gcdclust: gcdclust is used for calculating the solutions paths of...

In KarimOualkacha/HHSVM-ClusterNet: hhsvm classification using the cluster elastic net penality

Description

This function is used to calculat the solutions of HHSVM-CEN

Usage

gcdclust(x, y, KK, nlambda = 30, method = "hhsvm", lambda.factor = ifelse(nobs < nvars, 0.01, 1e-04), lambda = NULL, lambda2 = 0, pf = rep(1, nvars), pf2 = rep(1, nvars), exclude, dfmax = nvars + 1, pmax = min(dfmax * 1.2, nvars), standardize = TRUE, eps = 1e-05, maxit = 1e+06, delta = 2)

Arguments

x	The matrix or data
y	A response
KK	number of cluster in data
nlambda	The number of lambda1 used in the algorithm, the defaut value is 30
method	Our method HHSVM-CEN
lambda.factor	ifelse(nobs < nvars, 0.01, 1e-04), depends of data (relation betwin nobs and nvars)
lambda	A grid of lambda1 is calculating by our algorithm and we can use an other grid of lambda if we want.
lambda2	The defaut value is 0.
pf	A vector if we want penalized some predector
pf2	A vector if we want penalized some predector
exclude
dfmax
pmax
standardize
eps
maxit
delta

Value

The result is object of 9 items.

References

HHSVM-CEN

Examples

data(test)
x=test$x
y=test$y
m=gcdclust(x, y,KK=3,lambda2=1)

KarimOualkacha/HHSVM-ClusterNet documentation built on May 7, 2019, 12:28 p.m.