screen_control: Set Parameters for Screening Step of Fuzzy Forests

Description Usage Arguments Value Note References Examples

View source: R/tuning_parameters.R

Description

Creates screen_control object for controlling how feature selection will be carried out on each module.

Usage

1
2
screen_control(drop_fraction = 0.25, keep_fraction = 0.05,
  mtry_factor = 1, min_ntree = 500, ntree_factor = 1)

Arguments

drop_fraction

A number between 0 and 1. Percentage of features dropped at each iteration.

keep_fraction

A number between 0 and 1. Proportion of features from each module that are retained from screening step.

mtry_factor

In the case of regression, mtry is set to ceiling(√(p)*mtry_factor). In the case of classification, mtry is set to ceiling((p/3)*mtry_factor). If either of these numbers is greater than p, mtry is set to p.

min_ntree

Minimum number of trees grown in each random forest.

ntree_factor

A number greater than 1. ntree for each random forest is ntree_factor times the number of features. For each random forest, ntree is set to max(min_ntree, ntree_factor*p).

Value

An object of type screen_control.

Note

This work was partially funded by NSF IIS 1251151 and AMFAR 8721SC.

References

Conn, D., Ngun, T., Ramirez C.M., Li, G. (2019). "Fuzzy Forests: Extending Random Forest Feature Selection for Correlated, High-Dimensional Data." Journal of Statistical Software, 91(9). doi: 10.18637/jss.v091.i09

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
drop_fraction <- .25
keep_fraction <- .1
mtry_factor <- 1
min_ntree <- 5000
ntree_factor <- 5
screen_params <- screen_control(drop_fraction=drop_fraction,
                                keep_fraction=keep_fraction,
                                mtry_factor=mtry_factor,
                                min_ntree=min_ntree,
                                ntree_factor=ntree_factor)

fuzzyforest documentation built on March 25, 2020, 5:09 p.m.