PreProcessing: Preprocess the input to a classification function

In RSSL: Implementations of Semi-Supervised Learning Approaches for Classification

Preprocess the input to a classification function

Description

The following actions are carried out: 1. data.frames are converted to matrix form and labels converted to an indicator matrix 2. An intercept column is added if requested 3. centering and scaling is applied if requested.

Usage

PreProcessing(X, y, X_u = NULL, scale = FALSE, intercept = FALSE,
  x_center = FALSE, use_Xu_for_scaling = TRUE)

Arguments

X	Design matrix, intercept term is added within the function
y	Vector or factor with class assignments
X_u	Design matrix of the unlabeled observations
scale	If TRUE, apply a z-transform to the design matrix X
intercept	Whether to include an intercept in the design matrices
x_center	logical (default: TRUE); Whether the feature vectors should be centered
use_Xu_for_scaling	logical (default: TRUE); Should the unlabeled data be used to determine scaling?

Value

list object with the following objects:

X	design matrix of the labeled data
y	integer vector indicating the labels of the labeled data
X_u	design matrix of the unlabeled data
classnames	names of the classes corresponding to the integers in y
scaling	a scaling object used to scale the test observations in the same way as the training set
modelform	a formula object containing the used model

RSSL documentation built on March 31, 2023, 7:27 p.m.