# coding=utf-8
# Copyright 2020 Google and The HuggingFace Inc. team.
#
# 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.
import os
from shutil import copyfile
from typing import Any, Dict, List, Optional, Tuple
import sentencepiece as spm
from ...tokenization_utils import PreTrainedTokenizer
from ...utils import logging
SPIECE_UNDERLINE = "▁"
VOCAB_FILES_NAMES = {"vocab_file": "spiece.model"}
PRETRAINED_VOCAB_FILES_MAP = {
"vocab_file": {"google/pegasus-xsum": "https://huggingface.co/google/pegasus-xsum/resolve/main/spiece.model"}
}
PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
"google/pegasus-xsum": 512,
}
logger = logging.get_logger(__name__)
[docs]class PegasusTokenizer(PreTrainedTokenizer):
r"""
Construct a PEGASUS tokenizer. Based on `SentencePiece <https://github.com/google/sentencepiece>`__.
This tokenizer inherits from :class:`~transformers.PreTrainedTokenizer` which contains most of the main methods.
Users should refer to this superclass for more information regarding those methods.
Args:
vocab_file (:obj:`str`):
`SentencePiece <https://github.com/google/sentencepiece>`__ file (generally has a `.spm` extension) that
contains the vocabulary necessary to instantiate a tokenizer.
pad_token (:obj:`str`, `optional`, defaults to :obj:`"<pad>"`):
The token used for padding, for example when batching sequences of different lengths.
eos_token (:obj:`str`, `optional`, defaults to :obj:`"</s>"`):
The end of sequence token.
.. note::
When building a sequence using special tokens, this is not the token that is used for the end of
sequence. The token used is the :obj:`sep_token`.
unk_token (:obj:`str`, `optional`, defaults to :obj:`"<unk>"`):
The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this
token instead.
mask_token (:obj:`str`, `optional`, defaults to :obj:`"<mask_2>"`):
The token used for masking single token values. This is the token used when training this model with masked
language modeling (MLM). This is the token that the PEGASUS encoder will try to predict during pretraining.
It corresponds to `[MASK2]` in `PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive
Summarization <https://arxiv.org/pdf/1912.08777.pdf>`__.
mask_token_sent (:obj:`str`, `optional`, defaults to :obj:`"<mask_1>"`):
The token used for masking whole target sentences. This is the token used when training this model with gap
sentences generation (GSG). This is the sentence that the PEGASUS decoder will try to predict during
pretraining. It corresponds to `[MASK1]` in `PEGASUS: Pre-training with Extracted Gap-sentences for
Abstractive Summarization <https://arxiv.org/pdf/1912.08777.pdf>`__.
additional_special_tokens (:obj:`List[str]`, `optional`):
Additional special tokens used by the tokenizer. If no additional_special_tokens are provided <mask_2> and
<unk_2, ..., unk_102> are used as additional special tokens corresponding to the `original PEGASUS
tokenizer
<https://github.com/google-research/pegasus/blob/939830367bcf411193d2b5eca2f2f90f3f9260ca/pegasus/ops/pretrain_parsing_ops.cc#L66>`__
that uses the tokens 2 - 104 only for pretraining
sp_model_kwargs (:obj:`dict`, `optional`):
Will be passed to the ``SentencePieceProcessor.__init__()`` method. The `Python wrapper for SentencePiece
<https://github.com/google/sentencepiece/tree/master/python>`__ can be used, among other things, to set:
- ``enable_sampling``: Enable subword regularization.
- ``nbest_size``: Sampling parameters for unigram. Invalid for BPE-Dropout.
- ``nbest_size = {0,1}``: No sampling is performed.
- ``nbest_size > 1``: samples from the nbest_size results.
- ``nbest_size < 0``: assuming that nbest_size is infinite and samples from the all hypothesis (lattice)
using forward-filtering-and-backward-sampling algorithm.
- ``alpha``: Smoothing parameter for unigram sampling, and dropout probability of merge operations for
BPE-dropout.
"""
vocab_files_names = VOCAB_FILES_NAMES
vocab_files_names = VOCAB_FILES_NAMES
pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
model_input_names = ["input_ids", "attention_mask"]
def __init__(
self,
vocab_file,
pad_token="<pad>",
eos_token="</s>",
unk_token="<unk>",
mask_token="<mask_2>",
mask_token_sent="<mask_1>",
additional_special_tokens=None,
offset=103, # entries 2 - 104 are only used for pretraining
sp_model_kwargs: Optional[Dict[str, Any]] = None,
**kwargs
) -> None:
self.offset = offset
if additional_special_tokens is not None:
assert isinstance(
additional_special_tokens, list
), f"additional_special_tokens should be of type {type(list)}, but is {type(additional_special_tokens)}"
additional_special_tokens_extended = (
([mask_token_sent] + additional_special_tokens)
if mask_token_sent not in additional_special_tokens and mask_token_sent is not None
else additional_special_tokens
)
# fill additional tokens with ..., <unk_token_102> in case not all additional tokens are already taken
additional_special_tokens_extended += [
f"<unk_{i}>" for i in range(len(additional_special_tokens_extended), self.offset - 1)
]
if len(set(additional_special_tokens_extended)) != len(additional_special_tokens_extended):
raise ValueError(
f"Please make sure that the provided additional_special_tokens do not contain an incorrectly shifted list of <unk_x> tokens. Found {additional_special_tokens_extended}."
)
additional_special_tokens = additional_special_tokens_extended
else:
additional_special_tokens = [mask_token_sent] if mask_token_sent is not None else []
additional_special_tokens += [f"<unk_{i}>" for i in range(2, self.offset)]
self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
super().__init__(
eos_token=eos_token,
unk_token=unk_token,
mask_token=mask_token,
pad_token=pad_token,
mask_token_sent=mask_token_sent,
offset=offset,
additional_special_tokens=additional_special_tokens,
sp_model_kwargs=self.sp_model_kwargs,
**kwargs,
)
self.mask_token_sent = mask_token_sent
self.vocab_file = vocab_file
self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
self.sp_model.Load(vocab_file)
# add special tokens to encoder dict
self.encoder: Dict[int, str] = {
0: self.pad_token,
1: self.eos_token,
}
if self.mask_token_sent is not None:
self.encoder.update(
{
2: self.mask_token_sent,
3: self.mask_token,
}
)
if self.offset > 0:
# entries 2-104 are only used for pretraining and called <mask_1>, <mask_2>, unk_2, ...unk_102
# mask_token_sent is already added to list -> so start at 1
self.encoder.update({i + 3: additional_special_tokens[i] for i in range(1, self.offset - 1)})
self.decoder: Dict[str, int] = {v: k for k, v in self.encoder.items()}
@property
def vocab_size(self) -> int:
return len(self.sp_model) + self.offset
[docs] def get_vocab(self) -> Dict[str, int]:
vocab = {self.convert_ids_to_tokens(i): i for i in range(self.vocab_size)}
vocab.update(self.added_tokens_encoder)
return vocab
def __getstate__(self):
state = self.__dict__.copy()
state["sp_model"] = None
return state
def __setstate__(self, d):
self.__dict__ = d
# for backward compatibility
if not hasattr(self, "sp_model_kwargs"):
self.sp_model_kwargs = {}
self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
self.sp_model.Load(self.vocab_file)
def _tokenize(self, text: str) -> List[str]:
"""Take as input a string and return a list of strings (tokens) for words/sub-words"""
return self.sp_model.encode(text, out_type=str)
def _convert_token_to_id(self, token: str) -> int:
"""Converts a token (str) to an id using the vocab."""
if token in self.decoder:
return self.decoder[token]
elif token in self.added_tokens_decoder:
return self.added_tokens_decoder[token]
sp_id = self.sp_model.piece_to_id(token)
return sp_id + self.offset
def _convert_id_to_token(self, index: int) -> str:
"""Converts an index (integer) to a token (str) using the vocab."""
if index in self.encoder:
return self.encoder[index]
elif index in self.added_tokens_encoder:
return self.added_tokens_encoder[index]
else:
token = self.sp_model.IdToPiece(index - self.offset)
return token
[docs] def convert_tokens_to_string(self, tokens):
"""Converts a sequence of tokens (string) in a single string."""
out_string = self.sp_model.decode_pieces(tokens)
return out_string
[docs] def num_special_tokens_to_add(self, pair=False):
"""Just EOS"""
return 1
def _special_token_mask(self, seq):
all_special_ids = set(self.all_special_ids) # call it once instead of inside list comp
all_special_ids.remove(self.unk_token_id) # <unk> is only sometimes special
return [1 if x in all_special_ids else 0 for x in seq]
[docs] def get_special_tokens_mask(
self, token_ids_0: List, token_ids_1: Optional[List] = None, already_has_special_tokens: bool = False
) -> List[int]:
"""Get list where entries are [1] if a token is [eos] or [pad] else 0."""
if already_has_special_tokens:
return self._special_token_mask(token_ids_0)
elif token_ids_1 is None:
return self._special_token_mask(token_ids_0) + [1]
else:
return self._special_token_mask(token_ids_0 + token_ids_1) + [1]
[docs] def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] = None) -> Tuple[str]:
if not os.path.isdir(save_directory):
logger.error(f"Vocabulary path ({save_directory}) should be a directory")
return
out_vocab_file = os.path.join(
save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"]
)
if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file):
copyfile(self.vocab_file, out_vocab_file)
return (out_vocab_file,)