[source]

N-gram#

N-grams are sequences of n-words of a given string. The N-gram tokenizer outputs tokens of contiguous words ranging from min to max number of words per token.

Parameters#

# Param Default Type Description
1 min 2 int The minimum number of contiguous words to a token.
2 max 2 int The maximum number of contiguous words to a token.

Example#

use Rubix\ML\Extractors\Tokenizers\NGram;

$tokenizer = new NGram(1, 3);