Description object for task.
Result of multi-criteria tuning.
Get predictions from resample results.
Calculates feature filter values.
generateLearningCurveData
Generates a learning curve.
Crossover.
Configures the behavior of the package.
Boston Housing regression task.
Perform overall Friedman test for a BenchmarkResult.
Extract the feature selection results from a benchmark result.
Estimate the residual variance.
Convert BenchmarkResult to a rank-matrix.
European Union Agricultural Workforces clustering task.
Confusion matrix.
Get default measure.
Deprecated, use getPredictionProbabilities
instead.
Return error message of FailureModel.
Get the name(s) of the target column(s).
makeCostSensWeightedPairsWrapper
Wraps a classifier for cost-sensitive learning to produce a weighted pairs model.
getPredictionProbabilities
Get probabilities for some classes.
Aggregation methods.
Get the class levels for classification and multilabel tasks.
Extract costs in task.
Iris classification task.
List of supported learning algorithms.
Get the class weight parameter of a learner.
getStackedBaseLearnerPredictions
Returns the predictions for each base learner.
Get the tuned hyperparameter settings from a nested tuning.
Get the id of the task.
generateCritDifferencesData
Generate data for critical-differences plot.
Create a feature filter.
Instantiates a resampling strategy object.
Fuse learner with preprocessing.
Extract the feature selection results from a benchmark result.
Fuse learner with the bagging technique.
Motor Trend Car Road Tests clustering task.
Is the model a FailureModel?
Create a description object for a resampling strategy.
Create model multiplexer for model selection to tune over multiple possible models.
Get a summarizing task description.
mergeBenchmarkResultLearner
Merge different learners of BenchmarkResult objects.
Plots a ROC curve using ggplot2.
Construct performance measure.
Hyperparameter tuning for multiple measures at once.
Predict new data with an R learner.
Wisonsin Prognostic Breast Cancer (WPBC) survival task.
Set the probability threshold the learner should use.
Get response / truth from prediction object.
Yeast multilabel classification task.
Sonar classification task.
Create a classification, regression, survival, cluster, cost-sensitive classification or
multilabel task.
Benchmark experiment for multiple learners and tasks.
Return task ids used in benchmark.
Result of feature selection.
Create a bar chart for ranks in a BenchmarkResult.
Plot a benchmark summary.
Returns the selected feature set and optimization path after training.
Extract all models from benchmark result.
Returns the optimal hyperparameters and optimization path after training.
Perform a posthoc Friedman-Nemenyi test.
Filter features by thresholding filter values.
Converts predictions to a format package ROCR can handle.
Convert arguments to control structure.
ResampleResult object.
Get the type of the task.
Generate threshold vs. performance(s) for 2-class classification.
Aggregation object.
Set aggregation function of measure.
Calculates feature filter values.
Create box or violin plots for a BenchmarkResult.
Return learner ids used in benchmark.
getHomogeneousEnsembleModels
Deprecated, use getLearnerModel
instead.
Drop some features of task.
getNestedTuneResultsOptPathDf
Get the opt.path
s from each tuning step from the outer resampling.
Synthetic Minority Oversampling Technique to handle class imbalancy in binary classification.
Fuse learner with simple ove/underrsampling for imbalancy correction in binary classification.
Over- or undersample binary classification task to handle class imbalancy.
NCCTG Lung Cancer survival task.
Create control structures for tuning.
Prediction object.
Plots multi-criteria results after tuning using ggplot2.
Specify your own aggregation of measures.
Plot critical differences for a selected measure.
PimaIndiansDiabetes classification task.
Returns a list of mlr's options.
Get feature names of task.
Summarizes factors of a data.frame by tabling them.
Predict new data.
Get target data of task.
Extract the test performance values from a benchmark result.
Plot threshold vs. performance(s) for 2-class classification using ggplot2.
estimateRelativeOverfitting
Estimate relative overfitting.
Get formula of a task.
Return measures used in benchmark.
makeWeightedClassesWrapper
Wraps a classifier for weighted fitting where each class receives a weight.
Convert a machine learning benchmark / demo object from package mlbench to a task.
Wisconsin Breast Cancer classification task.
Set threshold of prediction object.
Query properties of learners.
Fuse learner with the bagging technique and oversampling for imbalancy correction.
Return learner short.names used in benchmark.
Prediction from resampling.
Create a stacked learner object.
Fuse learner with simple downsampling (subsampling).
Fuse learner with a feature filter method.
getMultilabelBinaryPerformances
Retrieve binary classification measures for multilabel classification predictions.
makeMultilabelBinaryRelevanceWrapper
Use binary relevance method to create a multilabel learner.
Find matching learning algorithms.
Convert large/infinite numeric values in a data.frame or task.
Deprecated, use hasLearnerProperties
instead.
Extract data in task.
plotTuneMultiCritResultGGVIS
Plots multi-criteria results after tuning using ggvis.
Set the type of predictions the learner should return.
Tune prediction threshold.
Get current parameter settings for a learner.
Get underlying R model of learner integrated into mlr.
Create a custom imputation method.
Generate dummy variables for factor features.
Re-impute a data set
Downsample (subsample) a task or a data.frame.
Fuse learner with multiclass method.
Creates a measure for non-standard misclassification costs.
Create control structures for feature selection.
Plot filter values using ggplot2.
Fit models according to a resampling strategy.
Find matching measures.
Merge different tasks of BenchmarkResult objects.
Plot learning curve data using ggplot2.
Remove hyperparameters settings of a learner.
Join some class existing levels to new, larger class levels for classification problems.
Wraps a regression learner for use in cost-sensitive learning.
makeCustomResampledMeasure
Construct your own resampled performance measure.
Train a learning algorithm.
Summarize columns of data.frame or task.
Get a description of all possible parameter settings for a learner.
Hyperparameter tuning.
Performance measures.
Plot filter values using ggvis.
Create control structures for multi-criteria tuning.
Get number of features in task.
Iris cost-sensitive classification task.
Set the id of a learner object.
Result of tuning.
Get number of observations in task.
Remove constant features from a data set.
Plot a partial prediction with ggplot2.
Subset data in task.
Extract the aggregated performance values from a benchmark result.
Set the hyperparameters of a learner object.
Induced model of learner.
Train an R learner.
Extract the tuning results from a benchmark result.
Internal construction / wrapping of learner object.
Failure model.
Fuse learner with feature selection.
Generate classifier calibration data.
Normalize features.
generatePartialPredictionData
Generate partial predictions.
Fuse learner with tuning.
BenchmarkResult object.
Generate a fixed holdout instance for resampling.
Show and visualize the steps of feature selection.
Get tuning parameters from a learner of the caret R-package.
Built-in imputation methods.
Plot learning curve data using ggvis.
Return learners used in benchmark.
Merges small levels of factors into new level.
Only exported for internal use.
Return measures IDs used in benchmark.
makeCostSensClassifWrapper
Wraps a classification learner for use in cost-sensitive learning.
Feature selection by wrapper approach.
Extract the predictions from a benchmark result.
Plot calibration data using ggplot2.
Impute and re-impute data
regression using randomForest.
plotPartialPredictionGGVIS
Plot a partial prediction using ggvis.
Plot threshold vs. performance(s) for 2-class classification using ggvis.
Visualize binary classification predictions via ViperCharts system.
makeModelMultiplexerParamSet
Creates a parameter set for model multiplexer tuning.
List filter methods.
Returns the filtered features.
Visualizes a learning algorithm on a 1D or 2D data set.
Fuse learner with an imputation method.
Create learner object.
Measure performance of prediction.
Fuse learner with SMOTE oversampling for imbalancy correction in binary classification.
Fuse learner with preprocessing.