mlr_pipeops_trafopred_regrsurv_pem: PipeOpPredRegrSurvPEM

In mlr-org/mlr3proba: Probabilistic Supervised Learning for 'mlr3'

PipeOpPredRegrSurvPEM

Description

Transform PredictionRegr to PredictionSurv. The predicted piece-wise constant hazards contained in PredictionRegr are transformed into survival probabilities and wrapped in a PredictionSurv object.

We compute the survival probability from the predicted hazards using the following relation:

S(t | \mathbf{x}) = \exp \left( - \int_{0}^{t} \lambda(s | \mathbf{x}) \, ds \right) = \exp \left( - \sum_{j = 1}^{J} \lambda(j | \mathbf{x}) d_j\, \right),

where j = 1, \ldots, J denotes the interval, t the time, and d_j the duration of interval j.

For a more detailed description of PEM, refer to pipeline_survtoregr_pem or the referred article.

Dictionary

This PipeOp can be instantiated via the dictionary mlr3pipelines::mlr_pipeops or with the associated sugar function mlr3pipelines::po():

PipeOpPredRegrSurvPEM$new()
mlr_pipeops$get("trafopred_regrsurv_pem")
po("trafopred_regrsurv_pem")

Input and Output Channels

The input consists of a PredictionRegr and a data.table containing the transformed data. The PredictionRegr is provided by the mlr3::LearnerRegr, while the data.table is generated by PipeOpTaskSurvRegrPEM. The output is the input PredictionRegr transformed to a PredictionSurv. Only works during prediction phase.

Super class

mlr3pipelines::PipeOp -> PipeOpPredRegrSurvPEM

Active bindings

predict_type

(character(1))
Returns the active predict type of this PipeOp, which is "crank"

Methods

Public methods

• PipeOpPredRegrSurvPEM$new()

• PipeOpPredRegrSurvPEM$clone()

Inherited methods

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

---

Method new()

Creates a new instance of this R6 class.

Usage

PipeOpPredRegrSurvPEM$new(id = "trafopred_regrsurv_pem")

Arguments

id

(character(1))
Identifier of the resulting object.

---

Method clone()

The objects of this class are cloneable with this method.

Usage

PipeOpPredRegrSurvPEM$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

References

Bender, Andreas, Groll, Andreas, Scheipl, Fabian (2018). “A generalized additive model approach to time-to-event analysis.” Statistical Modelling, 18(3-4), 299–321. https://doi.org/10.1177/1471082X17748083.

See Also

pipeline_survtoregr_pem

Other PipeOps: mlr_pipeops_survavg, mlr_pipeops_trafotask_survregr_pem

Other Transformation PipeOps: mlr_pipeops_trafopred_classifsurv_IPCW, mlr_pipeops_trafopred_classifsurv_disctime, mlr_pipeops_trafotask_survclassif_IPCW, mlr_pipeops_trafotask_survclassif_disctime, mlr_pipeops_trafotask_survregr_pem

mlr-org/mlr3proba documentation built on April 12, 2025, 4:38 p.m.