Class: Transformers::Bert::BertTokenizer

Inherits:
PreTrainedTokenizer show all
Defined in:
lib/transformers/models/bert/tokenization_bert.rb

Defined Under Namespace

Classes: BasicTokenizer, WordpieceTokenizer

Constant Summary

Constants included from SpecialTokensMixin

SpecialTokensMixin::SPECIAL_TOKENS_ATTRIBUTES

Instance Attribute Summary collapse

Attributes inherited from PreTrainedTokenizerBase

#init_kwargs, #model_max_length

Instance Method Summary collapse

Methods inherited from PreTrainedTokenizer

#_convert_token_to_id_with_added_voc, #_encode_plus, #convert_tokens_to_ids, #is_fast, #tokenize, #vocab_size

Methods inherited from PreTrainedTokenizerBase

#_eventual_warn_about_too_long_sequence, _from_pretrained, #call, from_pretrained

Methods included from ClassAttribute

#class_attribute

Methods included from SpecialTokensMixin

#bos_token_id, #cls_token_id, #eos_token_id, #pad_token_id, #sep_token_id, #special_tokens_map, #unk_token_id

Constructor Details

#initialize(vocab_file:, do_lower_case: true, do_basic_tokenize: true, never_split: nil, unk_token: "[UNK]", sep_token: "[SEP]", pad_token: "[PAD]", cls_token: "[CLS]", mask_token: "[MASK]", tokenize_chinese_chars: true, strip_accents: nil, **kwargs) ⇒ BertTokenizer

Returns a new instance of BertTokenizer.



49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
 
# File 'lib/transformers/models/bert/tokenization_bert.rb', line 49

def initialize(
  vocab_file:,
  do_lower_case: true,
  do_basic_tokenize: true,
  never_split: nil,
  unk_token: "[UNK]",
  sep_token: "[SEP]",
  pad_token: "[PAD]",
  cls_token: "[CLS]",
  mask_token: "[MASK]",
  tokenize_chinese_chars: true,
  strip_accents: nil,
  **kwargs
)
  if !File.exist?(vocab_file)
    raise ArgumentError,
      "Can't find a vocabulary file at path '#{vocab_file}'. To load the vocabulary from a Google pretrained" +
      " model use `tokenizer = BertTokenizer.from_pretrained(PRETRAINED_MODEL_NAME)`"
  end
  @vocab = load_vocab(vocab_file)
  @ids_to_tokens = @vocab.invert
  @do_basic_tokenize = do_basic_tokenize
  if do_basic_tokenize
    @basic_tokenizer =
      BasicTokenizer.new(
        do_lower_case: do_lower_case,
        never_split: never_split,
        tokenize_chinese_chars: tokenize_chinese_chars,
        strip_accents: strip_accents
      )
  end

  @wordpiece_tokenizer = WordpieceTokenizer.new(vocab: @vocab, unk_token: unk_token.to_s)

  super(
    do_lower_case: do_lower_case,
    do_basic_tokenize: do_basic_tokenize,
    never_split: never_split,
    unk_token: unk_token,
    sep_token: sep_token,
    pad_token: pad_token,
    cls_token: cls_token,
    mask_token: mask_token,
    tokenize_chinese_chars: tokenize_chinese_chars,
    strip_accents: strip_accents,
    **kwargs
  )
end

Instance Attribute Details

#basic_tokenizer ⇒ Object (readonly)

Returns the value of attribute basic_tokenizer.



47
48
49
 
# File 'lib/transformers/models/bert/tokenization_bert.rb', line 47

def basic_tokenizer
  @basic_tokenizer
end

#vocab ⇒ Object (readonly)

Returns the value of attribute vocab.



47
48
49
 
# File 'lib/transformers/models/bert/tokenization_bert.rb', line 47

def vocab
  @vocab
end

Instance Method Details

#_convert_token_to_id(token) ⇒ Object 



98
99
100
 
# File 'lib/transformers/models/bert/tokenization_bert.rb', line 98

def _convert_token_to_id(token)
  @vocab.fetch(token, @vocab.fetch(@unk_token))
end