Learn R Programming

BatchExperiments (version 1.4.3)

findExperiments: Find ids of experiments that match a query.

Description

Find job ids by querying problem/algorithm ids, problem/algorithm parameters or replication number.

Usage

findExperiments(
  reg,
  ids,
  prob.pattern,
  prob.pars,
  algo.pattern,
  algo.pars,
  repls,
  match.substring = TRUE,
  regexp = FALSE
)

Arguments

reg

[ExperimentRegistry] Registry.

ids

[integer] Ids of selected experiments to restrict to. Default is all experiments.

prob.pattern

[character(1)] If not missing, all problem ids that match this string are selected.

prob.pars

[R expression] If not missing, all problems whose parameters match the given expression are selected.

algo.pattern

[character(1)] If not missing, all algorithm ids that match this string are selected.

algo.pars

[R expression] If not missing, all algorithms whose parameters match the given expression are selected.

repls

[integer] If not missing, restrict to jobs with given replication numbers.

match.substring

[logical(1)] Is a match in prob.pattern and algo.pattern if the id contains the pattern as substring or must the id exactly match? Default is TRUE.

regexp

[logical(1)] Are prob.pattern and algo.pattern regular expressions? Note that this is significantly slower than substring matching. If set to TRUE the argument match.substring has no effect. Default is FALSE.

Value

[integer]. Ids for experiments which match the query.

Examples

Run this code
# NOT RUN {
reg = makeExperimentRegistry(id = "example1", file.dir = tempfile())
p1 = addProblem(reg, "one", 1)
p2 = addProblem(reg, "two", 2)
a = addAlgorithm(reg, "A", fun = function(static, n) static + n)
addExperiments(reg, algo.design = makeDesign(a, exhaustive = list(n = 1:4)))
findExperiments(reg, prob.pattern = "one")
findExperiments(reg, prob.pattern = "o")
findExperiments(reg, algo.pars = (n > 2))
# }

Run the code above in your browser using DataLab