mlr_pipeops_multioutsplit: PipeOpSplitMultiout

In mlr-org/mlr3multioutput: Multioutput extension for 'mlr3'

Description

Splits a multioutput Task into several Tasks.

For each target_type in the original multioutput Task, a new Task containing the respective target and task_type is constructed.

This PipeOp creates a Multiplicity, which means that subsequent PipeOps are executed multiple times.

Note that Multiplicity is currently an experimental feature and the implementation or UI may change.

Format

R6Class inheriting from PipeOp.

Construction

PipeOpSplitMultiout$new(id = "multioutsplit", param_vals = list())

• id :: character(1)
  Identifier of the resulting object, default "ovrsplit".

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

Input and Output Channels

PipeOpSplitMultiout has one input channel named "input" taking a TaskMultioutput both during training and prediction.

PipeOpSplitMultiout has one output channel named "output" returning a Multiplicity of Taskss both during training and prediction.

State

The $state contains the original task_types of the TaskMultioutput supplied during training.

Parameters

PipeOpSplitMultiout has no parameters.

Internals

The names of the element of the output Multiplicity are given by the levels of the target.

Should be used in combination with PipeOpPredictionMultioutUnite.

Fields

Only fields inherited from PipeOp.

Methods

Only methods inherited from PipeOp.

Super class

mlr3pipelines::PipeOp -> PipeOpSplitMultiout

Methods

Public methods

• PipeOpSplitMultiout$new()

• PipeOpSplitMultiout$clone()

Inherited methods

• mlr3pipelines::PipeOp$predict()

• mlr3pipelines::PipeOp$print()

• mlr3pipelines::PipeOp$train()

---

Method new()

Initialize a new R6 class.

Usage

PipeOpSplitMultiout$new(id = "targetsplit", param_vals = list())

Arguments

id

character(1)
Identifier of the resulting object, internally defaulting "targetsplit".

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

PipeOpSplitMultiout$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

See Also

Other PipeOps: mlr_pipeops_multioutlrn, mlr_pipeops_multioutunite

Other Multiplicity PipeOps: mlr_pipeops_multioutlrn, mlr_pipeops_multioutunite

Other Experimental Features: mlr_pipeops_multioutlrn, mlr_pipeops_multioutunite

Examples

library(mlr3)
task = tsk("linnerud")
po = po("multioutsplit")
po$train(list(task))
po$predict(list(task))

mlr-org/mlr3multioutput documentation built on Nov. 22, 2020, 1:17 p.m.