regPath.SDForest: Calculate the regularization path of an SDForest

In SDModels: Spectrally Deconfounded Models

Calculate the regularization path of an SDForest

Description

This function calculates the variable importance of an SDForest and the out-of-bag performance for different complexity parameters.

Usage

## S3 method for class 'SDForest'
regPath(object, cp_seq = NULL, X = NULL, Y = NULL, Q = NULL, copy = TRUE, ...)

Arguments

object	an SDForest object
cp_seq	A sequence of complexity parameters. If NULL, the sequence is calculated automatically using only relevant values.
X	The training data, if NULL the data from the forest object is used.
Y	The training response variable, if NULL the data from the forest object is used.
Q	The transformation matrix, if NULL the data from the forest object is used.
copy	Whether the tree should be copied for the regularization path. If FALSE, the pruning is done in place and will change the SDForest. This might be reasonable, if the SDForest is to large to copy.
...	Further arguments passed to or from other methods.

Value

An object of class paths containing

cp	The sequence of complexity parameters.
varImp_path	A matrix with the variable importance for each complexity parameter.
loss_path	A matrix with the out-of-bag performance for each complexity parameter.
cp_min	The complexity parameter with the lowest out-of-bag performance.
type	Path type

Author(s)

Markus Ulmer

See Also

plot.paths plotOOB regPath.SDTree prune get_cp_seq SDForest

Examples

set.seed(1)
n <- 10
X <- matrix(rnorm(n * 5), nrow = n)
y <- sign(X[, 1]) * 3 + sign(X[, 2]) + rnorm(n)
model <- SDForest(x = X, y = y, Q_type = 'no_deconfounding', cp = 0.5)
paths <- regPath(model)
plotOOB(paths)
plot(paths)

plot(paths, plotly = TRUE)

SDModels documentation built on April 11, 2025, 5:50 p.m.