quanteda (version 4.1.0)

tokens_sample: Randomly sample documents from a tokens object

Description

Take a random sample of documents of the specified size from a corpus, with or without replacement, optionally by grouping variables or with probability weights.

Usage

tokens_sample(
  x,
  size = NULL,
  replace = FALSE,
  prob = NULL,
  by = NULL,
  env = NULL
)

Value

a tokens object (re)sampled on the documents, containing the document variables for the documents sampled.

Arguments

x

a tokens object whose documents will be sampled

size

a positive number, the number of documents to select; when used with by, the number to select from each group or a vector equal in length to the number of groups defining the samples to be chosen in each category of by. By defining a size larger than the number of documents, it is possible to oversample when replace = TRUE.

replace

if TRUE, sample with replacement

prob

a vector of probability weights for obtaining the elements of the vector being sampled. May not be applied when by is used.

by

optional grouping variable for sampling. This will be evaluated in the docvars data.frame, so that docvars may be referred to by name without quoting. This also changes previous behaviours for by. See news(Version >= "2.9", package = "quanteda") for details.

env

an environment or a list object in which x is searched. Passed to substitute for non-standard evaluation.

See Also

sample

Examples

set.seed(123)
toks <- tokens(data_corpus_inaugural[1:6])
toks
tokens_sample(toks)
tokens_sample(toks, replace = TRUE) |> docnames()
tokens_sample(toks, size = 3, replace = TRUE) |> docnames()

# sampling using by
docvars(toks)
tokens_sample(toks, size = 2, replace = TRUE, by = Party) |> docnames()