File size: 17,743 Bytes
d11c6aa
 
 
 
 
 
 
cc96a22
 
 
 
d11c6aa
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
8492687
d11c6aa
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
c4575e7
d11c6aa
e1494f2
d11c6aa
 
 
 
 
 
 
 
 
 
 
cc96a22
d11c6aa
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cc96a22
d11c6aa
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1b54948
d11c6aa
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cc96a22
 
d11c6aa
 
 
 
 
 
 
 
cc96a22
373fd6b
cc96a22
 
 
d11c6aa
cc96a22
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
373fd6b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cc96a22
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
"""Tokenization classes for ChatGLM."""
from typing import List, Optional, Union
import os

from transformers.tokenization_utils import PreTrainedTokenizer
from icetk.text_tokenizer import TextTokenizer
import icetk.sentencepiece_model_pb2 as sp_model
from transformers.utils import logging, PaddingStrategy
from transformers.tokenization_utils_base import EncodedInput, BatchEncoding
from typing import Dict
import numpy as np

logger = logging.get_logger(__name__)

PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
    "THUDM/chatglm-6b": 2048,
}


class SPTokenizer:
    def __init__(
        self,
        vocab_file,
        max_blank_length=80,
        byte_fallback=True,
    ):
        assert vocab_file is not None
        self.vocab_file = vocab_file
        self.special_tokens = ["[MASK]", "[gMASK]", "[sMASK]", "<unused_0>", "<sop>", "<eop>", "<ENC>", "<dBLOCK>"]
        self.max_blank_length = max_blank_length
        self.byte_fallback = byte_fallback
        self.text_tokenizer = self._build_text_tokenizer(encode_special_tokens=False)
        self.special_text_tokenizer = self._build_text_tokenizer(encode_special_tokens=True)

    @staticmethod
    def _configure_tokenizer(
        text_tokenizer: TextTokenizer,
        special_tokens: List[str],
        max_blank_length: int,
        byte_fallback: bool,
        encode_special_tokens=False,
    ):
        # special token
        special_token_type = 4 if encode_special_tokens else 3  # 3 - CONTROL, 4 - USER_DEFINE
        for token in special_tokens:
            text_tokenizer.proto.pieces.append(
                sp_model.ModelProto.SentencePiece(piece=token, score=0.0, type=special_token_type)
            )
        # whitespaces
        for token in [SPTokenizer.get_tab_token()] + [
            SPTokenizer.get_blank_token(i) for i in range(2, max_blank_length + 1)
        ]:
            text_tokenizer.proto.pieces.append(sp_model.ModelProto.SentencePiece(piece=token, score=0.0, type=4))
        # byte fallback
        if byte_fallback:
            text_tokenizer.proto.trainer_spec.byte_fallback = True
            for i in range(256):
                text_tokenizer.proto.pieces.append(
                    sp_model.ModelProto.SentencePiece(piece="<0x{:02X}>".format(i), score=0.0, type=6)
                )
        text_tokenizer.refresh()

    def _build_text_tokenizer(self, encode_special_tokens=False):
        tokenizer = TextTokenizer(self.vocab_file)
        self._configure_tokenizer(
            tokenizer, self.special_tokens, self.max_blank_length, self.byte_fallback, encode_special_tokens
        )
        return tokenizer

    def _get_text_tokenizer(self, encode_special_tokens=False):
        if encode_special_tokens:
            return self.special_text_tokenizer
        else:
            return self.text_tokenizer

    @staticmethod
    def get_blank_token(length: int):
        assert length >= 2
        return f"<|blank_{length}|>"

    @staticmethod
    def get_tab_token():
        return f"<|tab|>"

    @property
    def num_image_tokens(self):
        return 20000

    @property
    def num_text_tokens(self):
        return self.text_tokenizer.num_tokens

    @property
    def num_tokens(self):
        return self.num_image_tokens + self.num_text_tokens

    @staticmethod
    def _encode_whitespaces(text: str, max_len: int = 80):
        text = text.replace("\t", SPTokenizer.get_tab_token())
        for i in range(max_len, 1, -1):
            text = text.replace(" " * i, SPTokenizer.get_blank_token(i))
        return text

    def _preprocess(self, text: str, linebreak=True, whitespaces=True):
        if linebreak:
            text = text.replace("\n", "<n>")
        if whitespaces:
            text = self._encode_whitespaces(text, max_len=self.max_blank_length)
        return text

    def encode(
        self, text: str, linebreak=True, whitespaces=True, special_tokens=False, add_dummy_prefix=True
    ) -> List[int]:
        """
        @param text: Text to encode.
        @param linebreak: Whether to encode newline (\n) in text.
        @param whitespaces: Whether to encode multiple whitespaces or tab in text, useful for source code encoding.
        @param special_tokens: Whether to encode special token ([MASK], [gMASK], etc.) in text.
        @param add_dummy_prefix: Whether to add dummy blank space in the beginning.
        """
        text = self._preprocess(text, linebreak, whitespaces)
        if not add_dummy_prefix:
            text = "<n>" + text
        tmp = self._get_text_tokenizer(encode_special_tokens=special_tokens).encode(text)
        tokens = [x + self.num_image_tokens for x in tmp]
        return tokens if add_dummy_prefix else tokens[2:]

    def decode(self, text_ids: List[int], special_tokens=False) -> str:
        ids = [int(_id) - self.num_image_tokens for _id in text_ids]
        ids = [_id for _id in ids if _id >= 0]
        text = self._get_text_tokenizer(encode_special_tokens=special_tokens).decode(ids)
        text = text.replace("<n>", "\n")
        text = text.replace(SPTokenizer.get_tab_token(), "\t")
        for i in range(2, self.max_blank_length + 1):
            text = text.replace(self.get_blank_token(i), " " * i)
        return text

    def tokenize(
        self, text: str, linebreak=True, whitespaces=True, special_tokens=False, add_dummy_prefix=True
    ) -> List[str]:
        """
        @param text: Text to encode.
        @param linebreak: Whether to encode newline (\n) in text.
        @param whitespaces: Whether to encode multiple whitespaces or tab in text, useful for source code encoding.
        @param special_tokens: Whether to encode special token ([MASK], [gMASK], etc.) in text.
        @param add_dummy_prefix: Whether to add dummy blank space in the beginning.
        """
        text = self._preprocess(text, linebreak, whitespaces)
        if not add_dummy_prefix:
            text = "<n>" + text
        tokens = self._get_text_tokenizer(encode_special_tokens=special_tokens).tokenize(text)
        return tokens if add_dummy_prefix else tokens[2:]

    def __getitem__(self, x: Union[int, str]):
        if isinstance(x, int):
            if x < self.num_image_tokens:
                return "<image_{}>".format(x)
            else:
                return self.text_tokenizer.convert_id_to_token(x - self.num_image_tokens)
        elif isinstance(x, str):
            if x.startswith("<image_") and x.endswith(">") and x[7:-1].isdigit():
                return int(x[7:-1])
            else:
                return self.text_tokenizer.convert_token_to_id(x) + self.num_image_tokens
        else:
            raise ValueError("The key should be str or int.")


class ChatGLMTokenizer(PreTrainedTokenizer):
    """
    Construct a ChatGLM tokenizer. Based on byte-level Byte-Pair-Encoding.

    Args:
        vocab_file (`str`):
            Path to the vocabulary file.
    """

    vocab_files_names = {"vocab_file": "ice_text.model"}
    max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
    model_input_names = ["input_ids", "attention_mask", "position_ids"]

    def __init__(
            self,
            vocab_file,
            do_lower_case=False,
            remove_space=False,
            bos_token='sop',
            eos_token='eos',
            eop_token='eop',
            mask_token='[MASK]',
            gmask_token='[gMASK]',
            padding_side="left",
            **kwargs
    ) -> None:
        super().__init__(
            do_lower_case=do_lower_case,
            remove_space=remove_space,
            padding_side=padding_side,
            **kwargs
        )

        self.do_lower_case = do_lower_case
        self.remove_space = remove_space
        self.vocab_file = vocab_file

        self.bos_token = bos_token
        self.eos_token = eos_token
        self.eop_token = eop_token
        self.mask_token = mask_token
        self.gmask_token = gmask_token

        self.sp_tokenizer = SPTokenizer(vocab_file)

        """ Initialisation """

    @property
    def eop_token_id(self) -> Optional[int]:
        """
        `Optional[int]`: Id of the end of sentence token in the vocabulary. Returns `None` if the token has not been
        set.
        """
        if self.eop_token is None:
            return None
        return self.convert_tokens_to_ids(self.eop_token)

    @property
    def vocab_size(self):
        """ Returns vocab size """
        return self.sp_tokenizer.num_tokens

    def get_vocab(self):
        """ Returns vocab as a dict """
        vocab = {self._convert_id_to_token(i): i for i in range(self.vocab_size)}
        vocab.update(self.added_tokens_encoder)
        return vocab

    def preprocess_text(self, inputs):
        if self.remove_space:
            outputs = " ".join(inputs.strip().split())
        else:
            outputs = inputs

        if self.do_lower_case:
            outputs = outputs.lower()

        return outputs

    def _tokenize(self, text, **kwargs):
        """ Returns a tokenized string. """
        text = self.preprocess_text(text)

        seq = self.sp_tokenizer.tokenize(text)

        return seq

    def decode(
            self,
            token_ids: Union[List[int], List[List[int]]],
            skip_special_tokens: bool = False,
            clean_up_tokenization_spaces: bool = True,
            spaces_between_special_tokens: bool = True,
            **kwargs
    ) -> str:
        if isinstance(token_ids[0], list):
            tokens = []
            for single_token_ids in token_ids:
                if self.pad_token_id in single_token_ids:  # remove pad
                    single_token_ids = list(filter((self.pad_token_id).__ne__, single_token_ids))
                tokens.append(self.sp_tokenizer.decode(single_token_ids))
            return (tokens)
        else:
            if self.pad_token_id in token_ids:  # remove pad
                token_ids = list(filter((self.pad_token_id).__ne__, token_ids))
            return self.sp_tokenizer.decode(token_ids)

    def _convert_token_to_id(self, token):
        """ Converts a token (str) in an id using the vocab. """
        return self.sp_tokenizer[token]

    def _convert_id_to_token(self, index):
        """Converts an index (integer) in a token (str) using the vocab."""
        return self.sp_tokenizer[index]

    def save_vocabulary(self, save_directory, filename_prefix=None):
        """
        Save the vocabulary and special tokens file to a directory.

        Args:
            save_directory (`str`):
                The directory in which to save the vocabulary.
            filename_prefix (`str`, *optional*):
                An optional prefix to add to the named of the saved files.

        Returns:
            `Tuple(str)`: Paths to the files saved.
        """
        if os.path.isdir(save_directory):
            vocab_file = os.path.join(
                save_directory, self.vocab_files_names["vocab_file"]
            )
        else:
            vocab_file = save_directory

        with open(self.vocab_file, 'rb') as fin:
            proto_str = fin.read()

        with open(vocab_file, "wb") as writer:
            writer.write(proto_str)

        return (vocab_file,)

    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 BERT 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.
        """
        mask_ids = self.sp_tokenizer[self.mask_token]
        gmask_ids = self.sp_tokenizer[self.gmask_token]
        eop_id = self.sp_tokenizer[self.eop_token]
        if mask_ids not in token_ids_0 and gmask_ids not in token_ids_0:
            token_ids_0 += [gmask_ids]

        if token_ids_0[-1] != mask_ids and token_ids_0[-1] != gmask_ids:
            token_ids_0 += [self.sp_tokenizer[self.eos_token]]

        token_ids_0 += [self.sp_tokenizer[self.bos_token]]

        if token_ids_1 is not None:
            if not token_ids_1 or token_ids_1[-1] != eop_id:
                token_ids_1 += [eop_id]
            token_ids_0 += token_ids_1

        return token_ids_0

    def _pad(
        self,
        encoded_inputs: Union[Dict[str, EncodedInput], BatchEncoding],
        max_length: Optional[int] = None,
        padding_strategy: PaddingStrategy = PaddingStrategy.DO_NOT_PAD,
        pad_to_multiple_of: Optional[int] = None,
        return_attention_mask: Optional[bool] = None,
    ) -> dict:
        """
        Pad encoded inputs (on left/right and up to predefined length or max length in the batch)

        Args:
            encoded_inputs:
                Dictionary of tokenized inputs (`List[int]`) or batch of tokenized inputs (`List[List[int]]`).
            max_length: maximum length of the returned list and optionally padding length (see below).
                Will truncate by taking into account the special tokens.
            padding_strategy: PaddingStrategy to use for padding.

                - PaddingStrategy.LONGEST Pad to the longest sequence in the batch
                - PaddingStrategy.MAX_LENGTH: Pad to the max length (default)
                - PaddingStrategy.DO_NOT_PAD: Do not pad
                The tokenizer padding sides are defined in self.padding_side:

                    - 'left': pads on the left of the sequences
                    - 'right': pads on the right of the sequences
            pad_to_multiple_of: (optional) Integer if set will pad the sequence to a multiple of the provided value.
                This is especially useful to enable the use of Tensor Core on NVIDIA hardware with compute capability
                `>= 7.5` (Volta).
            return_attention_mask:
                (optional) Set to False to avoid returning attention mask (default: set to model specifics)
        """
        # Load from model defaults
        bos_token_id = self.sp_tokenizer[self.bos_token]
        mask_token_id = self.sp_tokenizer[self.mask_token]
        gmask_token_id = self.sp_tokenizer[self.gmask_token]
        assert self.padding_side == "left"

        required_input = encoded_inputs[self.model_input_names[0]]
        seq_length = len(required_input)

        if padding_strategy == PaddingStrategy.LONGEST:
            max_length = len(required_input)

        if max_length is not None and pad_to_multiple_of is not None and (max_length % pad_to_multiple_of != 0):
            max_length = ((max_length // pad_to_multiple_of) + 1) * pad_to_multiple_of

        needs_to_be_padded = padding_strategy != PaddingStrategy.DO_NOT_PAD and len(required_input) != max_length

        # Initialize attention mask if not present.
        if max_length is not None:
            if "attention_mask" not in encoded_inputs:
                if bos_token_id in required_input:
                    context_length = required_input.index(bos_token_id)
                else:
                    context_length = seq_length
                attention_mask = np.ones((1, seq_length, seq_length))
                attention_mask = np.tril(attention_mask)
                attention_mask[:, :, :context_length] = 1
                attention_mask = np.bool_(attention_mask < 0.5)
                encoded_inputs["attention_mask"] = attention_mask

            if "position_ids" not in encoded_inputs:
                position_ids = np.arange(seq_length, dtype=np.int64)
                mask_token = mask_token_id if mask_token_id in required_input else gmask_token_id
                if mask_token in required_input:
                    mask_position = required_input.index(mask_token)
                    position_ids[context_length:] = mask_position
                block_position_ids = np.concatenate(
                    [np.zeros(context_length, dtype=np.int64),
                     np.arange(1, seq_length - context_length + 1, dtype=np.int64)])
                encoded_inputs["position_ids"] = np.stack([position_ids, block_position_ids], axis=0)

        if needs_to_be_padded:
            difference = max_length - len(required_input)

            if "attention_mask" in encoded_inputs:
                encoded_inputs["attention_mask"] = np.pad(encoded_inputs["attention_mask"],
                                                          pad_width=[(0, 0), (difference, 0), (difference, 0)],
                                                          mode='constant', constant_values=True)
            if "token_type_ids" in encoded_inputs:
                encoded_inputs["token_type_ids"] = [self.pad_token_type_id] * difference + encoded_inputs[
                    "token_type_ids"
                ]
            if "special_tokens_mask" in encoded_inputs:
                encoded_inputs["special_tokens_mask"] = [1] * difference + encoded_inputs["special_tokens_mask"]
            if "position_ids" in encoded_inputs:
                encoded_inputs["position_ids"] = np.pad(encoded_inputs["position_ids"],
                                                        pad_width=[(0, 0), (difference, 0)])
            encoded_inputs[self.model_input_names[0]] = [self.pad_token_id] * difference + required_input

        return encoded_inputs