Trainable radial basis neural network classifier
W = RBNC(A,UNITS)
A feed-forward neural network classifier with one hidden layer with UNITS radial basis units is computed for labeled dataset A. Default UNITS is number of objects * 0.2, but not more than 100. Uses the Mathworks' Neural Network toolbox.
If UNITS is NaN it is optimised by REGOPTC. This may take a long computing time and is often not significantly better than the default.
This routine escapes to KNNC for sample sizes < 6.
This routine calls MathWork's SOLVERB routine (Neural Network toolbox) as SOLVB.