Package: llama 0.10.1

llama: Leveraging Learning to Automatically Manage Algorithms

Provides functionality to train and evaluate algorithm selection models for portfolios.

Authors:Lars Kotthoff [aut,cre], Bernd Bischl [aut], Barry Hurley [ctb], Talal Rahwan [ctb], Damir Pulatov [ctb]

# Install 'llama' in R:

install.packages('llama', repos = c('https://larskotthoff.r-universe.dev', 'https://cloud.r-project.org'))

Uses libs:

Datasets:

satsolvers - Example data for Leveraging Learning to Automatically Manage Algorithms

On CRAN:

Last updated 4 years agofrom:f95206aaeb. Checks:7 OK. Indexed: yes.

Help page	Topics
Leveraging Learning to Automatically Manage Algorithms	llama-package llama
Analysis functions	contributions
Bootstrapping folds	bsFolds
Classification model	classify
Classification model for pairs of algorithms	classifyPairs
Cluster model	cluster
Cross-validation folds	cvFolds
Helpers	makeRLearner.classif.constant predictLearner.classif.constant print.llama.data print.llama.model trainLearner.classif.constant
Impute censored values	imputeCensored
Read data	input
Convenience functions	predTable singleBest singleBestByCount singleBestByPar singleBestBySuccesses vbs
Misclassification penalty	misclassificationPenalties
Normalize features	normalize
Penalized average runtime score	parscores
Plot convenience functions to visualise selectors	perfScatterPlot
Regression model	regression
Regression model for pairs of algorithms	regressionPairs
Example data for Leveraging Learning to Automatically Manage Algorithms	satsolvers
Success	successes
Train / test split	trainTest
Tune the hyperparameters of the machine learning algorithm underlying a model	tuneModel