SDROC Estimate ROC on a dataset with classifier outputs

    R=SDROC(OUT,options)
    R=SDROC(OPS,OUT) % use op.points OPS to estimate ROC
    R2=SDROC(R,OUT)  % re-estimate from OUT using op.points in R
    RA=SDROC({R1,R2,R3,...}) % average estimates in a set of ROC

 INPUT
    OUT   dataset with classifier outputs
    OPS   operating point set (see SDOPS)

 OUTPUT
    R,R2,RA  SDROC objects

 OPTIONS
  'target' Name of the target decision
  'non-target'  Name of the non-target decision in the resulting op.point.
  'reject'  Add a reject option and construct reject curve.
            - if (0,1) fraction is given, set threshold by rejecting
              percentage of all samples
            - if SDOPS or SDROC is given, use current op.point
  'measures' Cell array with measure names and parameters.

 DESCRIPTION
 SDROC performs ROC analysis on output dataset OUT.  If OUT contains more
 than two output features, weighting-based ROC is applied. OUT with a
 single output column leads to the thresholding-based ROC. The feature-label
 of the only column is considered a target class name, the only-other
 appearing among the OUT labels the non-target.  If mutliple classes are
 present in OUT, use SDRELAB to convert it to two-class dataset (use the
 CONVFLAB flag to update also the feature labels).
 A specific set of op.points may be used if supplied in SDDECIDE
 mapping WD. Similarly, new ROC R2 may be created from existing ROC object
 R re-estimating the performance from OUT on operating points in R.