File size: 9,430 Bytes
1ce5e18
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
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
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
# coding=utf-8
# Copyright Google AI and The HuggingFace Inc. team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""Tokenization classes for CANINE."""

from typing import Dict, List, Optional

from ...tokenization_utils import AddedToken, PreTrainedTokenizer
from ...utils import logging


logger = logging.get_logger(__name__)


PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
    "nielsr/canine-s": 2048,
}

# Unicode defines 1,114,112 total “codepoints”
UNICODE_VOCAB_SIZE = 1114112

# Below: Constants defining canonical codepoints for special, pseudo-characters.
# Copied from https://github.com/google-research/language/blob/master/language/canine/special_codepoints.py
PAD = 0
CLS = 0xE000
SEP = 0xE001
BOS = 0xE002
MASK = 0xE003
RESERVED = 0xE004

# Maps special codepoints to human-readable names.
SPECIAL_CODEPOINTS: Dict[int, str] = {
    # Special symbols are represented using codepoints values that are valid,
    # but designated as "Private Use", meaning that they will never be assigned
    # characters by the Unicode Consortium, and are thus safe for use here.
    #
    # NOTE: Do *NOT* add any sort of [UNK_CHAR] here. They are explicitly
    # excluded and should fail with a hard error.
    CLS: "[CLS]",
    SEP: "[SEP]",
    BOS: "[BOS]",
    MASK: "[MASK]",
    PAD: "[PAD]",
    RESERVED: "[RESERVED]",
}

# Maps special codepoint human-readable names to their codepoint values.
SPECIAL_CODEPOINTS_BY_NAME: Dict[str, int] = {name: codepoint for codepoint, name in SPECIAL_CODEPOINTS.items()}


class CanineTokenizer(PreTrainedTokenizer):
    r"""
    Construct a CANINE tokenizer (i.e. a character splitter). It turns text into a sequence of characters, and then
    converts each character into its Unicode code point.

    [`CanineTokenizer`] inherits from [`PreTrainedTokenizer`].

    Refer to superclass [`PreTrainedTokenizer`] for usage examples and documentation concerning parameters.

    Args:
        model_max_length (`int`, *optional*, defaults to 2048):
                The maximum sentence length the model accepts.
    """

    max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES

    def __init__(
        self,
        bos_token=chr(CLS),
        eos_token=chr(SEP),
        sep_token=chr(SEP),
        cls_token=chr(CLS),
        pad_token=chr(PAD),
        mask_token=chr(MASK),
        add_prefix_space=False,
        model_max_length=2048,
        **kwargs,
    ):
        bos_token = AddedToken(bos_token, lstrip=False, rstrip=False) if isinstance(bos_token, str) else bos_token
        eos_token = AddedToken(eos_token, lstrip=False, rstrip=False) if isinstance(eos_token, str) else eos_token
        sep_token = AddedToken(sep_token, lstrip=False, rstrip=False) if isinstance(sep_token, str) else sep_token
        cls_token = AddedToken(cls_token, lstrip=False, rstrip=False) if isinstance(cls_token, str) else cls_token
        pad_token = AddedToken(pad_token, lstrip=False, rstrip=False) if isinstance(pad_token, str) else pad_token

        # Mask token behave like a normal word, i.e. include the space before it
        mask_token = AddedToken(mask_token, lstrip=True, rstrip=False) if isinstance(mask_token, str) else mask_token

        # Creates a mapping for looking up the IDs of special symbols.
        self._special_codepoints: Dict[str, int] = {}
        for codepoint, name in SPECIAL_CODEPOINTS.items():
            self._special_codepoints[name] = codepoint

        # Creates a mapping for looking up the string forms of special symbol IDs.
        self._special_codepoint_strings: Dict[int, str] = {
            codepoint: name for name, codepoint in self._special_codepoints.items()
        }

        self._unicode_vocab_size = UNICODE_VOCAB_SIZE
        self._num_special_tokens = len(self._special_codepoints)

        super().__init__(
            bos_token=bos_token,
            eos_token=eos_token,
            sep_token=sep_token,
            cls_token=cls_token,
            pad_token=pad_token,
            mask_token=mask_token,
            add_prefix_space=add_prefix_space,
            model_max_length=model_max_length,
            **kwargs,
        )

    @property
    def vocab_size(self) -> int:
        return self._unicode_vocab_size

    def get_vocab(self):
        vocab = {chr(i): i for i in range(self.vocab_size)}
        vocab.update(self.added_tokens_encoder)
        return vocab

    def _tokenize(self, text: str) -> List[str]:
        """Tokenize a string (i.e. perform character splitting)."""
        return list(text)

    def _convert_token_to_id(self, token: str) -> int:
        """Converts a token (i.e. a Unicode character) in an id (i.e. its integer Unicode code point value)."""
        try:
            return ord(token)
        except TypeError:
            raise ValueError(f"invalid token: '{token}'")

    def _convert_id_to_token(self, index: int) -> str:
        """
        Converts a Unicode code point (integer) in a token (str). In case it's a special code point, convert to
        human-readable format.
        """
        try:
            if index in SPECIAL_CODEPOINTS:
                return SPECIAL_CODEPOINTS[index]
            return chr(index)
        except TypeError:
            raise ValueError(f"invalid id: {index}")

    def convert_tokens_to_string(self, tokens):
        return "".join(tokens)

    def build_inputs_with_special_tokens(
        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
    ) -> List[int]:
        """
        Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and
        adding special tokens. A CANINE sequence has the following format:

        - single sequence: `[CLS] X [SEP]`
        - pair of sequences: `[CLS] A [SEP] B [SEP]`

        Args:
            token_ids_0 (`List[int]`):
                List of IDs to which the special tokens will be added.
            token_ids_1 (`List[int]`, *optional*):
                Optional second list of IDs for sequence pairs.

        Returns:
            `List[int]`: List of [input IDs](../glossary#input-ids) with the appropriate special tokens.
        """
        sep = [self.sep_token_id]
        cls = [self.cls_token_id]

        result = cls + token_ids_0 + sep
        if token_ids_1 is not None:
            result += token_ids_1 + sep
        return result

    def get_special_tokens_mask(
        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None, already_has_special_tokens: bool = False
    ) -> List[int]:
        """
        Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding
        special tokens using the tokenizer `prepare_for_model` method.

        Args:
            token_ids_0 (`List[int]`):
                List of IDs.
            token_ids_1 (`List[int]`, *optional*):
                Optional second list of IDs for sequence pairs.
            already_has_special_tokens (`bool`, *optional*, defaults to `False`):
                Whether or not the token list is already formatted with special tokens for the model.

        Returns:
            `List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
        """
        if already_has_special_tokens:
            return super().get_special_tokens_mask(
                token_ids_0=token_ids_0, token_ids_1=token_ids_1, already_has_special_tokens=True
            )

        result = [1] + ([0] * len(token_ids_0)) + [1]
        if token_ids_1 is not None:
            result += ([0] * len(token_ids_1)) + [1]
        return result

    def create_token_type_ids_from_sequences(
        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
    ) -> List[int]:
        """
        Create a mask from the two sequences passed to be used in a sequence-pair classification task. A CANINE
        sequence pair mask has the following format:

        ```
        0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1
        | first sequence    | second sequence |
        ```

        If `token_ids_1` is `None`, this method only returns the first portion of the mask (0s).

        Args:
            token_ids_0 (`List[int]`):
                List of IDs.
            token_ids_1 (`List[int]`, *optional*):
                Optional second list of IDs for sequence pairs.

        Returns:
            `List[int]`: List of [token type IDs](../glossary#token-type-ids) according to the given sequence(s).
        """
        sep = [self.sep_token_id]
        cls = [self.cls_token_id]

        result = len(cls + token_ids_0 + sep) * [0]
        if token_ids_1 is not None:
            result += len(token_ids_1 + sep) * [1]
        return result

    # CanineTokenizer has no vocab file
    def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] = None):
        return ()