tokenization_transformerlm.py

# coding=utf-8
# Copyright 2018 Google AI, Google Brain 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.
""" Tokenization classes for ALBERT model."""


import os
import unicodedata
from shutil import copyfile
from typing import Any, Dict, List, Optional, Tuple

import sentencepiece as spm
from transformers.tokenization_utils import AddedToken, PreTrainedTokenizer


VOCAB_FILES_NAMES = {"vocab_file": "spiece.model"}

PRETRAINED_VOCAB_FILES_MAP = {
    "vocab_file": {
        "albert/albert-base-v1": "https://huggingface.co/albert/albert-base-v1/resolve/main/spiece.model",
        "albert/albert-large-v1": "https://huggingface.co/albert/albert-large-v1/resolve/main/spiece.model",
        "albert/albert-xlarge-v1": "https://huggingface.co/albert/albert-xlarge-v1/resolve/main/spiece.model",
        "albert/albert-xxlarge-v1": "https://huggingface.co/albert/albert-xxlarge-v1/resolve/main/spiece.model",
        "albert/albert-base-v2": "https://huggingface.co/albert/albert-base-v2/resolve/main/spiece.model",
        "albert/albert-large-v2": "https://huggingface.co/albert/albert-large-v2/resolve/main/spiece.model",
        "albert/albert-xlarge-v2": "https://huggingface.co/albert/albert-xlarge-v2/resolve/main/spiece.model",
        "albert/albert-xxlarge-v2": "https://huggingface.co/albert/albert-xxlarge-v2/resolve/main/spiece.model",
    }
}

PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
    "albert/albert-base-v1": 512,
    "albert/albert-large-v1": 512,
    "albert/albert-xlarge-v1": 512,
    "albert/albert-xxlarge-v1": 512,
    "albert/albert-base-v2": 512,
    "albert/albert-large-v2": 512,
    "albert/albert-xlarge-v2": 512,
    "albert/albert-xxlarge-v2": 512,
}

SPIECE_UNDERLINE = "▁"


class TransformerLMTokenizer(PreTrainedTokenizer):
    """
    Construct an ALBERT tokenizer. Based on [SentencePiece](https://github.com/google/sentencepiece).

    This tokenizer inherits from [`PreTrainedTokenizer`] which contains most of the main methods. Users should refer to
    this superclass for more information regarding those methods.

    Args:
        vocab_file (`str`):
            [SentencePiece](https://github.com/google/sentencepiece) file (generally has a *.spm* extension) that
            contains the vocabulary necessary to instantiate a tokenizer.
        do_lower_case (`bool`, *optional*, defaults to `True`):
            Whether or not to lowercase the input when tokenizing.
        remove_space (`bool`, *optional*, defaults to `True`):
            Whether or not to strip the text when tokenizing (removing excess spaces before and after the string).
        keep_accents (`bool`, *optional*, defaults to `False`):
            Whether or not to keep accents when tokenizing.
        bos_token (`str`, *optional*, defaults to `"[CLS]"`):
            The beginning of sequence token that was used during pretraining. Can be used a sequence classifier token.

            <Tip>

            When building a sequence using special tokens, this is not the token that is used for the beginning of
            sequence. The token used is the `cls_token`.

            </Tip>

        eos_token (`str`, *optional*, defaults to `"[SEP]"`):
            The end of sequence token.

            <Tip>

            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 `sep_token`.

            </Tip>

        unk_token (`str`, *optional*, defaults to `"<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.
        sep_token (`str`, *optional*, defaults to `"[SEP]"`):
            The separator token, which is used when building a sequence from multiple sequences, e.g. two sequences for
            sequence classification or for a text and a question for question answering. It is also used as the last
            token of a sequence built with special tokens.
        pad_token (`str`, *optional*, defaults to `"<pad>"`):
            The token used for padding, for example when batching sequences of different lengths.
        cls_token (`str`, *optional*, defaults to `"[CLS]"`):
            The classifier token which is used when doing sequence classification (classification of the whole sequence
            instead of per-token classification). It is the first token of the sequence when built with special tokens.
        mask_token (`str`, *optional*, defaults to `"[MASK]"`):
            The token used for masking values. This is the token used when training this model with masked language
            modeling. This is the token which the model will try to predict.
        sp_model_kwargs (`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.

    Attributes:
        sp_model (`SentencePieceProcessor`):
            The *SentencePiece* processor that is used for every conversion (string, tokens and IDs).
    """

    vocab_files_names = VOCAB_FILES_NAMES
    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
    max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES

    def __init__(
        self,
        vocab_file,
        do_lower_case=True,
        remove_space=True,
        keep_accents=False,
        bos_token="[CLS]",
        eos_token="[SEP]",
        unk_token="<unk>",
        sep_token="[SEP]",
        pad_token="<pad>",
        cls_token="[CLS]",
        mask_token="[MASK]",
        sp_model_kwargs: Optional[Dict[str, Any]] = None,
        **kwargs,
    ) -> None:
        # Mask token behave like a normal word, i.e. include the space before it and
        # is included in the raw text, there should be a match in a non-normalized sentence.
        mask_token = (
            AddedToken(mask_token, lstrip=True, rstrip=False, normalized=False)
            if isinstance(mask_token, str)
            else mask_token
        )

        self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs

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

        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
        self.sp_model.Load(vocab_file)

        super().__init__(
            do_lower_case=do_lower_case,
            remove_space=remove_space,
            keep_accents=keep_accents,
            bos_token=bos_token,
            eos_token=eos_token,
            unk_token=unk_token,
            sep_token=sep_token,
            pad_token=pad_token,
            cls_token=cls_token,
            mask_token=mask_token,
            sp_model_kwargs=self.sp_model_kwargs,
            **kwargs,
        )

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

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

        if not self.keep_accents:
            outputs = unicodedata.normalize("NFKD", outputs)
            outputs = "".join([c for c in outputs if not unicodedata.combining(c)])
        if self.do_lower_case:
            outputs = outputs.lower()

        return outputs

    def _tokenize(self, text: str) -> List[str]:
        """Tokenize a string."""
        text = self.preprocess_text(text)
        pieces = self.sp_model.encode(text, out_type=str)
        new_pieces = []
        for piece in pieces:
            if len(piece) > 1 and piece[-1] == str(",") and piece[-2].isdigit():
                # Logic to handle special cases see https://github.com/google-research/bert/blob/master/README.md#tokenization
                # `9,9` -> ['▁9', ',', '9'] instead of [`_9,`, '9']
                cur_pieces = self.sp_model.EncodeAsPieces(piece[:-1].replace(SPIECE_UNDERLINE, ""))
                if piece[0] != SPIECE_UNDERLINE and cur_pieces[0][0] == SPIECE_UNDERLINE:
                    if len(cur_pieces[0]) == 1:
                        cur_pieces = cur_pieces[1:]
                    else:
                        cur_pieces[0] = cur_pieces[0][1:]
                cur_pieces.append(piece[-1])
                new_pieces.extend(cur_pieces)
            else:
                new_pieces.append(piece)

        return new_pieces

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

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

    def convert_tokens_to_string(self, tokens):
        """Converts a sequence of tokens (string) in a single string."""
        current_sub_tokens = []
        out_string = ""
        prev_is_special = False
        for token in tokens:
            # make sure that special tokens are not decoded using sentencepiece model
            if token in self.all_special_tokens:
                if not prev_is_special:
                    out_string += " "
                out_string += self.sp_model.decode(current_sub_tokens) + token
                prev_is_special = True
                current_sub_tokens = []
            else:
                current_sub_tokens.append(token)
                prev_is_special = False
        out_string += self.sp_model.decode(current_sub_tokens)
        return out_string.strip()

    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. An ALBERT 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]
        if token_ids_1 is None:
            return cls + token_ids_0 + sep
        return cls + token_ids_0 + sep + token_ids_1 + sep

    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
            )

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

    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. An ALBERT
        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]

        if token_ids_1 is None:
            return len(cls + token_ids_0 + sep) * [0]
        return len(cls + token_ids_0 + sep) * [0] + len(token_ids_1 + sep) * [1]

    def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] = None) -> Tuple[str]:
        if not os.path.isdir(save_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) and os.path.isfile(self.vocab_file):
            copyfile(self.vocab_file, out_vocab_file)
        elif not os.path.isfile(self.vocab_file):
            with open(out_vocab_file, "wb") as fi:
                content_spiece_model = self.sp_model.serialized_model_proto()
                fi.write(content_spiece_model)

        return (out_vocab_file,)