twodimensiontwogaussian.readme

InsightToolkit-1.4.0(有大量的优化算法程序)

* Overview
        The "TwoDimensionTwoGaussian.dat" file has 2000 measurement
        vectors. Each measurement vector consist of two components of
        double. The measurement vectors are generated using GNU R
        statistical package (http://www.r-project.org). 

        Half of the measurement vectors are generated using the the
        "First Gaussian" class statistics and the other half are
        generated using the "Second Gaussian" class statistics in the
        following section. First 1000 rows in the file belong to the
        first class and the rest belong to the second class. 
       
       The data is generated by a multivariate random normal variate
       generator (rmvnorm function in mvtnorm pacakge). The exact GNU
       R script for genation is in the "GNU R script" section of this
       document.          

* Class Statistics
1) class statistics for random sample generation
First Gaussian class:
        mean 
                100 100
        covariance 
                800 0
                0 800

Second Gaussian class:
        mean 
                200 200
        covariance
                800 0
                0 800
2) class statistics of the generated samples
First Gaussian class:
        mean 
                99.261 100.078
        covariance 
                814.95741  38.40308
                38.40308 817.64446


Second Gaussian class:
        mean 
                200.1 201.3
        covariance
                859.785295  -3.617316
                -3.617316 848.991508

* GNU R Script 

library(mvtnorm)
mean1 = c(100, 100)
mean2 = c(200, 200)
sigma = matrix(data=c(800, 0, 0, 800), ncol=2, nrow=2, byrow=T)
gauss1 = rmvnorm(1000, mean1, sigma)
gauss2 = rmvnorm(1000, mean2, sigma)
mydata = rbind(gauss1, gauss2)
write.table(mydata, file="TwoDimensionTwoGaussian.dat", quote=F, row.names=F, col.names=F)