 Usage:  net=nnrnet (x, y, weights, size{, param {, wts}})  
 
 Input:

  x                      n x p matrix input variables 
                         
  y                      n x q matrix output variables 
                         
  weights                n x 1 vector of weights 
                         
  size                   scalar number of hidden units 
                         
  param                  7 x 1 vector of parameters 
                         
  wts                    vector of predefined weights 
                         
 Output:

  net.n                  3 x 1 vector number of input, hidden and output units 
                         
  net.nunits             scalar 
                         
  net.nconn              vector 
                         
  net.conn               vector 
                         
  net.decay              scalar weight decay parameter 
                         
  net.entropy            scalar 
                         
  net.softmax            scalar 
                         
  net.value              scalar value of error function 
                         
  net.wts                vector of weights 
                         
  net.yh                 n x q estimated y's 
                         
--------------------------------------------------------------
(C) MD*TECH Method and Data Technologies, 17.8.2000
