Learn R Programming

⚠️There's a newer version (2.19.1) of this package.Take me there.

mlr (version 1.1-18)

mlr: Machine Learning in R.

Description

Interface to a large number of classification and regression techniques, including machine-readable parameter descriptions. Generic resampling, including cross-validation, bootstrapping and subsampling. Hyperparameter tuning with modern optimization techniques. Filter and wrapper methods for feature selection. Extension of basic learners with additional operations. Nested resampling.

Copy Link

Version

Install

install.packages('mlr')

Monthly Downloads

7,379

Version

1.1-18

License

BSD_3_clause + file LICENSE

Issues

Pull Requests

Stars

Forks

Repository

https://github.com/berndbischl/mlr

Maintainer

Bernd Bischl

Last Published

August 29th, 2013

Functions in mlr (1.1-18)

Set aggregation function of measure.

Result of feature selection.

Aggregation methods.

makeClassifTask

Create a classification / regression task for a given data set.

Confusion matrix.

Get a description of all possible parameter settings for a learner.

getProbabilities

Get probabilities for some classes.

Prediction object.

Returns the optimal hyperparameters and optimization path.

asROCRPrediction

Converts predictions to a format package ROCR can handle.

ResamplePrediction

Prediction from resampling.

makeTuneWrapper

Fuse learner with tuning.

getBaggingModels

Returns the list of models fitted in bagging.

Description object for task.

Measure performance of prediction.

makeBaggingWrapper

Fuse learner with the bagging technique.

An aggregation method reduces the performance values of the test (and possibly the training sets) to a single value. To see all possible, implemented aggregations look at aggregations.

makeWrappedModel

Induced model of learner.

Feature selection by wrapper approach.

Get formula of a task.

makeTuneControlCMAES

Create control structures for tuning.

getLearnerModel

Get underlying R model of learner integrated into mlr.

getTaskFeatureNames

Get feature names of task.

Find matching learning algorithms.

Set the type of predictions the learner should return.

learnerArgsToControl

Convert arguments to control structure.

Configures the behaviour of the package.

Filter features by using a numerical importance criterion. Calculates numerical importance values for all features. Thresholding of these values can be used to select useful features. Look at package FSelector for details on the filter algorithms.

makeCostMeasure

Creates a measure for non-standard misclassification costs.

Fit models according to a resampling strategy.

Subset data in task.

makeResampleInstance

Instantiates a resampling strategy object.

Train an R learner.

Result of tuning.

Get target column of task.

Set the hyperparameters of a learner object.

Hyperparameter tuning.

Only exported for internal use.

Construct performance measure.

makeFilterWrapper

Fuse learner with filter method.

Set threshold of prediction object.

Create learner object.

Get current parameter settings for a learner.

makeFixedHoldoutInstance

Generate a fixed holdout instance for resampling.

List of supported learning algorithms.

makeCustomResampledMeasure

Construct your own resampled performance measure.

getFilteredFeatures

Returns the filtered features.

Train a learning algorithm.

makeFeatSelControlExhaustive

Create control structures for feature selection.

Predict new data with an R learner.

makePreprocWrapper

Fuse learner with preprocessing.

Internal construction / wrapping of learner object.

getTaskFormulaAsString

Get formula of a task as a string.

Tune prediction threshold.

predict.WrappedModel

Predict new data.

makeResampleDesc

Create a description object for a resampling strategy.

Extract data in task. Useful in trainLearner when you add a learning machine to the package.

Performance measures.