Unrecognized function or variable 'options'. for simulannealbnd, but works for fminunc

Question

Subrat Bastola 2023-4-10

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1944839-unrecognized-function-or-variable-options-for-simulannealbnd-but-works-for-fminunc

评论： Alan Weiss 2023-4-11

ft_inverse_dipolefit.m

When I change the optimization function to simulannealbnd instead of fminunc, MATLAB shows error " No variable named options".

What wrong am i doing?

I changed the code from :

if isempty(optimfun)
  % determine whether the MATLAB Optimization toolbox is available and can be used
  if ft_hastoolbox('optim')
    optimfun = @fminunc;
  else
    optimfun = @fminsearch;
  end
end
if isempty(maxiter)
  % set a default for the maximum number of iterations, depends on the optimization function
  if isequal(optimfun, @fminunc)
    maxiter = 1000;
  else
    maxiter = 3000;
  end
end
if isequal(optimfun, @fminunc)
  options = optimset(...
    'TolFun',1e-9,...
    'TypicalX',scale*ones(size(param)),...
    'LargeScale','off',...
    'HessUpdate','bfgs',...
    'MaxIter',maxiter,...
    'MaxFunEvals',2*maxiter*length(param),...
    'Display',display);
elseif isequal(optimfun, @fminsearch)
  options = optimset(...
    'MaxIter',maxiter,...
    'MaxFunEvals',2*maxiter*length(param),...
    'Display',display);

to:

if isempty(optimfun)
  % determine whether the MATLAB Optimization toolbox is available and can be used
  if ft_hastoolbox('optim')
    optimfun = @simulannealbnd;
  else
    optimfun = @fminsearch;
  end
end
if isempty(maxiter)
  % set a default for the maximum number of iterations, depends on the optimization function
  if isequal(optimfun, @simulannealbnd)
    maxiter = 1000;
  else
    maxiter = 3000;
  end
end
if isequal(optimfun, @simulannealbnd)
    options = optimoptions(...
        'TolFun',1e-9,...
        'TypicalX',scale*ones(size(param)),...
        'MaxIter',maxiter,...
        'MaxFunEvals',2*maxiter*length(param),...
        'Display',display);
end
elseif isequal(optimfun, @fminsearch)
  options = optimset(...
    'MaxIter',maxiter,...
    'MaxFunEvals',2*maxiter*length(param),...
    'Display',display);

% perform the optimization with either the fminsearch or fminunc function

  [param, fval, exitflag, output] = optimfun(@dipfit_error, param, options, dat, sens, headmodel, constr, metric, checkinside, mleweight, reducerank, normalize, normalizeparam, weight, backproject);

The complete file for code is attached here

I get an error saying ' Unrecognized function or variable 'options'.' How do i solve this?

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Alan Weiss 2023-4-11

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1944839-unrecognized-function-or-variable-options-for-simulannealbnd-but-works-for-fminunc#answer_1213649

在 MATLAB Online 中打开

optimoptions requires that the first argument be the name of the solver. Something like

options = optimoptions('simulannealbnd',...
        'TolFun',1e-9,...
        'TypicalX',scale*ones(size(param)),...
        'MaxIter',maxiter,...
        'MaxFunEvals',2*maxiter*length(param),...
        'Display',display);

Alan Weiss

MATLAB mathematical toolbox documentation

4 个评论
显示 2更早的评论隐藏 2更早的评论

Subrat Bastola 2023-4-11

Update: I fixed it by:

% Define the objective function

objfun = @(param) dipfit_error(param, dat, sens, headmodel, constr, metric, checkinside, mleweight, reducerank, normalize, normalizeparam, weight, backproject);

% Call `simulannealbnd`

[param, fval, exitflag, output] = simulannealbnd(objfun, param, [], [], options);

Alan Weiss 2023-4-11

Great! If you feel that I helped, please Accept my answer.

Alan Weiss

MATLAB mathematical toolbox documentation

请先登录，再进行评论。

Answer 2

Walter Roberson 2023-4-11

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1944839-unrecognized-function-or-variable-options-for-simulannealbnd-but-works-for-fminunc#answer_1213199

在 MATLAB Online 中打开

if isequal(optimfun, @simulannealbnd)
    options = optimoptions(...
        'TolFun',1e-9,...
        'TypicalX',scale*ones(size(param)),...
        'MaxIter',maxiter,...
        'MaxFunEvals',2*maxiter*length(param),...
        'Display',display);
end