Skip to contents

Record and restore the random number generator state

Source: R/repro.R
manifest_record_seed.Rd

manifest_record_seed() stores the generator's kind and its full state in the manifest; manifest_restore_seed() puts both back. Together they let a run that used randomness be repeated exactly.

Usage

manifest_record_seed(manifest)

manifest_restore_seed(manifest)

Arguments

manifest

A manifest, as from make_manifest().

Value

manifest_record_seed() the manifest with an rng element; manifest_restore_seed() the manifest, invisibly, having set the generator.

Details

Why the state and not just a seed. set.seed(1) is reproducible only if everything before it is too: a single extra draw anywhere upstream shifts every subsequent value. Recording .Random.seed as it stood pins the actual position in the stream, and recording RNGkind() alongside it pins what that position means – R has changed its default sample() algorithm before, and a seed replayed under a different kind gives different numbers with no warning.

This is opt-in because the state is 626 integers, which is a lot of manifest for an analysis with no randomness in it.

See also

capture_environment(), capsule_falsify().

Examples

set.seed(42)
m <- manifest_record_seed(make_manifest(list(a = 1),
                                        environment = FALSE))
first <- runif(3)

# any amount of other work can happen in between
invisible(runif(1000))

manifest_restore_seed(m)
identical(runif(3), first)
#> [1] TRUE