Labelling {svcR} | R Documentation |
Labelling methods try to identify clusters in a grid.
## S4 method for signature 'findModelCluster': Labelling.compute(x, MetLab = 1, MatriceKernel = MK, MatriceK = M_K, pp = vpp, Nu = 1, G = 1, q = 1, ncol = 1, nlin = 1, RadiusC = 2, r = 2, KernChoice = 0.01, NbClassInData = 0.01 ) ## S4 method for signature 'findModelCluster': AdjacencyPP(x, MatriceKernel=matrix(), Vec1=vector(), Vec2=vector()) ## S4 method for signature 'findModelCluster': Adjacency(x, MatriceKernel=matrix()) ## S4 method for signature 'findModelCluster': MST_labelling(x, MatriceKernel=matrix()) ## S4 method for signature 'findModelCluster': KNN_labelling(x, MatriceKernel=matrix())
x |
a findModelCluster object |
MetLab |
option taking value 1 (grid labelling) or 2 (mst labelling) or 3 (knn labelling) |
MatriceKernel |
kernel matrix with vector format |
MatriceK |
kernel matrix with vector format |
pp |
option taking value 0 (Euclidian) or 1 (RBF) or 2 (Exponential) |
Nu |
kernel parameter |
G |
kernel parameter |
q |
kernel parameter |
ncol |
number of attributes |
nlin |
size of variables |
RadiusC |
model radius |
r |
residual radius |
KernChoice |
kernel id |
NbClassInData |
number of classes in data |
Vec1 |
vector for computing adjacency |
Vec2 |
vector for computing adjacency |
An S4 object of class labelling
The object is the svc model along with
the slots :
ClassPoints |
class of grid points |
NumPoints |
value fo grid points |
slots can be accessed by object@slot
Nicolas Turenne - INRA France nicolas.turenne@jouy.inra.fr
N.Turenne , Some Heuristics to speed-up Support Vector Clustering , technical report 2006, INRA, France http://migale.jouy.inra.fr/~turenne/svc.pdf
## exemple with iris data fmc = findModelCluster.Test(); fmc@NumPoints; # values of the grid