mlr_pipeops_learner_cv: Wrap a Learner into a PipeOp with Cross-validated Predictions...
In mlr3pipelines: Preprocessing Operators and Pipelines for 'mlr3'
mlr_pipeops_learner_cv R Documentation
Wrap a Learner into a PipeOp with Cross-validated Predictions as Features
Description
Wraps an mlr3::Learner into a PipeOp.
Returns cross-validated predictions during training as a Task and stores a model of the
Learner trained on the whole data in $state. This is used to create a similar
Task during prediction.
Optionally, the fitted models obtained during the resampling phase can be reused for prediction by averaging
their predictions, avoiding the need for an additional fit on the complete training data.
The Task gets features depending on the capsuled Learner's
$predict_type. If the Learner's $predict.type is "response", a feature <ID>.response is created,
for $predict.type "prob" the <ID>.prob.<CLASS> features are created, and for $predict.type "se" the new columns
are <ID>.response and <ID>.se. <ID> denotes the $id of the PipeOpLearnerCV object.
Inherits the $param_set (and therefore $param_set$values) from the Learner it is constructed from.
PipeOpLearnerCV can be used to create "stacking" or "super learning" Graphs that use the output of one Learner
as feature for another Learner. Because the PipeOpLearnerCV erases the original input features, it is often
useful to use PipeOpFeatureUnion to bind the prediction Task to the original input Task.
Format
R6Class object inheriting from PipeOpTaskPreproc/PipeOp.
Construction
PipeOpLearnerCV$new(learner, id = NULL, param_vals = list())
-
learner :: Learner
Learner to use for cross validation / prediction, or a string identifying a
Learner in the mlr3::mlr_learners Dictionary.
This argument is always cloned; to access the Learner inside PipeOpLearnerCV by-reference, use $learner.
-
id :: character(1)
Identifier of the resulting object, internally defaulting to the id of the Learner being wrapped.
-
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
PipeOpLearnerCV has one input channel named "input", taking a Task specific to the Learner
type given to learner during construction; both during training and prediction.
PipeOpLearnerCV has one output channel named "output", producing a Task specific to the Learner
type given to learner during construction; both during training and prediction.
The output is a task with the same target as the input task, with features replaced by predictions made by the Learner.
During training, this prediction is the out-of-sample prediction made by resample, during prediction, this is the
ordinary prediction made on the data by a Learner trained on the training phase data.
State
The $state is set to the $state slot of the Learner object, together with the $state elements inherited from the
PipeOpTaskPreproc. It is a named list with the inherited members, as well as:
-
model :: any
Model created by the Learner's $.train() function.
-
train_log :: data.table with columns class (character), msg (character)
Errors logged during training.
-
train_time :: numeric(1)
Training time, in seconds.
-
predict_log :: NULL | data.table with columns class (character), msg (character)
Errors logged during prediction.
-
predict_time :: NULL | numeric(1)
Prediction time, in seconds.
-
predict_method :: character(1)
"full" when prediction uses a learner fitted on all training data, "cv_ensemble" when predictions are averaged over
models trained on resampling folds.
-
cv_model_states :: NULL | list
Present for predict_method = "cv_ensemble". Contains the states of the learners trained on each resampling fold.
This state is given the class "pipeop_learner_cv_state".
Parameters
The parameters are the parameters inherited from the PipeOpTaskPreproc, as well as the parameters of the Learner wrapped by this object.
Besides that, parameters introduced are:
-
resampling.method :: character(1)
Which resampling method do we want to use. Currently only supports "cv" and "insample". "insample" generates
predictions with the model trained on all training data.
-
resampling.folds :: numeric(1)
Number of cross validation folds. Initialized to 3. Only used for resampling.method = "cv".
-
keep_response :: logical(1)
Only effective during "prob" prediction: Whether to keep response values, if available. Initialized to FALSE.
-
resampling.predict_method :: character(1)
Controls how predictions are produced after training. "full" (default) fits the wrapped learner on the entire training data.
"cv_ensemble" reuses the models fitted during resampling and averages their predictions. This option currently supports
classification and regression learners together with resampling.method = "cv".
-
resampling.prob_aggr :: character(1)
Probability aggregation used when "cv_ensemble" predictions are produced for classification learners that can emit class probabilities.
Shares the semantics with PipeOpClassifAvg: "mean" (linear opinion pool, default) and "log" (log opinion pool / product of experts).
Only present for learners that support "prob" predictions.
-
resampling.prob_aggr_eps :: numeric(1)
Stabilization constant applied when resampling.prob_aggr = "log" to clamp probabilities before taking logarithms.
Defaults to 1e-12. Only present for learners that support "prob" predictions.
-
resampling.se_aggr :: character(1)
Standard error aggregation used when "cv_ensemble" predictions are produced for regression learners with predict_type
containing "se". Shares the definitions with PipeOpRegrAvg, i.e. "predictive", "mean", "within", "between", "none".
Initialized to "predictive" (within-fold variance plus between-fold disagreement) when constructed with a Learner that has predict_type = "se";
otherwise to "none".
Only present for learners that support "se" predictions.
-
resampling.se_aggr_rho :: numeric(1)
Equicorrelation parameter for resampling.se_aggr = "mean", interpreted as in PipeOpRegrAvg. Ignored otherwise.
Defaults to 0 when resampling.se_aggr = "mean".
Only present for learners that support "se" predictions.
Internals
The $state is currently not updated by prediction, so the $state$predict_log and $state$predict_time will always be NULL.
Fields
Fields inherited from PipeOp, as well as:
-
learner :: Learner
Learner that is being wrapped. Read-only.
-
learner_model :: Learner
Learner that is being wrapped. This learner contains the model if the PipeOp is trained. Read-only.
Methods
Methods inherited from PipeOpTaskPreproc/PipeOp.
See Also
https://mlr-org.com/pipeops.html
Other Meta PipeOps:
mlr_pipeops_learner,
mlr_pipeops_learner_pi_cvplus,
mlr_pipeops_learner_quantiles
Examples
library("mlr3")
task = tsk("iris")
learner = lrn("classif.rpart")
lrncv_po = po("learner_cv", learner)
lrncv_po$learner$predict_type = "response"
nop = mlr_pipeops$get("nop")
graph = gunion(list(
lrncv_po,
nop
)) %>>% po("featureunion")
graph$train(task)
graph$pipeops$classif.rpart$learner$predict_type = "prob"
graph$pipeops$classif.rpart$param_set$values$resampling.predict_method = "cv_ensemble"
graph$train(task)
mlr3pipelines documentation built on Nov. 7, 2025, 9:06 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
| mlr_pipeops_learner_cv | R Documentation |
Wrap a Learner into a PipeOp with Cross-validated Predictions as Features
Description
Wraps an mlr3::Learner into a PipeOp.
Returns cross-validated predictions during training as a Task and stores a model of the
Learner trained on the whole data in $state. This is used to create a similar
Task during prediction.
Optionally, the fitted models obtained during the resampling phase can be reused for prediction by averaging
their predictions, avoiding the need for an additional fit on the complete training data.
The Task gets features depending on the capsuled Learner's
$predict_type. If the Learner's $predict.type is "response", a feature <ID>.response is created,
for $predict.type "prob" the <ID>.prob.<CLASS> features are created, and for $predict.type "se" the new columns
are <ID>.response and <ID>.se. <ID> denotes the $id of the PipeOpLearnerCV object.
Inherits the $param_set (and therefore $param_set$values) from the Learner it is constructed from.
PipeOpLearnerCV can be used to create "stacking" or "super learning" Graphs that use the output of one Learner
as feature for another Learner. Because the PipeOpLearnerCV erases the original input features, it is often
useful to use PipeOpFeatureUnion to bind the prediction Task to the original input Task.
Format
R6Class object inheriting from PipeOpTaskPreproc/PipeOp.
Construction
PipeOpLearnerCV$new(learner, id = NULL, param_vals = list())
-
learner::Learner
Learnerto use for cross validation / prediction, or a string identifying aLearnerin themlr3::mlr_learnersDictionary. This argument is always cloned; to access theLearnerinsidePipeOpLearnerCVby-reference, use$learner.
-
id::character(1)Identifier of the resulting object, internally defaulting to theidof theLearnerbeing wrapped. -
param_vals:: namedlist
List of hyperparameter settings, overwriting the hyperparameter settings that would otherwise be set during construction. Defaultlist().
Input and Output Channels
PipeOpLearnerCV has one input channel named "input", taking a Task specific to the Learner
type given to learner during construction; both during training and prediction.
PipeOpLearnerCV has one output channel named "output", producing a Task specific to the Learner
type given to learner during construction; both during training and prediction.
The output is a task with the same target as the input task, with features replaced by predictions made by the Learner.
During training, this prediction is the out-of-sample prediction made by resample, during prediction, this is the
ordinary prediction made on the data by a Learner trained on the training phase data.
State
The $state is set to the $state slot of the Learner object, together with the $state elements inherited from the
PipeOpTaskPreproc. It is a named list with the inherited members, as well as:
-
model::any
Model created by theLearner's$.train()function. -
train_log::data.tablewith columnsclass(character),msg(character)
Errors logged during training. -
train_time::numeric(1)
Training time, in seconds. -
predict_log::NULL|data.tablewith columnsclass(character),msg(character)
Errors logged during prediction. -
predict_time::NULL|numeric(1)Prediction time, in seconds. -
predict_method::character(1)
"full"when prediction uses a learner fitted on all training data,"cv_ensemble"when predictions are averaged over models trained on resampling folds. -
cv_model_states::NULL|list
Present forpredict_method = "cv_ensemble". Contains the states of the learners trained on each resampling fold.
This state is given the class "pipeop_learner_cv_state".
Parameters
The parameters are the parameters inherited from the PipeOpTaskPreproc, as well as the parameters of the Learner wrapped by this object.
Besides that, parameters introduced are:
-
resampling.method::character(1)
Which resampling method do we want to use. Currently only supports"cv"and"insample"."insample"generates predictions with the model trained on all training data. -
resampling.folds::numeric(1)
Number of cross validation folds. Initialized to 3. Only used forresampling.method = "cv". -
keep_response::logical(1)
Only effective during"prob"prediction: Whether to keep response values, if available. Initialized toFALSE. -
resampling.predict_method::character(1)
Controls how predictions are produced after training."full"(default) fits the wrapped learner on the entire training data."cv_ensemble"reuses the models fitted during resampling and averages their predictions. This option currently supports classification and regression learners together withresampling.method = "cv". -
resampling.prob_aggr::character(1)
Probability aggregation used when"cv_ensemble"predictions are produced for classification learners that can emit class probabilities. Shares the semantics withPipeOpClassifAvg:"mean"(linear opinion pool, default) and"log"(log opinion pool / product of experts). Only present for learners that support"prob"predictions. -
resampling.prob_aggr_eps::numeric(1)
Stabilization constant applied whenresampling.prob_aggr = "log"to clamp probabilities before taking logarithms. Defaults to1e-12. Only present for learners that support"prob"predictions. -
resampling.se_aggr::character(1)
Standard error aggregation used when"cv_ensemble"predictions are produced for regression learners withpredict_typecontaining"se". Shares the definitions withPipeOpRegrAvg, i.e."predictive","mean","within","between","none". Initialized to"predictive"(within-fold variance plus between-fold disagreement) when constructed with aLearnerthat haspredict_type = "se"; otherwise to"none".
Only present for learners that support"se"predictions. -
resampling.se_aggr_rho::numeric(1)
Equicorrelation parameter forresampling.se_aggr = "mean", interpreted as inPipeOpRegrAvg. Ignored otherwise. Defaults to0whenresampling.se_aggr = "mean".
Only present for learners that support"se"predictions.
Internals
The $state is currently not updated by prediction, so the $state$predict_log and $state$predict_time will always be NULL.
Fields
Fields inherited from PipeOp, as well as:
-
learner::Learner
Learnerthat is being wrapped. Read-only. -
learner_model::Learner
Learnerthat is being wrapped. This learner contains the model if thePipeOpis trained. Read-only.
Methods
Methods inherited from PipeOpTaskPreproc/PipeOp.
See Also
https://mlr-org.com/pipeops.html
Other Meta PipeOps:
mlr_pipeops_learner,
mlr_pipeops_learner_pi_cvplus,
mlr_pipeops_learner_quantiles
Examples
library("mlr3")
task = tsk("iris")
learner = lrn("classif.rpart")
lrncv_po = po("learner_cv", learner)
lrncv_po$learner$predict_type = "response"
nop = mlr_pipeops$get("nop")
graph = gunion(list(
lrncv_po,
nop
)) %>>% po("featureunion")
graph$train(task)
graph$pipeops$classif.rpart$learner$predict_type = "prob"
graph$pipeops$classif.rpart$param_set$values$resampling.predict_method = "cv_ensemble"
graph$train(task)
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.