tokenization_hanbert.py

# coding=utf-8
# Modified by TwoBlock Ai, Jangwon Park
# for Korean, HanBert + Moran
#
# youngcho@tbai.info, youngcho@gmail.com
# adieujw@gmail.com (https://github.com/monologg)
#
# Copyright 2018 The Google AI Language Team Authors.
#
# 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."""
from __future__ import absolute_import, division, print_function, unicode_literals

import collections
import re
import unicodedata
import six
from ctypes import *

import logging
import os
from shutil import copyfile

import unicodedata
import six

from transformers import PreTrainedTokenizer

logger = logging.getLogger(__name__)

VOCAB_FILES_NAMES = {'vocab_file': 'vocab_54k.txt',
                     'moran_file': 'libmoran4dnlp.so'}

PRETRAINED_INIT_CONFIGURATION = {
    'HanBert-54kN-torch': {'do_lower_case': False},
    'HanBert-54kN-IP-torch': {'do_lower_case': False}
}


class MorAn16(object):
    def __init__(self, moran_file='libmoran4dnlp.so'):
        self.moran = CDLL(moran_file)
        self.moran.Moran4dnlp.restype = c_char_p
        self.moran.Moran4dnlp.argtypes = [c_char_p, c_char_p, c_int]

        self.moran.Moran4match.restype = c_char_p
        self.moran.Moran4match.argtypes = [c_char_p, c_char_p, c_char_p, c_int]

        self.moran_input = create_string_buffer(102400)
        self.moran_input2 = create_string_buffer(102400)
        self.moran_output = create_string_buffer(102400)
        self.moran.MorAn16_open_dbs()

    def close(self):
        self.moran.MorAn_close_dbs()

    def run(self, text):
        result = list()
        self.moran_input.value = text.encode()
        x = self.moran.Moran4dnlp(self.moran_input, self.moran_output, 102400)
        y = x.decode()
        result = y.split()
        return result

    def match(self, text, tokens):
        result = list()
        self.moran_input.value = text.encode()
        self.moran_input2.value = tokens.encode()
        x = self.moran.Moran4match(self.moran_input, self.moran_input2, self.moran_output, 102400)
        y = x.decode()
        result = y.split()
        return result


def convert_to_unicode(text):
    """Converts `text` to Unicode (if it's not already), assuming utf-8 input."""
    if six.PY3:
        if isinstance(text, str):
            return text
        elif isinstance(text, bytes):
            return text.decode("utf-8", "ignore")
        else:
            raise ValueError("Unsupported string type: %s" % (type(text)))
    elif six.PY2:
        if isinstance(text, str):
            return text.decode("utf-8", "ignore")
        elif isinstance(text, unicode):
            return text
        else:
            raise ValueError("Unsupported string type: %s" % (type(text)))
    else:
        raise ValueError("Not running on Python2 or Python 3?")


def load_vocab(vocab_file):
    """Loads a vocabulary file into a dictionary."""
    vocab = collections.OrderedDict()
    index = 0
    with open(vocab_file, 'r', encoding='utf-8') as reader:
        while True:
            token = convert_to_unicode(reader.readline())
            if not token:
                break
            token = token.strip()
            vocab[token] = index
            index += 1
    return vocab


def whitespace_tokenize(text):
    """Runs basic whitespace cleaning and splitting on a piece of text."""
    text = text.strip()
    if not text:
        return []
    tokens = text.split()
    return tokens


class BasicTokenizer(object):
    """Runs basic tokenization (punctuation splitting, lower casing, etc.)."""

    def __init__(self, use_moran=False, use_zwj=True, moran_file='libmoran4dnlp.so'):
        self.moran = None
        try:
            self.moran = MorAn16(moran_file)
        except:
            logger.warning("Only ubuntu is supported for HanBertTokenizer!")

        self.use_moran = use_moran
        self.use_zwj = use_zwj

    def tokenize(self, text):
        """Tokenizes a piece of text."""
        text = convert_to_unicode(text)
        text = self._clean_text(text)
        text = self._tokenize_chinese_chars(text)

        if self.use_moran:
            output_tokens = self.moran.run(text)
            return output_tokens
        else:
            orig_tokens = whitespace_tokenize(text)

        split_tokens = []
        for token in orig_tokens:
            token = token.lower()
            token = self._run_strip_accents(token)
            split_tokens.extend(self._run_split_on_punc(token))

        output_tokens = whitespace_tokenize(" ".join(split_tokens))
        return output_tokens

    def match(self, text, pattern):
        """Tokenizes a piece of text."""
        text = convert_to_unicode(text)
        text = self._clean_text(text)
        text = self._tokenize_chinese_chars(text)
        pattern = convert_to_unicode(pattern)
        pattern = self._clean_text(pattern)
        pattern = self._tokenize_chinese_chars(pattern)
        positions = self.moran.match(text, pattern)
        start = int(positions[0])
        end = int(positions[1])
        output_tokens = text[start:end]
        return output_tokens

    def _run_strip_accents(self, text):
        """Strips accents from a piece of text."""
        # Skip using _run_strip_accents() for Korean substrings, since normalizing
        # Korean characters with NFD and joining them back results in a seemingly
        # same but different text, which causes a bug.

        text = unicodedata.normalize("NFD", text)
        output = []
        for char in text:
            cat = unicodedata.category(char)
            if cat == "Mn":
                continue
            output.append(char)
        return "".join(output)

    def _run_split_on_punc(self, text):
        """Splits punctuation on a piece of text."""
        chars = list(text)
        i = 0
        start_new_word = True
        output = []
        while i < len(chars):
            char = chars[i]
            if _is_punctuation(char, self.use_zwj):
                output.append([char])
                start_new_word = True
            else:
                if start_new_word:
                    output.append([])
                start_new_word = False
                output[-1].append(char)
            i += 1

        return ["".join(x) for x in output]

    def _tokenize_chinese_chars(self, text):
        """Adds whitespace around any CJK character."""
        output = []
        for char in text:
            cp = ord(char)
            if self._is_chinese_char(cp):
                output.append(" ")
                output.append(char)
                output.append(" ")
            else:
                output.append(char)
        return "".join(output)

    def _is_chinese_char(self, cp):
        """Checks whether CP is the codepoint of a CJK character."""
        # This defines a "chinese character" as anything in the CJK Unicode block:
        #   https://en.wikipedia.org/wiki/CJK_Unified_Ideographs_(Unicode_block)
        #
        # Note that the CJK Unicode block is NOT all Japanese and Korean characters,
        # despite its name. The modern Korean Hangul alphabet is a different block,
        # as is Japanese Hiragana and Katakana. Those alphabets are used to write
        # space-separated words, so they are not treated specially and handled
        # like the all of the other languages.
        if ((cp >= 0x4E00 and cp <= 0x9FFF) or  #
            (cp >= 0x3400 and cp <= 0x4DBF) or  #
            (cp >= 0x20000 and cp <= 0x2A6DF) or  #
            (cp >= 0x2A700 and cp <= 0x2B73F) or  #
            (cp >= 0x2B740 and cp <= 0x2B81F) or  #
            (cp >= 0x2B820 and cp <= 0x2CEAF) or
            (cp >= 0xF900 and cp <= 0xFAFF) or  #
                (cp >= 0x2F800 and cp <= 0x2FA1F)):  #
            return True

        return False

    def _clean_text(self, text):
        """Performs invalid character removal and whitespace cleanup on text."""
        output = []
        for char in text:
            cp = ord(char)
            if cp == 0 or cp == 0xfffd or _is_control(char):
                continue
            if _is_whitespace(char):
                output.append(" ")
            else:
                output.append(char)
        return "".join(output)


class WordpieceTokenizer(object):
    """Runs WordPiece tokenziation."""

    def __init__(self, vocab, unk_token="[UNK]", max_input_chars_per_word=200):
        self.vocab = vocab
        self.unk_token = unk_token
        self.max_input_chars_per_word = max_input_chars_per_word

    def tokenize(self, text):
        """Tokenizes a piece of text into its word pieces.

        This uses a greedy longest-match-first algorithm to perform tokenization
        using the given vocabulary.

        For example:
          input = "unaffable"
          output = ["un", "##aff", "##able"]

        Args:
          text: A single token or whitespace separated tokens. This should have
            already been passed through `BasicTokenizer.

        Returns:
          A list of wordpiece tokens.
        """

        text = convert_to_unicode(text)
        output_tokens = []
        for token in whitespace_tokenize(text):
            chars = list(token)
            if len(chars) > self.max_input_chars_per_word:
                output_tokens.append(self.unk_token)
                continue

            is_bad = False
            start = 0
            sub_tokens = []
            while start < len(chars):
                end = len(chars)
                cur_substr = None
                while start < end:
                    substr = "".join(chars[start:end])
                    if start > 0:
                        substr = "##" + substr
                    if substr in self.vocab:
                        cur_substr = substr
                        break
                    end -= 1
                if cur_substr is None:
                    is_bad = True
                    break
                sub_tokens.append(cur_substr)
                start = end

            if is_bad:
                output_tokens.append(self.unk_token)
            else:
                output_tokens.extend(sub_tokens)
        return output_tokens


def _is_whitespace(char):
    """Checks whether `chars` is a whitespace character."""
    # \t, \n, and \r are technically contorl characters but we treat them
    # as whitespace since they are generally considered as such.
    if char == " " or char == "\t" or char == "\n" or char == "\r":
        return True
    cat = unicodedata.category(char)
    if cat == "Zs":
        return True
    return False


def _is_control(char):
    """Checks whether `chars` is a control character."""
    # These are technically control characters but we count them as whitespace
    # characters.
    if char == "\t" or char == "\n" or char == "\r":
        return False
    cat = unicodedata.category(char)
    if cat.startswith("C"):
        return True
    return False


def _is_punctuation(char, use_zwj=False):
    """Checks whether `chars` is a punctuation character."""
    if use_zwj and char == "~":
        return False
    cp = ord(char)
    # We treat all non-letter/number ASCII as punctuation.
    # Characters such as "^", "$", and "`" are not in the Unicode
    # Punctuation class but we treat them as punctuation anyways, for
    # consistency.
    if ((cp >= 33 and cp <= 47) or (cp >= 58 and cp <= 64) or
            (cp >= 91 and cp <= 96) or (cp >= 123 and cp <= 126)):
        return True
    cat = unicodedata.category(char)
    if cat.startswith("P"):
        return True
    return False


class HanBertTokenizer(PreTrainedTokenizer):
    r"""
    Constructs a BertTokenizer.
    :class:`~transformers.BertTokenizer` runs end-to-end tokenization: punctuation splitting + wordpiece

    Args:
        vocab_file: Path to a one-wordpiece-per-line vocabulary file
        do_lower_case: Whether to lower case the input. Only has an effect when do_wordpiece_only=False
        do_basic_tokenize: Whether to do basic tokenization before wordpiece.
        max_len: An artificial maximum length to truncate tokenized sequences to; Effective maximum length is always the
            minimum of this value (if specified) and the underlying BERT model's sequence length.
        never_split: List of tokens which will never be split during tokenization. Only has an effect when
            do_wordpiece_only=False
    """

    vocab_files_names = VOCAB_FILES_NAMES
    pretrained_init_configuration = PRETRAINED_INIT_CONFIGURATION

    def __init__(self, vocab_file, moran_file, do_lower_case=False, never_split=None, do_basic_tokenize=True, use_moran=True,
                 unk_token="[UNK]", sep_token="[SEP]", pad_token="[PAD]", cls_token="[CLS]",
                 mask_token="[MASK]", **kwargs):
        """Constructs a BertTokenizer.

        Args:
            **vocab_file**: Path to a one-wordpiece-per-line vocabulary file
            **do_lower_case**: (`optional`) boolean (default True)
                Whether to lower case the input
                Only has an effect when do_basic_tokenize=True
            **do_basic_tokenize**: (`optional`) boolean (default True)
                Whether to do basic tokenization before wordpiece.
            **never_split**: (`optional`) list of string
                List of tokens which will never be split during tokenization.
                Only has an effect when do_basic_tokenize=True
            **tokenize_chinese_chars**: (`optional`) boolean (default True)
                Whether to tokenize Chinese characters.
                This should likely be deactivated for Japanese:
                see: https://github.com/huggingface/pytorch-pretrained-BERT/issues/328
        """
        super(HanBertTokenizer, self).__init__(unk_token=unk_token, sep_token=sep_token,
                                               pad_token=pad_token, cls_token=cls_token,
                                               mask_token=mask_token, **kwargs)
        self.max_len = 512  # hard-coded
        self.max_len_single_sentence = self.max_len - 2  # take into account special tokens
        self.max_len_sentences_pair = self.max_len - 3  # take into account special tokens

        if not os.path.isfile(vocab_file):
            raise ValueError(
                "Can't find a vocabulary file at path '{}'. To load the vocabulary from a Google pretrained "
                "model use `tokenizer = HanBertTokenizer.from_pretrained(PRETRAINED_MODEL_NAME)`".format(vocab_file))
        self.vocab = load_vocab(vocab_file)
        self.ids_to_tokens = collections.OrderedDict(
            [(ids, tok) for tok, ids in self.vocab.items()])
        self.do_basic_tokenize = do_basic_tokenize
        self.do_lower_case = do_lower_case
        if do_basic_tokenize:
            self.basic_tokenizer = BasicTokenizer(use_moran=use_moran, moran_file=moran_file)
        self.wordpiece_tokenizer = WordpieceTokenizer(vocab=self.vocab, unk_token=self.unk_token)

    @property
    def vocab_size(self):
        return len(self.vocab)

    def _tokenize(self, text):
        split_tokens = []
        for token in self.basic_tokenizer.tokenize(text):
            for sub_token in self.wordpiece_tokenizer.tokenize(token):
                split_tokens.append(sub_token)
        return split_tokens

    def _convert_token_to_id(self, token):
        """ Converts a token (str/unicode) in an id using the vocab. """
        return self.vocab.get(token, self.vocab.get(self.unk_token))

    def _convert_id_to_token(self, index):
        """Converts an index (integer) in a token (string/unicode) using the vocab."""
        return self.ids_to_tokens.get(index, self.unk_token)

    def convert_tokens_to_string(self, tokens):
        """ Converts a sequence of tokens (string) in a single string. """
        tok_text = ' '.join(tokens)

        # De-tokenize WordPieces that have been split off.
        tok_text = tok_text.replace(" ##", "")
        tok_text = tok_text.replace(" ~~", "")
        tok_text = tok_text.replace(" ~", "")

        tok_text = tok_text.strip()
        tok_text = " ".join(tok_text.split())

        return tok_text

    def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
        """
        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]
        """
        if token_ids_1 is None:
            return [self.cls_token_id] + token_ids_0 + [self.sep_token_id]
        cls = [self.cls_token_id]
        sep = [self.sep_token_id]
        return cls + token_ids_0 + sep + token_ids_1 + sep

    def get_special_tokens_mask(self, token_ids_0, token_ids_1=None, already_has_special_tokens=False):
        """
        Retrieves 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`` or ``encode_plus`` methods.

        Args:
            token_ids_0: list of ids (must not contain special tokens)
            token_ids_1: Optional list of ids (must not contain special tokens), necessary when fetching sequence ids
                for sequence pairs
            already_has_special_tokens: (default False) Set to True if the token list is already formated with
                special tokens for the model

        Returns:
            A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
        """

        if already_has_special_tokens:
            if token_ids_1 is not None:
                raise ValueError("You should not supply a second sequence if the provided sequence of "
                                 "ids is already formated with special tokens for the model.")
            return list(map(lambda x: 1 if x in [self.sep_token_id, self.cls_token_id] else 0, token_ids_0))

        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, token_ids_1=None):
        """
        Creates a mask from the two sequences passed to be used in a sequence-pair classification task.
        A BERT sequence pair mask has the following format:
        0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1
        | first sequence    | second sequence

        if token_ids_1 is None, only returns the first portion of the mask (0'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, vocab_path):
        """Save the tokenizer vocabulary to a directory or file."""
        index = 0
        if os.path.isdir(vocab_path):
            vocab_file = os.path.join(vocab_path, VOCAB_FILES_NAMES['vocab_file'])
        else:
            vocab_file = vocab_path
        with open(vocab_file, "w", encoding="utf-8") as writer:
            for token, token_index in sorted(self.vocab.items(), key=lambda kv: kv[1]):
                if index != token_index:
                    logger.warning("Saving vocabulary to {}: vocabulary indices are not consecutive."
                                   " Please check that the vocabulary is not corrupted!".format(vocab_file))
                    index = token_index
                writer.write(token + u'\n')
                index += 1
        return (vocab_file,)