AMA_step_size: Compute step size Anderson-Morely upper bound on the largest...
In scvxclustr: Sparse Convex Clustering

Description Usage Arguments Examples

View source: R/util.r

AMA_step_size computes a step size based on the better of two bounds derived by Anderson and Morely.

1	AMA_step_size(w, n)

`w`	vector of weights
`n`	number of points to cluster

data(mammals)
X <- as.matrix(mammals[,-1])
X <- t(scale(X,center=TRUE,scale=FALSE))
n <- ncol(X)

## Pick some weights and a sequence of regularization parameters.
k <- 5
phi <- 0.5
w <- kernel_weights(X,phi)
w <- knn_weights(w,k,n)
AMA_step_size(w,n)