Learn R Programming
⚠️There's a newer version (0.1.3) of this package.Take me there.

aslib (version 0.1.1)

Interface to the Algorithm Selection Benchmark Library

Description

Provides an interface to the algorithm selection benchmark library at and the 'LLAMA' package () for building algorithm selection models; see Bischl et al. (2016) .

Copy Link

Version

Install

install.packages('aslib')

Monthly Downloads

293

Version

0.1.1

License

GPL-3

Issues

4

Pull Requests

0

Stars

8

Forks

6

Maintainer

Lars Kotthoff

Last Published

May 24th, 2020

Functions in aslib (0.1.1)

summarizeAlgoPerf

Creates summary data.frame for algorithm performance values across all instances.
convertAlgoPerfToWideFormat

Converts algo.runs object of a scenario to wide format.
runLlamaModels

Creates a registry which can be used for running several Llama models on a cluster.
convertToLlama

Convert an ASScenario scenario object to a llama data object.
writeASScenario

Writes an algorithm selection scenario to a directory.
findDominatedAlgos

Creates a table that shows the dominance of one algorithm over another one.
getFeatureStepNames

Returns feature step names of scenario.
fixFeckingPresolve

Bakes presolving stuff into a LLAMA data frame.
getFeatureNames

Returns feature names of scenario.
getInstanceNames

Returns instance names of scenario.
plotAlgoPerf

EDA plots for performance values of algorithms across all instances.
getNumberOfCVFolds

Returns number of CV folds.
ASScenarioDesc

S3 class for ASScenarioDesc.
checkDuplicatedInstances

Checks the feature data set for duplicated instances.
getCostsAndPresolvedStatus

Return wether an instance was presolved and which step did it.
getSummedFeatureCosts

Returns feature costs of scenario, summed over all instances.
getAlgorithmNames

Returns algorithm names of scenario.
convertToLlamaCVFolds

Convert an ASScenario scenario object to a llama data object with cross-validation folds.
summarizeFeatureValues

Creates summary data.frame for feature values across all instances.
createCVSplits

Create cross-validation splits for a scenario.
imputeAlgoPerf

Imputes algorithm performance for runs which have NA performance values.
summarizeAlgoRunstatus

Creates summary data.frame for algorithm runstatus across all instances.
getNumberOfCVReps

Returns number of CV repetitions.
getProvidedFeatures

Return features that are useable for a given set of feature steps.
summarizeFeatureSteps

Creates a data.frame that summarizes the feature steps.
plotAlgoCorMatrix

Plots the correlation matrix of the algorithms.
getDefaultFeatureStepNames

Returns the default feature step names of scenario.
getCosealASScenario

Retrieves a scenario from the Coseal Github repository and parses into an S3 object.
parseASScenario

Parses the data files of an algorithm selection scenario into an S3 object.