generate_unigrams: Generate unigrams

In biogram: N-Gram Analysis of Biological Sequences

Description

Generates an alphabet of unigrams based on given list of properties.

Usage

generate_unigrams(unigram_list, unigram_names = NULL, prop_names = NULL)

Arguments

unigram_list	a list of unigrams' parameters. See Details.
unigram_names	names of unigrams. If not NULL, will overwrite any existing unigram names.
prop_names	names of properties. If not NULL, will overwrite any existing names.

Details

Unigram parameters are represented as a list of intervals, where each interval corresponds to a different property. The function generate unigrams randomly choosing values of properties from given intervals using uniform distribution. All lists of ranges should have the same length, which equils to describing each unigram using the same properties.

Examples

props1 <- list(P1 = c(0, 0.5), 
               P2 = c(0.2, 0.4),
               P3 = c(0.5, 1),
               P4 = c(0, 0))

props2 <- list(P1 = c(0.5, 1), 
               P2 = c(0.4, 1),
               P3 = c(0, 0.5),
               P4 = c(1, 1))


alph <- generate_unigrams(c(replicate(8, props1, simplify = FALSE),
                          replicate(12, props2, simplify = FALSE)),
                          unigram_names = letters[1L:20])
                          

biogram documentation built on March 31, 2020, 5:14 p.m.