Package: BatchExperiments 1.4.3

Michel Lang

BatchExperiments: Statistical Experiments on Batch Computing Clusters

Extends the BatchJobs package to run statistical experiments on batch computing clusters. For further details see the project web page.

Authors:Bernd Bischl <[email protected]>, Michel Lang <[email protected]>, Olaf Mersmann <[email protected]>

# Install 'BatchExperiments' in R:

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

On CRAN:

Last updated 3 years agofrom:93ca900b7e. Checks:3 OK, 6 NOTE. Indexed: yes.

Help page	Topics
Add an algorithm to registry.	addAlgorithm Algorithm
Add experiemts to the registry.	addExperiments Experiment
Add a problem to registry.	addProblem Problem
The BatchExperiments package	BatchExperiments
ExperimentJob	ExperimentJob
Find ids of experiments that match a query.	findExperiments
Generate dynamic part of problem.	generateProblemInstance Instance
Get algorithm from registry by id.	getAlgorithm
Get ids of algorithms in registry.	getAlgorithmIds
Get all parts required to run a single job.	getExperimentParts
Group experiments.	getIndex
Get jobs (here: experiments) from registry by id.	getJobs.ExperimentRegistry
Get problem from registry by id.	getProblem
Get ids of problems in registry.	getProblemIds
Get variable groups of reduced results.	getResultVars
Create parameter designs for problems and algorithms.	Design makeDesign
Construct a registry object for experiments.	ExperimentRegistry makeExperimentRegistry
Reduce results into a data.frame with all relevant information.	ReducedResultsExperiments reduceResultsExperiments
Reduce very many results in parallel.	reduceResultsExperimentsParallel
Remove algorithm from registry.	removeAlgorithm
Remove jobs from registry.	removeExperiments
Remove problem from registry.	removeProblem
Summarize selected experiments.	summarizeExperiments