mirror of
https://github.com/discourse/discourse-ai.git
synced 2025-03-09 02:40:50 +00:00
15 lines
294 B
Ruby
15 lines
294 B
Ruby
# frozen_string_literal: true
|
|
|
|
module DiscourseAi
|
|
class Tokenizer
|
|
def self.tokenizer
|
|
@@tokenizer ||=
|
|
Tokenizers.from_file("./plugins/discourse-ai/tokenizers/bert-base-uncased.json")
|
|
end
|
|
|
|
def self.size(text)
|
|
tokenizer.encode(text).tokens.size
|
|
end
|
|
end
|
|
end
|