Error using repmat function

Error using repmat

Replication factors must be a row vector of integers or integer scalars.

Can anyone tell me why i get this error when I am trying to use

    x{i} = repmat(shiftdim(varargin{i}(j,:),1-i),repvec);

[Walter Roberson]

What is in repvec ? What is class(repvec) ?

[Nik Sam]

function fzr = rtrm(func, varargin)

% RTRM Fast, vectorized implementation of the
reduced transformation method
% using multi-dimensional arrays.
% FZR = RTRM(FUNC, FZ1, FZ2, ..., FZN)
% calls the reduced transformation method
% with N fuzzy numbers in alpha-cut repre-
% sentation. The function FUNC must be an
% analytical expression using \textsc{Matlab}'s
% array arithmetic (or logic) operators.
%

% Example:

% f = inline('x.^2 - y');

% fz1 = [0 5; 1 4; 2 3; 2.5 2.5];

% fz2 = [1 3; 1.5 2.5; 1.75 2.25; 2 2];

% fzr = rtrm(f, fz1, fz2)

% plot(fzr,linspace(0,1,4))

n = nargin - 1;

m = size(varargin{1},1) - 1;

fzr = repmat([inf -inf],m+1,1);
for j = m+1:-1:1

for i = 1:n

repvec = 2*ones(n,1);

repvec(i) = 1;

x{i} = repmat(shiftdim(...

varargin{i}(j,:)',1-i),repvec);

end

z = reshape(feval(func,x{:}),2^n,1);

fzr(j,1) = min(min(z),fzr(min(j,m)+1,1));

fzr(j,2) = max(max(z),fzr(min(j,m)+1,2));

end

This all the code

Answer 1

[Walter Roberson]

Where you have

repvec = 2*ones(n,1);

change that to

repvec = 2*ones(1,n);

[Nik Sam]

Now I get this.Why?

Error using inlineeval (line 14)

Error in inline expression ==> x.^2 - y

 Matrix dimensions must agree.

Error in inline/feval (line 33)

        INLINE_OUT_ = inlineeval(INLINE_INPUTS_, INLINE_OBJ_.inputExpr, INLINE_OBJ_.expr);

Error in rtrm (line 27)

z = reshape(feval(func,x{:}),2^n,1);

[Walter Roberson]

With just the change I mentioned, I get results. I have enclosed the edited file. I tested with

f = inline('x.^2 - y');
fz1 = [0 5; 1 4; 2 3; 2.5 2.5];
fz2 = [1 3; 1.5 2.5; 1.75 2.25; 2 2];
fzr = rtrm(f, fz1, fz2)