L_score: Leave one feature out cross validation loss

In Monoxido45/PhyloHclust: What the Package Does (One Line, Title Case)

Leave one feature out cross validation loss

Description

Leave one feature out cross validation loss for a given dendrogram and dataset. This is an auxiliary function to the main functions CVL and PFIS but can be used separately.

Usage

L_score(dend, original_data, seed = 99, ncores = 2, score = FALSE)

Arguments

dend	Dendrogram of interest (an object of class "phylo")
original_data	Entire dataset (data.frame or matrix) or feature of interest (numeric or factor vector) to compute cross validate loss
seed	Fixed random seed for SIMMAP algorithm. Default is 99.
ncores	Number of cores to paralelize the computing process. Default is 2.
score	Set whether a vector of leave one feature out loss should be computed (TRUE) or compute the overall mean of the loss directly (FALSE)

Value

A numeric vector of leave one feature out losses.

Monoxido45/PhyloHclust documentation built on Sept. 25, 2024, 3:17 a.m.