest2','uov','utm','utd1','utd2','urf'};
 outputvars= stack(tbl,vars,'NewDataVariableName','Amt','IndexVariableName','Var')

this is what matlab says:

"Error using struct2table (line 33) , Input structure must be a scalar structure, or a structure array with one column or one row."

Error in realmodifiedcode (line 213)

tbl=struct2table(intlinsol);

Answer 1

[Matt J]

I don't know why you think struct2table is relevant to this problem if intlinsol is not a struct. You have also not described how the 16 variables are distributed over the 768 elements of linsol. If each successive group of 48 elements belong to one of the 16 variables, then you could build tbl as follows:

tbl=array2table( reshape(intlinsol,[],16) )

You then present 32 variable names vars in your code rather than the 16 that you say you have. Not sure how those are related to anything.

[Timmy Lehin]

thank you very much for the response.

i really didn't know struct2table wouldn't work for intlinsol.

i re-strutured my code and i had to solve for 32 variables consisting of 48 elements each.

After the simulation, i was wanting to create a table for those variables or better still i will say i wanted a table to be created so i could easily see the answers (48 elements each) to the 32 variables because it is hard to recognize the format at which it was solved manually, e.g., the variable that was solved first or that own the first 48 digits and the variable with the next 48 and so on.

[Timmy Lehin]

initally i use the code

showproblem(prob) and when i do that i can easily identify the other in which the problem was solved (e.g., maybe the variable 'xwm1' came first or another) but this worked when i had just 16 variables with 48 elements and i was able to see the full scope when i used that code (i.e., showproblem(prob)) but after restructuring the code to 32 variables with 48 elements, i used the same code but this time around, it wasn't showing me the full scope of how the problem was solved (e.g., the variable that came first and even i couldn't see the full objective function unlike when the first time i used 16 variables with 48 enteries)

that is why i resulted back to trying to use the code hat could automatically draft out the varaibles and the appropriate 48 elements

i used this:

tbl=array2table( reshape(intlinsol,[],32) );
vars={'Xwm1','Xwm2','Xdw1','Xdw2','Xvc1','Xvc2','Xmw','Xtt','Xsi','Xest1','Xest2','Xov','Xtm','Xtd1','Xtd2','Xrf','Uwm1','Uwm2','Udw1','Udw2','Uvc1','Uvc2','Umw','Utt','Usi','Uest1','Uest2','Uov','Utm','Utd1','Utd2','Urf'};
outputvars= stack(tbl,vars,'NewDataVariableName','Amt','IndexVariableName','Var')

NB: 'Xwm1' and co reprsents the variable names each with 48 elements already defined from the beginning

this is what i got:

Error using tabular/stack (line 115)

Unrecognized table variable name 'Xwm1'.

Error in COMPLETEDCODE1 (line 283)

outputvars=stack(tbl,vars,'NewDataVariableName','Amt','IndexVariableName','Var')

[Timmy Lehin]

This is how i had defined my unknown 32 variableswith 48 elements each

nos=48;
% Define unknown main decision variables
Xwm1= optimvar('xwm1',nos,'LowerBound',0,'UpperBound',1,'Type','integer'); 
Xwm2= optimvar('xwm2',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xdw1= optimvar('xdw1',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xdw2= optimvar('xdw2',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xvc1= optimvar('xvc1',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xvc2= optimvar('xvc2',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xmw = optimvar('xmw',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xtt = optimvar('xtt',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xsi = optimvar('xsi',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xest1=optimvar('xest1',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xest2=optimvar('xest2',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xov=  optimvar('xov',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xtm = optimvar('xtm',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xtd1= optimvar('xtd1',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xtd2= optimvar('xtd2',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xrf = optimvar('xrf',nos,'LowerBound',0,'UpperBound',1,'Type','integer');

% Define unknown auxiliary decision variables
Uwm1= optimvar('uwm1',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Uwm2= optimvar('uwm2',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Udw1= optimvar('udw1',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Udw2= optimvar('udw2',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Uvc1= optimvar('uvc1',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Uvc2= optimvar('uvc2',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Umw = optimvar('umw',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Utt = optimvar('utt',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Usi = optimvar('usi',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Uest1=optimvar('uest1',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Uest2=optimvar('uest2',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Uov = optimvar('uov',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Utm = optimvar('utm',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Utd1= optimvar('utd1',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Utd2= optimvar('utd2',nos,'LowerBound',0,'UpperBound',1,'Type','integer');

Answer 2

[Steven Lord]

You described intlinsol as "intlinsol is a 768 x 1 (these are binary integers numbers of 1's and 0's)" which to me implies that it is a logical or double vector. In that case, struct2table is not the right tool to convert that into a table array, since intlinsol is not a struct. Just call table on it to create a 768-by-1 table.

v = randi(5, 8, 1)

v = 8×1

1 1 5 4 3 2 5 2

t = table(v)

t = 8×1 table
    v
    _

    1
    1
    5
    4
    3
    2
    5
    2

[Timmy Lehin]

this worked but how to make sure those binary digits are assigned to the 32 variables which is represented by ('Xwm1', 'Xwm2'.....until the last) of 48 binary elements each is the challenge

[Timmy Lehin]

after defining all my variables (32 different variables with 48 elements) and constraints i used the code below to solve and it gave me the 1536 x 1 binary numbers which i would love to know which belongs to which variable

problem = prob2struct(prob);
[sol,fval,exitflag,output]= intlinprog(problem); 

[Timmy Lehin]

This is how i had defined my unknown 32 variableswith 48 elements each

nos=48;
% Define unknown main decision variables
Xwm1= optimvar('xwm1',nos,'LowerBound',0,'UpperBound',1,'Type','integer'); 
Xwm2= optimvar('xwm2',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xdw1= optimvar('xdw1',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xdw2= optimvar('xdw2',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xvc1= optimvar('xvc1',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xvc2= optimvar('xvc2',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xmw = optimvar('xmw',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xtt = optimvar('xtt',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xsi = optimvar('xsi',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xest1=optimvar('xest1',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xest2=optimvar('xest2',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xov=  optimvar('xov',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xtm = optimvar('xtm',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xtd1= optimvar('xtd1',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xtd2= optimvar('xtd2',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Xrf = optimvar('xrf',nos,'LowerBound',0,'UpperBound',1,'Type','integer');

% Define unknown auxiliary decision variables
Uwm1= optimvar('uwm1',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Uwm2= optimvar('uwm2',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Udw1= optimvar('udw1',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Udw2= optimvar('udw2',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Uvc1= optimvar('uvc1',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Uvc2= optimvar('uvc2',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Umw = optimvar('umw',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Utt = optimvar('utt',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Usi = optimvar('usi',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Uest1=optimvar('uest1',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Uest2=optimvar('uest2',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Uov = optimvar('uov',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Utm = optimvar('utm',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Utd1= optimvar('utd1',nos,'LowerBound',0,'UpperBound',1,'Type','integer');
Utd2= optimvar('utd2',nos,'LowerBound',0,'UpperBound',1,'Type','integer');

[Steven Lord]

So in this case sol is likely a double array containing only 1's and 0's. Is that correct?

In that case reshape it into an array. For example, if I had a vector of twenty 0's and 1's:

x = [1 0 0 1, ... % 9
     0 1 0 0, ... % 4
     0 1 1 0, ... % 6
     0 0 1 1, ... % 3
     1 0 0 0]     % 8

x = 1×20

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

I could reshape it. Remember that when you reshape MATLAB fills in down the first column then down the second etc. instead of across the first row then across the second etc.

digitsAsColumns = reshape(x, 4, 5) % each column represents a digit

digitsAsColumns = 4×5

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

values = [8 4 2 1]*digitsAsColumns

values = 1×5

9 4 6 3 8

digitsAsRows = digitsAsColumns.'

digitsAsRows = 5×4

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

values = digitsAsRows*[8; 4; 2; 1]

values = 5×1

9 4 6 3 8

In this context I'm not sure where converting to a table array comes into play.

This assumes that the bits of the numbers are stored consecutively. If instead the values were stored with all the highest order bits first, then the next highest, etc.

y = [1  0  0  1, ... 
     0  1  0  0, ... 
     0  1  1  0, ... 
     0  0  1  1, ... 
     1  0  0  0]     

y = 1×20

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

   %17 12  6 18

then you'd need to reshape a bit differently.

recreateY = reshape(y, 4, 5).'

recreateY = 5×4

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

yValues = 2.^(4:-1:0)*recreateY

yValues = 1×4

17 12 6 18

[Timmy Lehin]

unfrotunately, i couldn't wrap my brain through it but i was eventuall able to get through it.

i observed that the variables are solved and given in binary form in accordance to the alphabetical order so i was able to bye-pass it by logically writing a few codes in that regards like the one below to seperate each variables with their 48 elements:

% Writing an equation to be able to extract variables for plotting the graph 
udsw_sol=sol(1:nos);
udw_sol=sol(1*nos+1:2*nos);
ues_sol=sol(2*nos+1:3*nos);
uest_sol=sol(3*nos+1:4*nos);
umw_sol=sol(4*nos+1:5*nos);
uov_sol=sol(5*nos+1:6*nos);
urf_sol=sol(6*nos+1:7*nos);
usi_sol=sol(7*nos+1:8*nos);
utd_sol=sol(8*nos+1:9*nos);
utm_sol=sol(9*nos+1:10*nos);
utmd_sol=sol(10*nos+1:11*nos);
utt_sol=sol(11*nos+1:12*nos);
uvc_sol=sol(12*nos+1:13*nos);
uvcc_sol=sol(13*nos+1:14*nos);
uwm_sol=sol(14*nos+1:15*nos);
uwsm_sol=sol(15*nos+1:16*nos);

xdsw_sol=sol(16*nos+1:17*nos);
xdw_sol=sol(17*nos+1:18*nos);
xes_sol=sol(18*nos+1:19*nos);
xest_sol=sol(19*nos+1:20*nos);
xmw_sol=sol(20*nos+1:21*nos);
xov_sol=sol(21*nos+1:22*nos);
xrf_sol=sol(22*nos+1:23*nos);
xsi_sol=sol(23*nos+1:24*nos);
xtd_sol=sol(24*nos+1:25*nos);
xtm_sol=sol(25*nos+1:26*nos);
xtmd_sol=sol(26*nos+1:27*nos);
xtt_sol=sol(27*nos+1:28*nos);
xvc_sol=sol(28*nos+1:29*nos);
xvcc_sol=sol(29*nos+1:30*nos);
xwm_sol=sol(30*nos+1:31*nos);
xwsm_sol=sol(31*nos+1:32*nos);

thanks alot for your concern, it is extremely appreciated.

[Timmy Lehin]

I have a question on stacked barchart, got any idea about that?

[Steven Lord]

Since that doesn't sound really related to the topic of this question you should ask it as a separate question with a descriptive title.

[Timmy Lehin]

alright, thanks. i will do that