Error using () Subscripting into a table...

Hi all, just updated to R2023 in order to use the "combinations" function, and all of sudden another section of my code no longer works. It seems like Matlab is misinterpreting "dot(" (for dot product) as a variable? I provides this error message:

Error using ()

Subscripting into a table using one subscript (as in t(i)) is not supported. Specify a row subscript and a variable subscript, as

in t(rows,vars). To select variables, use t(:,i) or for one variable t.(i). To select rows, use t(i,:).

Error in dot (line 29)

a = a(:);

Error in hw4_problem4 (line 41)

dot( ...

I haven't named any variable as dot, obviously. The code for this section is below.

for n = 1:(number_of_vectors)
    for o = 1:(number_of_slip_directions)
        for p = 1:(number_of_slip_planes)
            schmid_factor(n,o,p) =     ( ...
                dot( ...
                stress_vectors(n,:), ...
                slip_planes(p,:) ...
                ) ...
                / ...
                sqrt( ...
                dot(stress_vectors(n,:),stress_vectors(n,:)) ...
                * ...
                dot(slip_planes(p,:),slip_planes(p,:)) ...
                ) ...
                )...
                *...
                ( ...
                dot( ...
                stress_vectors(n,:), ...
                slip_directions(o,:) ...
                )...
                / ...
                sqrt( ...
                dot(stress_vectors(n,:),stress_vectors(n,:)) ...
                *...
                dot(slip_directions(o,:),slip_directions(o,:)) ...
                ) ...
                );
        end
    end
end

[JORDAN]

Guess I'll post the full code here, just in case.

clear;clc;close all
%% Vector Setup
% This section inputs all of the vectors for the applied stress, the slip
% directions, and the slip plane normals vectors.
stress_vectors = unique(combinations([0 1 -1 1/2 -1/2],[0 1 -1 1/2 -1/2],[0 1 -1 1/2 -1/2]));
slip_directions = [ 1  1  0
    -1 -1  0
    1  0  1
    -1  0 -1
    0  1  1
    0 -1 -1];
slip_planes = [ 1  1  1
    -1 -1 -1];
%% Counting
% This section finds how many vectors of each type there are for later
% calculations.
[number_of_vectors,dimensions1]             = size(stress_vectors);
[number_of_slip_directions,dimensions2]     = size(slip_directions);
[number_of_slip_planes,dimensions3]         = size(slip_planes);
%% Schmid Factor Calculation
% This section calculates Schmid's Factor for each direction of applied
% stress. Schmid's Factor is found by multiplying two quantities: the first
% quantity is the dot product of the applied stress vector and the plane
% normal vector divided by the product of the magnitudes of each vector,
% and the second quantity is the dot product of the applied stress and the
% slip direction vector divided by the product of each vector's magnitude
schmid_factor = zeros(number_of_vectors,number_of_slip_directions,number_of_slip_planes);
for n = 1:(number_of_vectors)
    for o = 1:(number_of_slip_directions)
        for p = 1:(number_of_slip_planes)
            schmid_factor(n,o,p) =     ( ...
                dot( ...
                stress_vectors(n,:), ...
                slip_planes(p,:) ...
                ) ...
                / ...
                sqrt( ...
                dot(stress_vectors(n,:),stress_vectors(n,:)) ...
                * ...
                dot(slip_planes(p,:),slip_planes(p,:)) ...
                ) ...
                )...
                *...
                ( ...
                dot( ...
                stress_vectors(n,:), ...
                slip_directions(o,:) ...
                )...
                / ...
                sqrt( ...
                dot(stress_vectors(n,:),stress_vectors(n,:)) ...
                *...
                dot(slip_directions(o,:),slip_directions(o,:)) ...
                ) ...
                );
        end
    end
end
%% Finding the number of non-zero slip systems per applied stress vector
% This section counts the number of Schimd's Factors per applied stress
% vector that don't equal zero, then appends that number to the original
% set of stress vectors for use in an fprintf function that outputs a
% statement listing the number of slip systems per applied stress vector.
non_zero_slip_systems = zeros(number_of_vectors,1);
for q=1:number_of_vectors
    non_zero_slip_systems(q,1) = numel(find(schmid_factor(q,:,:) ~= 0));
end
applied_stress_and_system_count_matrix = sortrows([stress_vectors,non_zero_slip_systems],4);
formatSpec = 'Applied stress vector [%1.0f %1.0f %1.0f] has %1g slip systems\n';
%% ANSWERS
fprintf(formatSpec,(applied_stress_and_system_count_matrix'))

Answer 1

[Walter Roberson]

You are passing a table() object as a parameter to dot()

It looks like either stress_vectors or else slip_panes is a table() object.

[JORDAN]

Oh, duh. I didn't examine the combinations function output very well, I'm surprised that it generated a table... guess I'll try to sort that out. Thank you for the quick reply, this is my first time posting despite having used MATLAB for a couple of years and I wasn't sure how long it would take.

Answer 2

[Steven Lord]

Line 29 of the dot.m included with MATLAB doesn't match what you showed. MATLAB Answers uses release R2023b (Update 2, but I don't believe dot.m has changed in the 23b Update releases.)

version

ans = '23.2.0.2410171 (R2023b) Update 2'

dbtype 25:35 dot.m

25        end
26        tc = sum(conj(ta).*tb,dim);
27    end
28    end
29    

I suspect you've defined your own dot.m that's taking precedence over the one included with MATLAB. To check, see what this command displays:

which -all dot

/MATLAB/toolbox/matlab/specfun/dot.m
/MATLAB/toolbox/matlab/bigdata/@tall/dot.m              % tall method
/MATLAB/toolbox/parallel/gpu/@gpuArray/dot.m            % gpuArray method
/MATLAB/toolbox/parallel/parallel/@codistributed/dot.m  % codistributed method

[Walter Roberson]

MATLAB Version: 23.2.0.2365128 (R2023b)

has the call to with a(:) at line 29 of toolbox/matlab/specfun/dot.m

[Steven Lord]

Hmm. You're right. Ah, I'm guessing that dbtype call displayed a different dot.m than I expected.

dbtype 25:35 toolbox/matlab/specfun/dot.m

25    if nargin == 2
26    
27        if isvector(a) && isvector(b)
28            % Special case: A and B are vectors and dim not supplied
29            a = a(:);
30            b = b(:);
31            if length(a) ~= length(b)
32                error(message('MATLAB:dot:InputSizeMismatch'));
33            end
34            if isreal(a) && isreal(b)
35                c = a'*b;

It looks like instead dtype displayed the overload of dot for tall arrays.

dbtype 1:5 dot.m

1     function tc = dot(ta,tb,dim)
2     %DOT Vector dot product for tall arrays.
3     %   C = DOT(A,B)
4     %   C = DOT(A,B,DIM)
5     %