prtDistanceCustom   Custom distance function
  
    DIST = prtDistanceCityBlock(D1,D2,DISTFUNHANDLE)  calculates the
    distance from all the observations in dataset D1 to dataset D2, using
    the function handle DISTFUNHANDLE. The output is a distance matrix of
    size D1.nObservations x D2.nObservations. D1 and D2 must have the same
    number of features.
 
    DISTFUNHANDLE should be a function handle that accepts two 1xn
    vectors and outputs the scalar distance between them.  For example, 
 
    DISTFUNHANDLE = @(x,y)sqrt(sum((x-y).^2,2)); %euclidean distance
 
    Note: This is provided as an example only, use prtDistanceEuclidean to
    calculate Euclidean distances, as it is significantly faster than
    prtDistanceCustom.
    
    %  Example:
 
    % Create 2 data sets
    dsx = prtDataSetStandard('Observations',[0 0; 1 1]);
    dsy = prtDataSetStandard('Observations',[1 0; 2 2; 3 3;]);
    % Compute their distance based on function handle
    distance = prtDistanceCustom(dsx,dsy,@(x,y)sqrt(sum((x-y).^2,2)))