feature_selector: Create a feature selection specification

In bbuchsbaum/rMVPA: Multivoxel Pattern Analysis in R

Create a feature selection specification

Description

This function creates a feature selection specification using the provided method, cutoff type, and cutoff value.

Usage

feature_selector(method, cutoff_type, cutoff_value)

Arguments

method	The type of feature selection method to use. Supported methods are "FTest" and "catscore".
cutoff_type	The type of threshold used to select features. Supported cutoff types are "top_k" and "top_p".
cutoff_value	The numeric value of the threshold cutoff.

Details

The available feature selection methods are: - FTest: Computes a one-way ANOVA for every column in the feature matrix. - catscore: Computes a correlation adjusted t-test for every column in the matrix using sda.ranking from the sda package.

Value

A list with a class name equal to the method argument.

Examples

fsel <- feature_selector("FTest", "top_k", 1000)
fsel <- feature_selector("FTest", "top_p", .1)
class(fsel) == "FTest"

bbuchsbaum/rMVPA documentation built on April 28, 2024, 6:30 a.m.