I wondered if anybody would be able to help with some R code for unsupervised recursive feature elimination (uRFE), please?

In short, I ran an unsupervised SOM, partitioned with k-medoids, but would also like to identify uRFE using the silhouette score.

Attached some dummy code below. If possible, I would appreciate help with some code to do uRFE on the SOM, and therefore identify which features are most relevant when clustering

library(aweSOM)

#IRIS DATA

full.data <- iris
train.data <- full.data[, c("Sepal.Length", "Sepal.Width", "Petal.Length", "Petal.Width")]
train.data <- scale(train.data)

#SOM INITITIALISATION + MAP

set.seed(1465)
init <- somInit(train.data, 4, 4)
iris.som <- kohonen::som(train.data, grid = kohonen::somgrid(4, 4, "hexagonal"), 
                         rlen = 100, alpha = c(0.05, 0.01), radius = c(2.65,-2.65), 
                         dist.fcts = "sumofsquares", init = init)

#SOM CLUSTERS

superclust_pam <- cluster::pam(iris.som$codes[[1]], 3)
superclasses_pam <- superclust_pam$clustering