mlr_pipeops_fda.fourier: Fast Fourier transform features

In mlr3fda: Extending 'mlr3' to Functional Data Analysis

Fast Fourier transform features

Description

This PipeOp extracts Fourier coefficients from functional columns. For more details, see stats::fft(), which is called internally. Only the one-sided spectrum is returned since the input is real-valued (Oppenheim and Schafer, 2010).

Parameters

The parameters are the parameters inherited from PipeOpTaskPreprocSimple, as well as the following parameters:

• type :: character(1)
  Which feature to extract from the Fourier coefficients. "amplitude" returns the magnitude. "phase" returns the phase shift in degrees (values in [-180, 180]). Initial value is "phase".

Super classes

mlr3pipelines::PipeOp -> mlr3pipelines::PipeOpTaskPreproc -> mlr3pipelines::PipeOpTaskPreprocSimple -> PipeOpFDAFourier

Methods

Public methods

• PipeOpFDAFourier$new()

• PipeOpFDAFourier$clone()

Inherited methods

• mlr3pipelines::PipeOp$help()
• mlr3pipelines::PipeOp$predict()
• mlr3pipelines::PipeOp$print()
• mlr3pipelines::PipeOp$train()

---

Method new()

Initializes a new instance of this Class.

Usage

PipeOpFDAFourier$new(id = "fda.fourier", param_vals = list())

Arguments

id

(character(1))
Identifier of resulting object, default is "fda.fourier".

param_vals

(named list())
List of hyperparameter settings, overwriting the hyperparameter settings that would otherwise be set during construction. Default list().

---

Method clone()

The objects of this class are cloneable with this method.

Usage

PipeOpFDAFourier$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

References

Oppenheim, V A, Schafer, W R (2010). Discrete-Time Signal Processing, 3rd edition. Pearson. ISBN 978-0131988422.

Examples

task = tsk("fuel")
po_fourier = po("fda.fourier")
task_fourier = po_fourier$train(list(task))[[1L]]
task_fourier$data()

mlr3fda documentation built on March 31, 2026, 1:06 a.m.