mlr (version 2.8)

generateFilterValuesData: Calculates feature filter values.

Description

Calculates numerical filter values for features. For a list of features, use listFilterMethods.

Usage

generateFilterValuesData(task, method = "rf.importance", nselect = getTaskNFeats(task), ...)

Arguments

task

[Task] The task.

method

[character] Filter method(s), see above. Default is “rf.importance”.

nselect

[integer(1)] Number of scores to request. Scores are getting calculated for all features per default.

...

[any] Passed down to selected method.

Value

[FilterValues]. A list containing:

task.desc

[TaskDesc] Task description.

data

[data.frame] with columns:

• name Name of feature.
• type Feature column type.
• A column for each method with the feature importance values.

See Also

Other filter: filterFeatures, getFilterValues, getFilteredFeatures, makeFilterWrapper, plotFilterValuesGGVIS, plotFilterValues

Other generate_plot_data: generateCalibrationData, generateCritDifferencesData, generateLearningCurveData, generatePartialPredictionData, generateThreshVsPerfData, getFilterValues