machine learning - Singular Covariance and NA's for z value using Random Projection in R -


i'm getting na's z value in following code. it's due singular covariance. i've run similar test on iris dataset, , don't same error. doing wrong or nature of data?

code in r:

install.packages('mclust') library('mclust')  mydata <- read.table("http://archive.ics.uci.edu/ml/machine-learning-databases/wine-quality/winequality-red.csv", sep=";", header=true);  est <- mevvv(mydata[,-12], unmap(mydata[,12]))  randproj(mydata[,-12], seeds=200, parameters = variance, z = est$z, truth = iris[,5],          = "errors", identify = true, scale = true) 

thanks,

dgene

take @

table(mydata[, 12])  # 3   4   5   6   7   8  # 10  53 681 638 199  18 

this means using 10 observations in 1 cluster, while have 11 features (variables) in model, warning message na values.

if want specify model vvv use code :

 fm <- mclust(mydata[, -12],  modelnames="vvv") 

Comments

Popular posts from this blog

google chrome - Developer tools - How to inspect the elements which are added momentarily (by JQuery)? -

angularjs - Showing an empty as first option in select tag -

php - Cloud9 cloud IDE and CakePHP -