Skip to contents

Runs a resampling (possibly in parallel): Repeatedly apply Learner learner on a training set of Task task to train a model, then use the trained model to predict observations of a test set. Training and test sets are defined by the Resampling resampling.


  store_models = FALSE,
  store_backends = TRUE,
  encapsulate = NA_character_,
  allow_hotstart = FALSE,
  clone = c("task", "learner", "resampling"),
  unmarshal = TRUE









Store the fitted model in the resulting object= Set to TRUE if you want to further analyse the models or want to extract information like variable importance.


Keep the DataBackend of the Task in the ResampleResult? Set to TRUE if your performance measures require a Task, or to analyse results more conveniently. Set to FALSE to reduce the file size and memory footprint after serialization. The current default is TRUE, but this eventually will be changed in a future release.


If not NA, enables encapsulation by setting the field Learner$encapsulate to one of the supported values: "none" (disable encapsulation), "try" (captures errors but output is printed to the console and not logged), "evaluate" (execute via evaluate) and "callr" (start in external session via callr). If NA, encapsulation is not changed, i.e. the settings of the individual learner are active. Additionally, if encapsulation is set to "evaluate" or "callr", the fallback learner is set to the featureless learner if the learner does not already have a fallback configured.


Determines if learner(s) are hot started with trained models in $hotstart_stack. See also HotstartStack.


Select the input objects to be cloned before proceeding by providing a set with possible values "task", "learner" and "resampling" for Task, Learner and Resampling, respectively. Per default, all input objects are cloned.


Whether to unmarshal learners that were marshaled during the execution. Setting this to FALSE does not guarantee that the learners are marshaled. For example, with sequential execution and no encapsulation, marshaling is not necessary. If you want to ensure that all learners are in marshaled form, you need to call $marshal() on the result object.


The fitted models are discarded after the predictions have been computed in order to reduce memory consumption. If you need access to the models for later analysis, set store_models to TRUE.

Predict Sets

If you want to compare the performance of a learner on the training with the performance on the test set, you have to configure the Learner to predict on multiple sets by setting the field predict_sets to c("train", "test") (default is "test"). Each set yields a separate Prediction object during resampling. In the next step, you have to configure the measures to operate on the respective Prediction object:

m1 = msr("classif.ce", id = "ce.train", predict_sets = "train")
m2 = msr("classif.ce", id = "ce.test", predict_sets = "test")

The (list of) created measures can finally be passed to $aggregate() or $score().


This function can be parallelized with the future package. One job is one resampling iteration, and all jobs are send to an apply function from future.apply in a single batch. To select a parallel backend, use future::plan().

Progress Bars

This function supports progress bars via the package progressr. Simply wrap the function call in progressr::with_progress() to enable them. Alternatively, call progressr::handlers() with global = TRUE to enable progress bars globally. We recommend the progress package as backend which can be enabled with progressr::handlers("progress").


The mlr3 uses the lgr package for logging. lgr supports multiple log levels which can be queried with getOption("lgr.log_levels").

To suppress output and reduce verbosity, you can lower the log from the default level "info" to "warn":


To get additional log output for debugging, increase the log level to "debug" or "trace":


To log to a file or a data base, see the documentation of lgr::lgr-package.

See also


task = tsk("penguins")
learner = lrn("classif.rpart")
resampling = rsmp("cv")

# Explicitly instantiate the resampling for this task for reproduciblity

rr = resample(task, learner, resampling)
#> <ResampleResult> with 10 resampling iterations
#>   task_id    learner_id resampling_id iteration warnings errors
#>  penguins classif.rpart            cv         1        0      0
#>  penguins classif.rpart            cv         2        0      0
#>  penguins classif.rpart            cv         3        0      0
#>  penguins classif.rpart            cv         4        0      0
#>  penguins classif.rpart            cv         5        0      0
#>  penguins classif.rpart            cv         6        0      0
#>  penguins classif.rpart            cv         7        0      0
#>  penguins classif.rpart            cv         8        0      0
#>  penguins classif.rpart            cv         9        0      0
#>  penguins classif.rpart            cv        10        0      0

# Retrieve performance
#>      task_id    learner_id resampling_id iteration classif.ce
#>       <char>        <char>        <char>     <int>      <num>
#>  1: penguins classif.rpart            cv         1 0.00000000
#>  2: penguins classif.rpart            cv         2 0.00000000
#>  3: penguins classif.rpart            cv         3 0.02857143
#>  4: penguins classif.rpart            cv         4 0.00000000
#>  5: penguins classif.rpart            cv         5 0.17647059
#>  6: penguins classif.rpart            cv         6 0.05882353
#>  7: penguins classif.rpart            cv         7 0.05882353
#>  8: penguins classif.rpart            cv         8 0.02941176
#>  9: penguins classif.rpart            cv         9 0.11764706
#> 10: penguins classif.rpart            cv        10 0.05882353
#> Hidden columns: task, learner, resampling, prediction
#> classif.ce 
#> 0.05285714 

# merged prediction objects of all resampling iterations
pred = rr$prediction()
#>            truth
#> response    Adelie Chinstrap Gentoo
#>   Adelie       145         6      0
#>   Chinstrap      7        59      2
#>   Gentoo         0         3    122

# Repeat resampling with featureless learner
rr_featureless = resample(task, lrn("classif.featureless"), resampling)

# Convert results to BenchmarkResult, then combine them
bmr1 = as_benchmark_result(rr)
bmr2 = as_benchmark_result(rr_featureless)
#> <BenchmarkResult> of 20 rows with 2 resampling runs
#>  nr  task_id          learner_id resampling_id iters warnings errors
#>   1 penguins       classif.rpart            cv    10        0      0
#>   2 penguins classif.featureless            cv    10        0      0