optimize Optimize action parameter by exhaustive function maximization.
 
   OPTIMACT = optimize(DS, EVALFN, PARAMNAME, PARAMVALS)
   returns an optimized prtAction object, with parameter
   PARAMNAME set to the optimal value. DS must be a prtDataSet
   object. EVALFN must be a function handle that returns a
   scalar value that indicates a performance metric for the
   prtAction object, for example a prtEval function. PARAMNAME
   must be a string that indicates the parameter of the
   prtAction that is to be optimized. PARAMVALS must be a
   vector of possible values of the parameter at which
   prtAction will be evaluated.
 
   [OPTIMACT, PERF]  = optimize(...) returns a vector of
   performance values that correspond to each element of
   PARAMVALS.
 
  Example:
 
   ds = prtDataGenBimodal;  % Load a data set
   knn = prtClassKnn;       % Create a classifier
   kVec = 3:5:50;           % Create a vector of parameters to
                            % optimze over
 
  % Optimize over the range of k values, using the area under
  % the receiver operating curve as the evaluation metric.
  % Validation is performed by a k-folds cross validation with
  % 10 folds as specified by the call to prtEvalAuc.
            
  [knnOptimize, percentCorrects] = knn.optimize(ds, @(class,ds)prtEvalAuc(class,ds,10), 'k',kVec);
  plot(kVec, percentCorrects)

Help for prtClassKmeansPrototypes/optimize is inherited from superclass prtAction