M-file for exercise!!!

Hi!I have to create an m-file for an exercise and I wanted to ask you what is the difference between a function and a script.Which of them is better??

[per isakson]

the debugger and the profiler works better with functions

[evi]

So,do you think that the error message I get has to do with the debugger and the profiler??

Answer 1

[sixwwwwww]

If you want to re-use some operation again and again then function is better and if you just wanted to start writing codes in MATLAB then script is better at first. Better or not depends upon functionality usually

[evi]

Which would you suggest me,for example,for numerical methods,like the implementation of the Jacobi method??

[sixwwwwww]

I will suggest you using script first

[evi]

And is there a difference between these??Because I wrote my code as a function and also as a script and don't get the same result...

[sixwwwwww]

If you using same parameters in the script and function then you should get the same results. You should take care of values which you input to function and compare those values in the script. If they are same and also implementation in script and function is same then you should get same results beside some minor numerical noise

[evi]

When I run the script file for n,that is the maximum number of iterations,I get this message error: ??? Error using ==> mldivide Matrix dimensions must agree

but when I run the function file It gives a result..Why does this happen???

[sixwwwwww]

can you show me both your script and you function? to find the reason for error

[evi]

The function is like that:

function[]=test(x_0,dimension,e,maxit) H=hilb(dimension); A=H; for j=1:dimension b(j)=sum(A(j,:)); exact(j)=1; end b=b'; exact=exact'; P=diag(diag(A)); F=A-P; x=-P\F*x_0+P\b; end

and the script is like that:

   dimension=input('Give dimension: ');
   H=hilb(n);
   A=H;
   e=input('Give e: ');
   maxit=input('Give maxit: ');
   x_0=input('Give x_0: ');
   for j=1:dimension
       b(j)=sum(A(j,:));
       exact(j)=1;
   end
   b=b';
   exact=exact';
   P=diag(diag(A));
   F=A-P;
   x=-P\F*x_0+P\b;

When I run the function for dimension=50,e: 0.00001,maxit: 50,x_0: zeros(50,1),I get a result,but when I run the script I get this message: ??? Error using ==> mtimes Inner matrix dimensions must agree.

Error in ==> ll at 15 x=-P\F*x_0+P\b

Why does this happen???What have I done wrong???

[sixwwwwww]

I am getting result both from your script and function. Following are the inputs: For function

test(zeros(50, 1), 50, 0.00001, 50)

For script

Give dimension: 50
Give e: 0.00001
Give maxit: 50
Give x_0: zeros(50, 1)

then why you dont get correct result. Can you check what inputs you give to both your function and script?

Answer 2

[Andreas Goser]

The documentation for this is also quite good:

http://www.mathworks.com/help/matlab/scripts.html

http://www.mathworks.com/help/matlab/functions.html