Given a sub-partion (xmin,xmax) of[0,s], returns the value of feature FeatureIdx, which minimizes
the ratio (lowercount(feature,c1)/(uppercount(feature,c2)+x)) in the interval.
If there are missing data in the NonMar part of the units to be classified, this function
identifies the intervals in which the range [0,s] has to be sub-partitioned.
Counts that correspond to counts-after-dropping-missing for MarFeatures,
bivariate count: frequency are computed for each output class and for each class of the
feature.