fdataTransform: Create a preprocessing pipeline function

In maierhofert/classiFunc: Classification of Functional Data

Description

Internal function, documented due to the importance of its concept. Creates a pipeline function to do all the preprocessing needed in classiKnn and classiKernel. This is helpful to ensure that the data preprocessing (imputation of missing values, derivation) is carried out in exactly the same way for the training and the test set in predict.classiKnn and predict.classiKernel.

Usage

fdataTransform(grid, nderiv, derived, evenly.spaced, no.missing, deriv.method,
  ...)

Arguments

grid, nderiv, derived, evenly.spaced, no.missing, deriv.method	see classiKnn
...	additional arguments to fda::smooth.basis

Value

Pipeline function taking one argument fdata. The returned function carries out all the preprocessing needed for the calling model of class classiKnn.

maierhofert/classiFunc documentation built on May 21, 2019, 11:06 a.m.