Creating symbolic function with array for argument

Hi,

I am trying to understand how to use mutli dimension array with symbolic function. I am not sure why I get an error for the symolic function when the input argument is two dimension.

x = sym ('x', [2 1]);
y = sym( 'y', [1 2]);
z = sym( 'z', [2 5]);  
f(x,y,z)=(x(1,1)+y(1,1)+z(1,1));

Error using sym/cat>checkDimensions
CAT arguments dimensions not consistent.

Error in sym/cat>catMany (line 33)
[resz, ranges] = checkDimensions(sz,dim);

Error in sym/cat (line 25)
ySym = catMany(dim, args);

Error in sym/horzcat (line 19)
ySym = cat(2,args{:});

Error in indexing (line 1033)
C = symfun(sym(R),[inds{:}]);

if I change the variable to

x = sym ('x', [1 1]);
y = sym( 'y', [1 2]);
z = sym( 'z', [1 5]);  
f(x,y,z)=(x(1,1)+y(1,1)+z(1,1));

it works fine, I am not sure I understand why

[VBBV]

x = sym ('sh', [2 1]);
y = sym( 'ao', [1 2]);
z = sym( 'aa', [2 5]);  
f=(x(1,1)+y(1,1)+z(1,1))  % The LHS of equation is creating problem since x, y, z are symbolic arrays !!

f = 

[Benoit Cabot]

Yes, but if I understand proprelly it is not a function anymore. I want to be able to evaluate the symbolic function for specific values in a later stage without converting it.

[VBBV]

Do you mean something like this ?

x = sym ('sh');
y = sym( 'ao');
z = sym( 'aa');  
f(x,y,z) =  symfun(x + y + z,[x,y,z])  % 

f(sh, ao, aa) = 

f(1,2,1)

ans = 

4

The RHS of the below equation is already evaluated for specfic points in the symbolic arrays

f=(x(1,1)+y(1,1)+z(1,1));

[Benoit Cabot]

Thank you for your time, maybe I simplified it to much for the question. but the input argument need to be 2D array because the real function is more complicated. maybe here is a better example.

x = sym ('x', [2 1]);
y = sym( 'y', [1 2]);
z = sym( 'z', [2 5]);  
f(x,y,z)=symfun(x(1,1)+x(2,1)+y(1,1)+y(1,2)+z(1,1)+z(2,5),[x,y,z]);

Error using sym/cat>checkDimensions
CAT arguments dimensions not consistent.

Error in sym/cat>catMany (line 33)
[resz, ranges] = checkDimensions(sz,dim);

Error in sym/cat (line 25)
ySym = catMany(dim, args);

Error in sym/horzcat (line 19)
ySym = cat(2,args{:});

X=[1 ; 1];
Y=[1  1];
Z=[1 2 3 4 5; 6 7 8 9 0];
f(X,Y,Z)

[VBBV]

symfun operates using scalars more efficiently. The function f has three independent variables, x y and z . So, the function needs inputs to three variables

x = sym ('x');
y = sym( 'y');
z = sym( 'z');  
f(x,y,z)=symfun(x+y+z,[x,y,z])

f(x, y, z) = 

X=[1 ; 1];
Y=[1  1];
Z=[1 2 3 4 5; 6 7 8 9 0];

f(X(1,1),Y(1,1),Z(1,1)) + f(X(2,1),Y(1,1),Z(1,1)) + f(X(1,1),Y(1,1),Z(1,1)) + f(X(1,1),Y(1,2),Z(1,1)) + ...
    f(X(1,1),Y(1,1),Z(1,1)) + f(X(1,1),Y(1,1),Z(2,5))

ans = 

17

[Benoit Cabot]

thank you

Answer 1

[Walter Roberson]

You cannot create symfun that expect non-scalars as parameters.

symfun takes all of the provided parameters, horzcat()'s them together, and creates individual parameters for the results.

x = sym ('x', [1 2]);
y = sym ('y', [1 2]);
f(x, y) = x(1) + x(2) + y(1) + y(2)

f(x1, x2, y1, y2) = 

Notice that it has become a function of four parameters.

x = sym ('x', [2 1]);
y = sym ('y', [2 1]);
f(x, y) = x(1) + x(2) + y(1) + y(2)

Error using symfun.validateArgNames
Second argument must be a scalar or vector of unique symbolic variables.

Error in symfun (line 102)
y.vars = symfun.validateArgNames(inputs);

Error in indexing (line 1033)
C = symfun(sym(R),[inds{:}]);

Notice the only difference here is the orientation of x and y, but it just fails.

[Benoit Cabot]

thank you