Multiple outputs from anonymous function in fminsearch

Question

I am trying to return results from an fminsearch optimization. I am using fminsearch to find the optimum hyperparameters (variable z) for an SVM. The anonymous function is minimizing classification error ('Crit') but I also wish to return another variable (the selected feature for given hyperparameters) obtained at the same iteration ('Features'):

fun = @(z)SVM_min_fn(Data,Labels,exp(z(1)),exp(z(2)),num_folds);
[z_opt,Crit] = fminsearch(fun,z0,opts);

function [Crit Features] = SVM_min_fn(Data,Labels,rbf_sigma,boxconstraint,num_folds)
direction = 'forward';
opts = statset('display','iter');
kernel = 'rbf';

disp(sprintf('RBF sigma: %1.4f. Boxconstraint: %1.4f',rbf_sigma,boxconstraint))
c = cvpartition(Labels,'k',num_folds);
opts = statset('display','iter','TolFun',1e-3);
fun = @(x_train,y_train,x_test,y_test)SVM_class_fun(x_train,y_train,x_test,y_test,kernel,rbf_sigma,boxconstraint);
[fs,history] = sequentialfs(fun,Data,Labels,'cv',c,'direction',direction,'options',opts);

Features = find(fs==1);        % Features selected for given sigma and C
[Crit,h] = min(history.Crit);  % Mean classification error

Is there a way to make 'fminsearch' return both 'Crit' and Features ? Saving to workspace does not work as the Features are not the correct ones for the hyperparameters returned by 'fminsearch'

Answer 1

It's simplest if you do one more function evaluation after fminsearch completes:

fun = @(z)SVM_min_fn(Data,Labels,exp(z(1)),exp(z(2)),num_folds);
[z_opt,Crit] = fminsearch(fun,z0,opts);

[~, Features_opt] = fun(z_opt);