makeOMLEstimationProcedure: Construct OMLEstimationProcedure.

Description Usage Arguments Value

View source: R/OMLEstimationProcedure_Class.R

Description

Constructs an estimation procedure.

Usage

1
2
makeOMLEstimationProcedure(type, data.splits.url = NA_character_,
  data.splits = NULL, parameters = NULL)

Arguments

type

[character(1)]
The type of procedure used for performance estimation: either cross-validation, holdout, bootstrap or sampling.

data.splits.url

[character(1)]
The url from which a file with train-test splits can be downloaded.

data.splits

[data.frame]
The train-test splits. Default is NULL, which means data splits have not yet been downloaded.

parameters

[list]
Any parameters for the estimation procedure, e.g. the number of repeats and folds in cross-validation.

Value

[OMLEstimationProcedure]


openml/openml-r documentation built on May 24, 2019, 2:32 p.m.