normal.l2.cluster: Clustering of some normal data in 2d with the l2 clusterpath

In tdhock/directlabels: Direct Labels for Multicolor Plots

Clustering of some normal data in 2d with the l2 clusterpath

Description

The l2 clustering algorithm from the clusterpath package was applied to some randomly generated data in 2 dimensions, and the solutions found using the descent algorithm are stored in this data frame.

Usage

data(normal.l2.cluster)

Format

The format is: List of 2 $ pts :'data.frame': 320 obs. of 3 variables: ..$ class: Factor w/ 8 levels "1","2","3","4",..: 1 1 1 1 1 1 1 1 1 1 ... ..$ x : num [1:320] -2.73 -3.63 -2.13 -1.27 -2.98 ... ..$ y : num [1:320] -3.89 -3.43 -3.42 -3.17 -2.75 ... $ path:Classes 'l2', 'clusterpath' and 'data.frame': 21760 obs. of 7 variables: ..$ x : num [1:21760] -2.73 -3.63 -2.13 -1.27 -2.98 ... ..$ y : num [1:21760] -3.89 -3.43 -3.42 -3.17 -2.75 ... ..$ lambda: num [1:21760] 0 0 0 0 0 0 0 0 0 0 ... ..$ row : Factor w/ 320 levels "1","2","3","4",..: 1 2 3 4 5 6 7 8 9 10 ... ..$ gamma : Factor w/ 1 level "0.1": 1 1 1 1 1 1 1 1 1 1 ... ..$ norm : Factor w/ 1 level "2": 1 1 1 1 1 1 1 1 1 1 ... ..$ solver: Factor w/ 1 level "descent.nocheck": 1 1 1 1 1 1 1 1 1 1 ... ..- attr(*, "data")= num [1:320, 1:2] -2.73 -3.63 -2.13 -1.27 -2.98 ... .. ..- attr(*, "dimnames")=List of 2 .. .. ..$ : NULL .. .. ..$ : chr [1:2] "x" "y" ..- attr(*, "alphacolnames")= chr [1:2] "x" "y" ..- attr(*, "weight.pts")= num [1:320, 1:2] -2.73 -3.63 -2.13 -1.27 -2.98 ... .. ..- attr(*, "dimnames")=List of 2 .. .. ..$ : NULL .. .. ..$ : chr [1:2] "x" "y"

Source

clusterpath package

References

clusterpath article

Examples

data(normal.l2.cluster)
if(require(ggplot2)){
  p <- ggplot(normal.l2.cluster$path,aes(x,y))+
    geom_path(aes(group=row),colour="grey")+
    geom_point(aes(size=lambda),colour="grey")+
    geom_point(aes(colour=class),data=normal.l2.cluster$pts)+
    coord_equal()
  print(direct.label(p))
}

tdhock/directlabels documentation built on April 21, 2024, 11:32 a.m.