mlr v2.10


Monthly downloads



by Bernd Bischl

Machine Learning in R

Interface to a large number of classification and regression techniques, including machine-readable parameter descriptions. There is also an experimental extension for survival analysis, clustering and general, example-specific cost-sensitive learning. Generic resampling, including cross-validation, bootstrapping and subsampling. Hyperparameter tuning with modern optimization techniques, for single- and multi-objective problems. Filter and wrapper methods for feature selection. Extension of basic learners with additional operations common in machine learning, also allowing for easy nested resampling. Most operations can be parallelized.

Functions in mlr

Name Description
benchmark Benchmark experiment for multiple learners and tasks.
capLargeValues Convert large/infinite numeric values in a data.frame or task.
convertMLBenchObjToTask Convert a machine learning benchmark / demo object from package mlbench to a task.
classif.featureless Featureless classification learner.
costiris.task Iris cost-sensitive classification task.
dropFeatures Drop some features of task.
downsample Downsample (subsample) a task or a data.frame.
crossover Crossover.
configureMlr Configures the behavior of the package.
createDummyFeatures Generate dummy variables for factor features.
convertBMRToRankMatrix Convert BenchmarkResult to a rank-matrix.
friedmanTestBMR Perform overall Friedman test for a BenchmarkResult.
FeatSelControl Create control structures for feature selection.
generateFeatureImportanceData Generate feature importance.
friedmanPostHocTestBMR Perform a posthoc Friedman-Nemenyi test.
estimateRelativeOverfitting Estimate relative overfitting.
generateCritDifferencesData Generate data for critical-differences plot.
generateCalibrationData Generate classifier calibration data.
estimateResidualVariance Estimate the residual variance.
FailureModel Failure model.
filterFeatures Filter features by thresholding filter values.
getBMRFeatSelResults Extract the feature selection results from a benchmark result.
generateFilterValuesData Calculates feature filter values.
generateHyperParsEffectData Generate hyperparameter effect data.
getBMRLearnerIds Return learner ids used in benchmark.
generateThreshVsPerfData Generate threshold vs. performance(s) for 2-class classification.
generateLearningCurveData Generates a learning curve.
getBMRFilteredFeatures Extract the feature selection results from a benchmark result.
generateFunctionalANOVAData Generate a functional ANOVA decomposition
getBMRAggrPerformances Extract the aggregated performance values from a benchmark result.
generatePartialDependenceData Generate partial dependence.
getBMRTuneResults Extract the tuning results from a benchmark result.
getBMRPerformances Extract the test performance values from a benchmark result.
getBMRMeasures Return measures used in benchmark.
getBMRTaskIds Return task ids used in benchmark.
getBMRTaskDescriptions Extract all task descriptions from benchmark result.
getBMRMeasureIds Return measures IDs used in benchmark.
getBMRLearners Return learners used in benchmark.
getBMRModels Extract all models from benchmark result.
getBMRLearnerShortNames Return learner short.names used in benchmark.
getBMRPredictions Extract the predictions from a benchmark result.
getFeatSelResult Returns the selected feature set and optimization path after training.
getFilteredFeatures Returns the filtered features.
getFilterValues Calculates feature filter values.
getCaretParamSet Get tuning parameters from a learner of the caret R-package.
getClassWeightParam Get the class weight parameter of a learner.
getHomogeneousEnsembleModels Deprecated, use
getHyperPars Get current parameter settings for a learner.
getConfMatrix Confusion matrix.
getFailureModelMsg Return error message of FailureModel.
getDefaultMeasure Get default measure.
getMlrOptions Returns a list of mlr's options.
getRRTaskDescription Get task description from resample results.
getProbabilities Deprecated, use
getTaskClassLevels Get the class levels for classification and multilabel tasks.
getTaskCosts Extract costs in task.
getLearnerPackages Get the required R packages of the learner.
getStackedBaseLearnerPredictions Returns the predictions for each base learner.
getLearnerParamSet Get the parameter set of the learner.
getPredictionResponse Get response / truth from prediction object.
getMultilabelBinaryPerformances Retrieve binary classification measures for multilabel classification predictions.
getFeatureImportance Calculates feature importance values for trained models.
getLearnerShortName Get the short name of the learner.
getFeatureImportanceLearner.regr.randomForestSRC Calculates feature importance values for a given learner.
getRRPredictions Get predictions from resample results.
getRRPredictionList Get list of predictions for train and test set of each single resample iteration.
getTaskNFeats Get number of features in task.
getTaskTargetNames Get the name(s) of the target column(s).
getTaskSize Get number of observations in task.
getTaskId Get the id of the task.
getLearnerType Get the type of the learner.
listFilterMethods List filter methods.
listLearners Find matching learning algorithms.
makeFeatSelWrapper Fuse learner with feature selection.
makeModelMultiplexer Create model multiplexer for model selection to tune over multiple possible models.
makeModelMultiplexerParamSet Creates a parameter set for model multiplexer tuning.
getParamSet Get a description of all possible parameter settings for a learner.
getPredictionProbabilities Get probabilities for some classes.
makeWeightedClassesWrapper Wraps a classifier for weighted fitting where each class receives a weight.
makeWrappedModel Induced model of learner.
makeDownsampleWrapper Fuse learner with simple downsampling (subsampling).
makeBaggingWrapper Fuse learner with the bagging technique.
makeAggregation Specify your own aggregation of measures.
analyzeFeatSelResult Show and visualize the steps of feature selection.
makeFilterWrapper Fuse learner with a feature filter method.
makeOverBaggingWrapper Fuse learner with the bagging technique and oversampling for imbalancy correction.
makePreprocWrapper Fuse learner with preprocessing.
makeFilter Create a feature filter.
makeMultilabelBinaryRelevanceWrapper Use binary relevance method to create a multilabel learner.
agri.task European Union Agricultural Workforces clustering task.
makeMulticlassWrapper Fuse learner with multiclass method.
plotFilterValues Plot filter values using ggplot2.
plotFilterValuesGGVIS Plot filter values using ggvis.
plotHyperParsEffect Plot the hyperparameter effects data
getLearnerId Get the ID of the learner.
bh.task Boston Housing regression task.
summarizeColumns Summarize columns of data.frame or task.
subsetTask Subset data in task.
plotLearnerPrediction Visualizes a learning algorithm on a 1D or 2D data set.
selectFeatures Feature selection by wrapper approach.
RLearner Internal construction / wrapping of learner object.
getNestedTuneResultsOptPathDf Get the
getLearnerModel Get underlying R model of learner integrated into mlr.
getNestedTuneResultsX Get the tuned hyperparameter settings from a nested tuning.
getTaskFormula Get formula of a task.
getTaskFeatureNames Get feature names of task.
iris.task Iris classification task.
isFailureModel Is the model a FailureModel?
LearnerProperties Query properties of learners.
aggregations Aggregation methods.
addRRMeasure Compute new measures for existing ResampleResult
getTaskType Get the type of the task.
makeCustomResampledMeasure Construct your own resampled performance measure.
makeLearner Create learner object.
makePreprocWrapperCaret Fuse learner with preprocessing.
makeCostSensWeightedPairsWrapper Wraps a classifier for cost-sensitive learning to produce a weighted pairs model.
getTaskTargets Get target data of task.
makeImputeWrapper Fuse learner with an imputation method.
makeUndersampleWrapper Fuse learner with simple ove/underrsampling for imbalancy correction in binary classification.
makeTuneWrapper Fuse learner with tuning.
makeRemoveConstantFeaturesWrapper Fuse learner with removal of constant features preprocessing.
pid.task PimaIndiansDiabetes classification task.
plotBMRBoxplots Create box or violin plots for a BenchmarkResult.
plotPartialDependence Plot a partial dependence with ggplot2.
plotPartialDependenceGGVIS Plot a partial dependence using ggvis.
setAggregation Set aggregation function of measure.
regr.randomForest regression using randomForest.
reimpute Re-impute a data set
setHyperPars Set the hyperparameters of a learner object.
calculateConfusionMatrix Confusion matrix.
calculateROCMeasures Calculate receiver operator measures.
getLearnerParVals Get the parameter values of the learner.
imputations Built-in imputation methods.
impute Impute and re-impute data
joinClassLevels Join some class existing levels to new, larger class levels for classification problems.
getTaskData Extract data in task.
getTaskDescription Get a summarizing task description.
getLearnerPredictType Get the predict type of the learner.
mergeSmallFactorLevels Merges small levels of factors into new level.
mlrFamilies mlr documentation families
plotTuneMultiCritResult Plots multi-criteria results after tuning using ggplot2.
resample Fit models according to a resampling strategy.
plotTuneMultiCritResultGGVIS Plots multi-criteria results after tuning using ggvis.
setPredictType Set the type of predictions the learner should return.
setThreshold Set threshold of prediction object.
tuneThreshold Tune prediction threshold.
tuneParamsMultiCrit Hyperparameter tuning for multiple measures at once.
ResamplePrediction Prediction from resampling.
bc.task Wisconsin Breast Cancer classification task.
asROCRPrediction Converts predictions to a format package ROCR can handle.
getTuneResult Returns the optimal hyperparameters and optimization path after training.
hasProperties Deprecated, use
listMeasures Find matching measures.
makeConstantClassWrapper Wraps a classification learner to support problems where the class label is (almost) constant.
makeCostMeasure Creates a measure for non-standard misclassification costs.
makeFixedHoldoutInstance Generate a fixed holdout instance for resampling.
makeImputeMethod Create a custom imputation method.
lung.task NCCTG Lung Cancer survival task.
makeCostSensClassifWrapper Wraps a classification learner for use in cost-sensitive learning.
makeCostSensRegrWrapper Wraps a regression learner for use in cost-sensitive learning.
learners List of supported learning algorithms.
makeLearners Create multiple learners at once.
makeMeasure Construct performance measure.
plotThreshVsPerf Plot threshold vs. performance(s) for 2-class classification using ggplot2.
plotThreshVsPerfGGVIS Plot threshold vs. performance(s) for 2-class classification using ggvis.
makeMultilabelStackingWrapper Use stacking method (stacked generalization) to create a multilabel learner.
plotCalibration Plot calibration data using ggplot2.
plotCritDifferences Plot critical differences for a selected measure.
makeMultilabelNestedStackingWrapper Use nested stacking method to create a multilabel learner.
predict.WrappedModel Predict new data.
plotViperCharts Visualize binary classification predictions via ViperCharts system.
setLearnerId Set the ID of a learner object.
setPredictThreshold Set the probability threshold the learner should use.
summarizeLevels Summarizes factors of a data.frame by tabling them.
mergeBenchmarkResults Merge different BenchmarkResult objects.
train Train a learning algorithm.
makeSMOTEWrapper Fuse learner with SMOTE oversampling for imbalancy correction in binary classification.
makeStackedLearner Create a stacked learner object.
measures Performance measures.
plotBMRRanksAsBarChart Create a bar chart for ranks in a BenchmarkResult.
plotBMRSummary Plot a benchmark summary.
plotLearningCurve Plot learning curve data using ggplot2.
plotLearningCurveGGVIS Plot learning curve data using ggvis.
predictLearner Predict new data with an R learner.
makeMultilabelClassifierChainsWrapper Use classifier chains method (CC) to create a multilabel learner.
makeMultilabelDBRWrapper Use dependent binary relevance method (DBR) to create a multilabel learner.
learnerArgsToControl Convert arguments to control structure.
regr.featureless Featureless regression learner.
mtcars.task Motor Trend Car Road Tests clustering task.
oversample Over- or undersample binary classification task to handle class imbalancy.
normalizeFeatures Normalize features.
plotResiduals Create residual plots for prediction objects or benchmark results.
performance Measure performance of prediction.
plotROCCurves Plots a ROC curve using ggplot2.
removeHyperPars Remove hyperparameters settings of a learner.
removeConstantFeatures Remove constant features from a data set.
yeast.task Yeast multilabel classification task.
wpbc.task Wisonsin Prognostic Breast Cancer (WPBC) survival task.
trainLearner Train an R learner.
TuneControl Create control structures for tuning.
TuneMultiCritControl Create control structures for multi-criteria tuning.
tuneParams Hyperparameter tuning.
setHyperPars2 Only exported for internal use.
setId Set the id of a learner object.
smote Synthetic Minority Oversampling Technique to handle class imbalancy in binary classification.
sonar.task Sonar classification task.
BenchmarkResult BenchmarkResult object.
Aggregation Aggregation object.
ConfusionMatrix Confusion matrix
FeatSelResult Result of feature selection.
makeResampleInstance Instantiates a resampling strategy object.
makeResampleDesc Create a description object for a resampling strategy.
ResampleResult ResampleResult object.
Prediction Prediction object.
makeClassifTask Create a classification, regression, survival, cluster, cost-sensitive classification or
TaskDesc Description object for task.
TuneResult Result of tuning.
TuneMultiCritResult Result of multi-criteria tuning.
No Results!

Last month downloads


License BSD_2_clause + file LICENSE
Encoding UTF-8
LazyData yes
ByteCompile yes
VignetteBuilder knitr
RoxygenNote 6.0.1
NeedsCompilation yes
Packaged 2017-02-06 21:24:04 UTC; bischl
Repository CRAN
Date/Publication 2017-02-07 10:08:42
suggests ada , adabag , bartMachine , brnn , bst , C50 , care , caret (>= 6.0-57) , class , clue , cluster , clusterSim (>= 0.44-5) , clValid , cmaes , CoxBoost , crs , Cubist , deepnet , DiceKriging , DiceOptim , DiscriMiner , e1071 , earth , elasticnet , elmNN , emoa , evtree , extraTrees , fields , flare , FNN , fpc , frbs , FSelector , gbm , GenSA , ggvis , glmnet , GPfit , h2o (>= , Hmisc , ipred , irace (>= 2.0) , kernlab , kknn , klaR , knitr , kohonen , laGP , LiblineaR , lqa , MASS , mboost , mco , mda , mlbench , mldr , modeltools , mRMRe , neuralnet , nnet , nodeHarvest (>= 0.7-3) , numDeriv , pamr , party , penalized (>= 0.9-47) , pls , PMCMR (>= 4.1) , pROC (>= 1.8) , randomForest , randomForestSRC (>= 2.2.0) , ranger (>= 0.6.0) , RCurl , Rfast , rFerns , rjson , rknn , rmarkdown , robustbase , ROCR , rotationForest , rpart , RRF , rrlda , rsm , RSNNS , RWeka , sda , shiny , smoof , sparsediscrim , sparseLDA , stepPlr , svglite , SwarmSVM , testthat , tgp , , xgboost (>= 0.6-2) , XML
imports backports , BBmisc (>= 1.10) , checkmate (>= 1.8.1) , data.table , ggplot2 , methods , parallelMap (>= 1.3) , stats , stringi , survival , utils
depends ParamHelpers (>= 1.8) , R (>= 3.0.2)
Contributors Pascal Kerschke, Giuseppe Casalicchio, Michel Lang, Jakob Bossek, Tobias Kuehn, Erich Studerus, Lars Kotthoff, Zachary Jones, Schiffner Julia, Bernd Bischl, Jakob Richter, Leonard Judt, Florian Fendt, Janek Thomas, Mason Gallo, Philipp Probst, Xudong Sun, Bruno Vieira

Include our badge in your README