Machine Learning in R

[Up] [Top]

Documentation for package ‘mlr’ version 2.3

Help Pages

A B C D E F G H I J L M N O P R S T U W

-- A --

acc Performance measures.
addProperties Set, add, remove or query properties of learners
Aggregation Aggregation object.
aggregations Aggregation methods.
agri.task European Union Agricultural Workforces clustering task
analyzeFeatSelResult Show and visualize the steps of feature selection.
asROCRPrediction Converts predictions to a format package ROCR can handle.
auc Performance measures.

-- B --

b632 Aggregation methods.
b632plus Aggregation methods.
bac Performance measures.
bc.task Wisconsin Breast Cancer classification task
benchmark Benchmark experiment for multiple learners and tasks.
BenchmarkResult Result of a benchmark run.
ber Performance measures.
bh.task Boston Housing regression task
bootstrapB632 Fit models according to a resampling strategy.
bootstrapB632plus Fit models according to a resampling strategy.
bootstrapOOB Fit models according to a resampling strategy.

-- C --

capLargeValues Convert large/infinite numeric values in a data.frame or task.
cindex Performance measures.
ClassifTask Create a classification, regression, survival, cluster, or cost-sensitive classification task.
ClusterTask Create a classification, regression, survival, cluster, or cost-sensitive classification task.
configureMlr Configures the behavior of the package.
costiris.task Iris cost-sensitive classification task
CostSensClassifModel Wraps a classification learner for use in cost-sensitive learning.
CostSensClassifWrapper Wraps a classification learner for use in cost-sensitive learning.
CostSensRegrModel Wraps a regression learner for use in cost-sensitive learning.
CostSensRegrWrapper Wraps a regression learner for use in cost-sensitive learning.
CostSensTask Create a classification, regression, survival, cluster, or cost-sensitive classification task.
CostSensWeightedPairsModel Wraps a classifier for cost-sensitive learning to produce a weighted pairs model.
CostSensWeightedPairsWrapper Wraps a classifier for cost-sensitive learning to produce a weighted pairs model.
createDummyFeatures Generate dummy variables for factor features.
crossover crossover
crossval Fit models according to a resampling strategy.

-- D --

db Performance measures.
downsample Downsample (subsample) a task or a data.frame.
dropFeatures Drop some features of task.
dunn Performance measures.

-- E --

estimateResidualVariance Estimate the residual variance

-- F --

f1 Performance measures.
FailureModel Failure model.
fdr Performance measures.
featperc Performance measures.
FeatSelControl Create control structures for feature selection.
FeatSelControlExhaustive Create control structures for feature selection.
FeatSelControlGA Create control structures for feature selection.
FeatSelControlRandom Create control structures for feature selection.
FeatSelControlSequential Create control structures for feature selection.
FeatSelResult Result of feature selection.
filterFeatures Filter features by thresholding filter values.
FilterValues Result of 'getFilterValues'.
fn Performance measures.
fnr Performance measures.
fp Performance measures.
fpr Performance measures.

-- G --

G1 Performance measures.
G2 Performance measures.
getBMRAggrPerformances Extract the aggregated performance values from a benchmark result.
getBMRFeatSelResults Extract the feature selection results from a benchmark result.
getBMRFilteredFeatures Extract the feature selection results from a benchmark result.
getBMRLearnerIds Return learner ids used in benchmark.
getBMRPerformances Extract the test performance values from a benchmark result.
getBMRPredictions Extract the predictions from a benchmark result.
getBMRTaskIds Return task ids used in benchmark.
getBMRTuneResults Extract the tuning results from a benchmark result.
getConfMatrix Confusion matrix.
getFailureModelMsg Return error message of FailureModel.
getFeatSelResult Returns the selected feature set and optimization path after training.
getFilteredFeatures Returns the filtered features.
getFilterValues Calculates feature filter values.
getHomogeneousEnsembleModels Returns the list of fitted models.
getHyperPars Get current parameter settings for a learner.
getLearnerModel Get underlying R model of learner integrated into mlr.
getMlrOptions Returns a list of mlr's options
getParamSet Get a description of all possible parameter settings for a learner.
getProbabilities Get probabilities for some classes.
getStackedBaseLearnerPredictions Returns the predictions for each base learner.
getTaskCosts Extract costs in task.
getTaskData Extract data in task.
getTaskDescription Get a summarizing task description.
getTaskFeatureNames Get feature names of task.
getTaskFormula Get formula of a task.
getTaskFormulaAsString Get formula of a task.
getTaskId Get the id of the task.
getTaskNFeats Get number of features in task.
getTaskTargetNames Get the name(s) of the target column(s).
getTaskTargets Get target column of task.
getTaskType Get the type of the task.
getTuneResult Returns the optimal hyperparameters and optimization path after training.
gmean Performance measures.
gpr Performance measures.

-- H --

hasProperties Set, add, remove or query properties of learners
holdout Fit models according to a resampling strategy.

-- I --

imputations Built in imputation methods The built-ins are: • 'imputeConstant(const)' for imputation using a constant value, • 'imputeMedian()' for imputation using the median, • 'imputeMode()' for imputation using the mode, • 'imputeMin(multiplier)' for imputing constant values shifted below the minimum using 'min(x) - multiplier * diff(range(x))', • 'imputeMin(multiplier)' for imputing constant values shifted above the maximum using 'max(x) + multiplier * diff(range(x))', • 'imputeNormal(mean, sd)' for imputation using normally distributed random values. Mean and standard deviation will be calculated from the data if not provided. • 'imputeHist(breaks, use.mids)' for imputation using random values with probabilities calculated using 'table' or 'hist'. • 'imputeLearner(learner, preimpute)' for imputations using the response of a classification or regression learner.
impute Impute and re-impute data
imputeConstant Built in imputation methods The built-ins are: • 'imputeConstant(const)' for imputation using a constant value, • 'imputeMedian()' for imputation using the median, • 'imputeMode()' for imputation using the mode, • 'imputeMin(multiplier)' for imputing constant values shifted below the minimum using 'min(x) - multiplier * diff(range(x))', • 'imputeMin(multiplier)' for imputing constant values shifted above the maximum using 'max(x) + multiplier * diff(range(x))', • 'imputeNormal(mean, sd)' for imputation using normally distributed random values. Mean and standard deviation will be calculated from the data if not provided. • 'imputeHist(breaks, use.mids)' for imputation using random values with probabilities calculated using 'table' or 'hist'. • 'imputeLearner(learner, preimpute)' for imputations using the response of a classification or regression learner.
imputeHist Built in imputation methods The built-ins are: • 'imputeConstant(const)' for imputation using a constant value, • 'imputeMedian()' for imputation using the median, • 'imputeMode()' for imputation using the mode, • 'imputeMin(multiplier)' for imputing constant values shifted below the minimum using 'min(x) - multiplier * diff(range(x))', • 'imputeMin(multiplier)' for imputing constant values shifted above the maximum using 'max(x) + multiplier * diff(range(x))', • 'imputeNormal(mean, sd)' for imputation using normally distributed random values. Mean and standard deviation will be calculated from the data if not provided. • 'imputeHist(breaks, use.mids)' for imputation using random values with probabilities calculated using 'table' or 'hist'. • 'imputeLearner(learner, preimpute)' for imputations using the response of a classification or regression learner.
imputeLearner Built in imputation methods The built-ins are: • 'imputeConstant(const)' for imputation using a constant value, • 'imputeMedian()' for imputation using the median, • 'imputeMode()' for imputation using the mode, • 'imputeMin(multiplier)' for imputing constant values shifted below the minimum using 'min(x) - multiplier * diff(range(x))', • 'imputeMin(multiplier)' for imputing constant values shifted above the maximum using 'max(x) + multiplier * diff(range(x))', • 'imputeNormal(mean, sd)' for imputation using normally distributed random values. Mean and standard deviation will be calculated from the data if not provided. • 'imputeHist(breaks, use.mids)' for imputation using random values with probabilities calculated using 'table' or 'hist'. • 'imputeLearner(learner, preimpute)' for imputations using the response of a classification or regression learner.
imputeMax Built in imputation methods The built-ins are: • 'imputeConstant(const)' for imputation using a constant value, • 'imputeMedian()' for imputation using the median, • 'imputeMode()' for imputation using the mode, • 'imputeMin(multiplier)' for imputing constant values shifted below the minimum using 'min(x) - multiplier * diff(range(x))', • 'imputeMin(multiplier)' for imputing constant values shifted above the maximum using 'max(x) + multiplier * diff(range(x))', • 'imputeNormal(mean, sd)' for imputation using normally distributed random values. Mean and standard deviation will be calculated from the data if not provided. • 'imputeHist(breaks, use.mids)' for imputation using random values with probabilities calculated using 'table' or 'hist'. • 'imputeLearner(learner, preimpute)' for imputations using the response of a classification or regression learner.
imputeMean Built in imputation methods The built-ins are: • 'imputeConstant(const)' for imputation using a constant value, • 'imputeMedian()' for imputation using the median, • 'imputeMode()' for imputation using the mode, • 'imputeMin(multiplier)' for imputing constant values shifted below the minimum using 'min(x) - multiplier * diff(range(x))', • 'imputeMin(multiplier)' for imputing constant values shifted above the maximum using 'max(x) + multiplier * diff(range(x))', • 'imputeNormal(mean, sd)' for imputation using normally distributed random values. Mean and standard deviation will be calculated from the data if not provided. • 'imputeHist(breaks, use.mids)' for imputation using random values with probabilities calculated using 'table' or 'hist'. • 'imputeLearner(learner, preimpute)' for imputations using the response of a classification or regression learner.
imputeMedian Built in imputation methods The built-ins are: • 'imputeConstant(const)' for imputation using a constant value, • 'imputeMedian()' for imputation using the median, • 'imputeMode()' for imputation using the mode, • 'imputeMin(multiplier)' for imputing constant values shifted below the minimum using 'min(x) - multiplier * diff(range(x))', • 'imputeMin(multiplier)' for imputing constant values shifted above the maximum using 'max(x) + multiplier * diff(range(x))', • 'imputeNormal(mean, sd)' for imputation using normally distributed random values. Mean and standard deviation will be calculated from the data if not provided. • 'imputeHist(breaks, use.mids)' for imputation using random values with probabilities calculated using 'table' or 'hist'. • 'imputeLearner(learner, preimpute)' for imputations using the response of a classification or regression learner.
imputeMin Built in imputation methods The built-ins are: • 'imputeConstant(const)' for imputation using a constant value, • 'imputeMedian()' for imputation using the median, • 'imputeMode()' for imputation using the mode, • 'imputeMin(multiplier)' for imputing constant values shifted below the minimum using 'min(x) - multiplier * diff(range(x))', • 'imputeMin(multiplier)' for imputing constant values shifted above the maximum using 'max(x) + multiplier * diff(range(x))', • 'imputeNormal(mean, sd)' for imputation using normally distributed random values. Mean and standard deviation will be calculated from the data if not provided. • 'imputeHist(breaks, use.mids)' for imputation using random values with probabilities calculated using 'table' or 'hist'. • 'imputeLearner(learner, preimpute)' for imputations using the response of a classification or regression learner.
imputeMode Built in imputation methods The built-ins are: • 'imputeConstant(const)' for imputation using a constant value, • 'imputeMedian()' for imputation using the median, • 'imputeMode()' for imputation using the mode, • 'imputeMin(multiplier)' for imputing constant values shifted below the minimum using 'min(x) - multiplier * diff(range(x))', • 'imputeMin(multiplier)' for imputing constant values shifted above the maximum using 'max(x) + multiplier * diff(range(x))', • 'imputeNormal(mean, sd)' for imputation using normally distributed random values. Mean and standard deviation will be calculated from the data if not provided. • 'imputeHist(breaks, use.mids)' for imputation using random values with probabilities calculated using 'table' or 'hist'. • 'imputeLearner(learner, preimpute)' for imputations using the response of a classification or regression learner.
imputeNormal Built in imputation methods The built-ins are: • 'imputeConstant(const)' for imputation using a constant value, • 'imputeMedian()' for imputation using the median, • 'imputeMode()' for imputation using the mode, • 'imputeMin(multiplier)' for imputing constant values shifted below the minimum using 'min(x) - multiplier * diff(range(x))', • 'imputeMin(multiplier)' for imputing constant values shifted above the maximum using 'max(x) + multiplier * diff(range(x))', • 'imputeNormal(mean, sd)' for imputation using normally distributed random values. Mean and standard deviation will be calculated from the data if not provided. • 'imputeHist(breaks, use.mids)' for imputation using random values with probabilities calculated using 'table' or 'hist'. • 'imputeLearner(learner, preimpute)' for imputations using the response of a classification or regression learner.
imputeUniform Built in imputation methods The built-ins are: • 'imputeConstant(const)' for imputation using a constant value, • 'imputeMedian()' for imputation using the median, • 'imputeMode()' for imputation using the mode, • 'imputeMin(multiplier)' for imputing constant values shifted below the minimum using 'min(x) - multiplier * diff(range(x))', • 'imputeMin(multiplier)' for imputing constant values shifted above the maximum using 'max(x) + multiplier * diff(range(x))', • 'imputeNormal(mean, sd)' for imputation using normally distributed random values. Mean and standard deviation will be calculated from the data if not provided. • 'imputeHist(breaks, use.mids)' for imputation using random values with probabilities calculated using 'table' or 'hist'. • 'imputeLearner(learner, preimpute)' for imputations using the response of a classification or regression learner.
iris.task Iris classification task
isFailureModel Is the model a FailureModel?

-- J --

joinClassLevels Join some class existing levels to new, larger class levels for classification problems.

-- L --

Learner Create learner object.
learnerArgsToControl Convert arguments to control structure.
LearnerProperties Set, add, remove or query properties of learners
learners List of supported learning algorithms.
listFilterMethods List filter methods
listLearners Find matching learning algorithms.
listLearners.character Find matching learning algorithms.
listLearners.default Find matching learning algorithms.
listLearners.Task Find matching learning algorithms.
listMeasures Find matching measures.
listMeasures.character Find matching measures.
listMeasures.default Find matching measures.
listMeasures.Task Find matching measures.
lung.task NCCTG Lung Cancer survival task

-- M --

mae Performance measures.
makeAggregation Specifiy your own aggregation of measures
makeBaggingWrapper Fuse learner with the bagging technique.
makeClassifTask Create a classification, regression, survival, cluster, or cost-sensitive classification task.
makeClusterTask Create a classification, regression, survival, cluster, or cost-sensitive classification task.
makeCostMeasure Creates a measure for non-standard misclassification costs.
makeCostSensClassifWrapper Wraps a classification learner for use in cost-sensitive learning.
makeCostSensRegrWrapper Wraps a regression learner for use in cost-sensitive learning.
makeCostSensTask Create a classification, regression, survival, cluster, or cost-sensitive classification task.
makeCostSensWeightedPairsWrapper Wraps a classifier for cost-sensitive learning to produce a weighted pairs model.
makeCustomResampledMeasure Construct your own resampled performance measure.
makeDownsampleWrapper Fuse learner with simple downsampling (subsampling).
makeFeatSelControlExhaustive Create control structures for feature selection.
makeFeatSelControlGA Create control structures for feature selection.
makeFeatSelControlRandom Create control structures for feature selection.
makeFeatSelControlSequential Create control structures for feature selection.
makeFeatSelWrapper Fuse learner with feature selection.
makeFilter Create a feature filter
makeFilterWrapper Fuse learner with a feature filter method.
makeFixedHoldoutInstance Generate a fixed holdout instance for resampling.
makeImputeMethod Create a custom imputation method.
makeImputeWrapper Fuse learner with an imputation method.
makeLearner Create learner object.
makeMeasure Construct performance measure.
makeModelMultiplexer Create model multiplexer for model selection to tune over multiple possible models.
makeModelMultiplexerParamSet Creates a parameter set for model multiplexer tuning.
makeMulticlassWrapper Fuse learner with multiclass method.
makeOverBaggingWrapper Fuse learner with the bagging technique and oversampling for imbalancy correction.
makeOversampleWrapper Fuse learner with simple ove/underrsampling for imbalancy correction in binary classification.
makePreprocWrapper Fuse learner with preprocessing.
makePreprocWrapperCaret Fuse learner with preprocessing
makeRegrTask Create a classification, regression, survival, cluster, or cost-sensitive classification task.
makeResampleDesc Create a description object for a resampling strategy.
makeResampleInstance Instantiates a resampling strategy object.
makeRLearner Internal construction / wrapping of learner object.
makeRLearnerClassif Internal construction / wrapping of learner object.
makeRLearnerCluster Internal construction / wrapping of learner object.
makeRLearnerRegr Internal construction / wrapping of learner object.
makeRLearnerSurv Internal construction / wrapping of learner object.
makeSMOTEWrapper Fuse learner with SMOTE oversampling for imbalancy correction in binary classification.
makeStackedLearner Create a stacked learner object.
makeSurvTask Create a classification, regression, survival, cluster, or cost-sensitive classification task.
makeTuneControlCMAES Create control structures for tuning.
makeTuneControlGenSA Create control structures for tuning.
makeTuneControlGrid Create control structures for tuning.
makeTuneControlIrace Create control structures for tuning.
makeTuneControlRandom Create control structures for tuning.
makeTuneMultiCritControlGrid Create control structures for multi-criteria tuning.
makeTuneMultiCritControlNSGA2 Create control structures for multi-criteria tuning.
makeTuneMultiCritControlRandom Create control structures for multi-criteria tuning.
makeTuneWrapper Fuse learner with tuning.
makeUndersampleWrapper Fuse learner with simple ove/underrsampling for imbalancy correction in binary classification.
makeWeightedClassesWrapper Wraps a classifier for weighted fitting where each class receives a weight.
makeWrappedModel Induced model of learner.
mcc Performance measures.
mcp Performance measures.
meancosts Performance measures.
Measure Construct performance measure.
measureACC Performance measures.
measureAUC Performance measures.
measureBAC Performance measures.
measureFDR Performance measures.
measureFN Performance measures.
measureFNR Performance measures.
measureFP Performance measures.
measureFPR Performance measures.
measureGMEAN Performance measures.
measureGPR Performance measures.
measureMAE Performance measures.
measureMCC Performance measures.
measureMEDAE Performance measures.
measureMEDSE Performance measures.
measureMMCE Performance measures.
measureMSE Performance measures.
measureNPV Performance measures.
measurePPV Performance measures.
measureRMSE Performance measures.
measures Performance measures.
measureSAE Performance measures.
measureSSE Performance measures.
measureTN Performance measures.
measureTNR Performance measures.
measureTP Performance measures.
measureTPR Performance measures.
medae Performance measures.
medse Performance measures.
mergeSmallFactorLevels Merges small levels of factors into new level.
mmce Performance measures.
ModelMultiplexer Create model multiplexer for model selection to tune over multiple possible models.
mse Performance measures.
mtcars.task Motor Trend Car Road Tests clustering task
multiclass.auc Performance measures.

-- N --

normalizeFeatures Normalize features
npv Performance measures.

-- O --

oversample Over- or undersample binary classification task to handle class imbalancy.

-- P --

performance Measure performance of prediction.
pid.task PimaIndiansDiabetes classification task
plotFilterValues Plot filter values.
plotLearnerPrediction Visualizes a learning algorithm on a 1D or 2D data set.
plotROCRCurves Visualize binary classification predictions via ROCR ROC curves.
plotThreshVsPerf Plot threshold vs. performance(s) for 2-class classification.
plotTuneMultiCritResult Plots multi-criteria results after tuning.
plotViperCharts Visualize binary classification predictions via ViperCharts system.
ppv Performance measures.
predict.WrappedModel Predict new data.
Prediction Prediction object.
predictLearner Predict new data with an R learner.

-- R --

RegrTask Create a classification, regression, survival, cluster, or cost-sensitive classification task.
reimpute Re-impute a data set
removeConstantFeatures Remove constant features from a data set.
removeHyperPars Remove hyperparameters settings of a learner.
removeProperties Set, add, remove or query properties of learners
repcv Fit models according to a resampling strategy.
resample Fit models according to a resampling strategy.
ResampleDesc Create a description object for a resampling strategy.
ResampleInstance Instantiates a resampling strategy object.
ResamplePrediction Prediction from resampling.
ResampleResult ResampleResult object.
RLearner Internal construction / wrapping of learner object.
RLearnerClassif Internal construction / wrapping of learner object.
RLearnerRegr Internal construction / wrapping of learner object.
RLearnerSurv Internal construction / wrapping of learner object.
rmse Performance measures.

-- S --

sae Performance measures.
selectFeatures Feature selection by wrapper approach.
setAggregation Set aggregation function of measure.
setHyperPars Set the hyperparameters of a learner object.
setHyperPars2 Only exported for internal use.
setId Set the id of a learner object.
setPredictThreshold Set the probability threshold the learner should use.
setPredictType Set the type of predictions the learner should return.
setProperties Set, add, remove or query properties of learners
setThreshold Set threshold of prediction object.
showHyperPars Display all possible hyperparameter settings for a learner that mlr knows.
silhouette Performance measures.
smote Synthetic Minority Oversampling Technique to handle class imbalancy in binary classification.
sonar.task Sonar classification task
sse Performance measures.
subsample Fit models according to a resampling strategy.
subsetTask Subset data in task.
summarizeColumns Summarize columns of data.frame or task.
summarizeLevels Summarizes factors of a data.frame by tabling them.
SurvTask Create a classification, regression, survival, cluster, or cost-sensitive classification task.

-- T --

Task Create a classification, regression, survival, cluster, or cost-sensitive classification task.
TaskDesc Description object for task.
test.join Aggregation methods.
test.max Aggregation methods.
test.mean Aggregation methods.
test.median Aggregation methods.
test.min Aggregation methods.
test.range Aggregation methods.
test.sd Aggregation methods.
test.sqrt.of.mean Aggregation methods.
test.sum Aggregation methods.
testgroup.mean Aggregation methods.
timeboth Performance measures.
timepredict Performance measures.
timetrain Performance measures.
tn Performance measures.
tnr Performance measures.
tp Performance measures.
tpr Performance measures.
train Train a learning algorithm.
train.max Aggregation methods.
train.mean Aggregation methods.
train.median Aggregation methods.
train.min Aggregation methods.
train.range Aggregation methods.
train.sd Aggregation methods.
train.sqrt.of.mean Aggregation methods.
train.sum Aggregation methods.
trainLearner Train an R learner.
TuneControl Create control structures for tuning.
TuneControlCMAES Create control structures for tuning.
TuneControlGenSA Create control structures for tuning.
TuneControlGrid Create control structures for tuning.
TuneControlIrace Create control structures for tuning.
TuneControlRandom Create control structures for tuning.
TuneMultiCritControl Create control structures for multi-criteria tuning.
TuneMultiCritControlGrid Create control structures for multi-criteria tuning.
TuneMultiCritControlNSGA2 Create control structures for multi-criteria tuning.
TuneMultiCritControlRandom Create control structures for multi-criteria tuning.
TuneMultiCritResult Result of multi-criteria tuning.
tuneParams Hyperparameter tuning.
tuneParamsMultiCrit Hyperparameter tuning for multiple measures at once.
TuneResult Result of tuning.
tuneThreshold Tune prediction threshold.

-- U --

undersample Over- or undersample binary classification task to handle class imbalancy.

-- W --

wpbc.task Wisonsin Prognostic Breast Cancer (WPBC) survival task
WrappedModel Induced model of learner.