From e12bbe3b4d4acab06a88b8e05fa6aecfe565b33a Mon Sep 17 00:00:00 2001 From: Shyam Sudhakaran Date: Wed, 5 Oct 2022 06:15:14 -0700 Subject: [PATCH] Remove bert interdependency from clip tokenizer (#19332) --- .../models/clip/tokenization_clip.py | 164 +++++++++++++++++- 1 file changed, 162 insertions(+), 2 deletions(-) diff --git a/src/transformers/models/clip/tokenization_clip.py b/src/transformers/models/clip/tokenization_clip.py index c6870cc69f5..29707ee5817 100644 --- a/src/transformers/models/clip/tokenization_clip.py +++ b/src/transformers/models/clip/tokenization_clip.py @@ -16,13 +16,13 @@ import json import os +import unicodedata from functools import lru_cache from typing import List, Optional, Tuple import regex as re -from transformers.models.bert.tokenization_bert import BasicTokenizer -from ...tokenization_utils import AddedToken, PreTrainedTokenizer +from ...tokenization_utils import AddedToken, PreTrainedTokenizer, _is_control, _is_punctuation, _is_whitespace from ...utils import logging @@ -97,6 +97,166 @@ def whitespace_clean(text): return text +# Copied from transformers.models.bert.tokenization_bert.whitespace_tokenize +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 + + +# Copied from transformers.models.bert.tokenization_bert.BasicTokenizer +class BasicTokenizer(object): + """ + Constructs a BasicTokenizer that will run basic tokenization (punctuation splitting, lower casing, etc.). + + Args: + do_lower_case (`bool`, *optional*, defaults to `True`): + Whether or not to lowercase the input when tokenizing. + never_split (`Iterable`, *optional*): + Collection of tokens which will never be split during tokenization. Only has an effect when + `do_basic_tokenize=True` + tokenize_chinese_chars (`bool`, *optional*, defaults to `True`): + Whether or not to tokenize Chinese characters. + + This should likely be deactivated for Japanese (see this + [issue](https://github.com/huggingface/transformers/issues/328)). + strip_accents (`bool`, *optional*): + Whether or not to strip all accents. If this option is not specified, then it will be determined by the + value for `lowercase` (as in the original BERT). + """ + + def __init__(self, do_lower_case=True, never_split=None, tokenize_chinese_chars=True, strip_accents=None): + if never_split is None: + never_split = [] + self.do_lower_case = do_lower_case + self.never_split = set(never_split) + self.tokenize_chinese_chars = tokenize_chinese_chars + self.strip_accents = strip_accents + + def tokenize(self, text, never_split=None): + """ + Basic Tokenization of a piece of text. Split on "white spaces" only, for sub-word tokenization, see + WordPieceTokenizer. + + Args: + never_split (`List[str]`, *optional*) + Kept for backward compatibility purposes. Now implemented directly at the base class level (see + [`PreTrainedTokenizer.tokenize`]) List of token not to split. + """ + # union() returns a new set by concatenating the two sets. + never_split = self.never_split.union(set(never_split)) if never_split else self.never_split + text = self._clean_text(text) + + # This was added on November 1st, 2018 for the multilingual and Chinese + # models. This is also applied to the English models now, but it doesn't + # matter since the English models were not trained on any Chinese data + # and generally don't have any Chinese data in them (there are Chinese + # characters in the vocabulary because Wikipedia does have some Chinese + # words in the English Wikipedia.). + if self.tokenize_chinese_chars: + text = self._tokenize_chinese_chars(text) + orig_tokens = whitespace_tokenize(text) + split_tokens = [] + for token in orig_tokens: + if token not in never_split: + if self.do_lower_case: + token = token.lower() + if self.strip_accents is not False: + token = self._run_strip_accents(token) + elif self.strip_accents: + token = self._run_strip_accents(token) + split_tokens.extend(self._run_split_on_punc(token, never_split)) + + output_tokens = whitespace_tokenize(" ".join(split_tokens)) + return output_tokens + + def _run_strip_accents(self, text): + """Strips accents from a piece of text.""" + 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, never_split=None): + """Splits punctuation on a piece of text.""" + if never_split is not None and text in never_split: + return [text] + chars = list(text) + i = 0 + start_new_word = True + output = [] + while i < len(chars): + char = chars[i] + if _is_punctuation(char): + 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 CLIPTokenizer(PreTrainedTokenizer): """ Construct a CLIP tokenizer. Based on byte-level Byte-Pair-Encoding.