R/datasets/classification/twoGaussianMixture.R

In RamboWANG/RegularizedCrossValidation: Provides several types of regularized cross-validation methods.

twoGaussianMixture.DataGenerator<-function(n,p)
{   
  n1<-trunc(n/2)
  n0<-n-n1  
  mu1<-rep(0,p)
  mu01<-rep(0.4,p/3)
  mu02<-rep(0.3,p/3)
  mu03<-rep(0,p/3)
  sigma1<-diag(p)
  sigma0<-diag(p/3)
  x1<-mvrnorm(n1,mu1,sigma1)
  x01<-mvrnorm(n0,mu01,sigma0)
  x02<-mvrnorm(n0,mu02,sigma0)
  x03<-mvrnorm(n0,mu03,sigma0)  
  x0<-cbind(x01,x02,x03)
  x<-rbind(x1,x0)
  y<-c(rep(1,n1),rep(0,n0))
  return(list(x=x,y=y))
}