Extract Random Effects from Functional Columns

This is the class that extracts random effects, specifically random intercepts and random slopes, from functional columns. This PipeOp fits a linear mixed model, specifically a random intercept and random slope model, using the lme4::lmer() function. The target variable is the value of the functional feature which is regressed on the functional feature's argument while subject id determines the grouping structure. After model estimation, the random effects are extracted and assigned to the correct id.

Parameters

The parameters are the parameters inherited from PipeOpTaskPreprocSimple.

Naming

The new names append _random_intercept and _random_slope to the corresponding column name of the functional feature.

Super classes

mlr3pipelines::PipeOp -> mlr3pipelines::PipeOpTaskPreproc -> mlr3pipelines::PipeOpTaskPreprocSimple -> PipeOpFDARandomEffect

Methods

Public methods

• PipeOpFDARandomEffect$new()

• PipeOpFDARandomEffect$clone()

Inherited methods

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

---

PipeOpFDARandomEffect$new()

Initializes a new instance of this Class.

Usage

PipeOpFDARandomEffect$new(id = "fda.random_effect", param_vals = list())

Arguments

id

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

param_vals

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

---

PipeOpFDARandomEffect$clone()

The objects of this class are cloneable with this method.

Usage

PipeOpFDARandomEffect$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

task = tsk("dti")
po_fre = po("fda.random_effect")
task_fre = po_fre$train(list(task))[[1L]]