Tokenization behave the same as original XLM proprocessing for most languages except zh, ja and th; Change API to allow specifying language in tokenize

pytorch_transformers/tokenization_xlm.py

@@ -20,8 +20,11 @@ import json
 import logging
 import os
 import re
+import unicodedata
 from io import open
 
+import sacremoses as sm
+
 from .tokenization_utils import PreTrainedTokenizer
 from .tokenization_bert import BasicTokenizer
 
@@ -95,6 +98,93 @@ def text_standardize(text):
     text = re.sub(r'[^\S\n]+', ' ', text)
     return text.strip()
 
+
+def lowercase_and_remove_accent(text):
+    """
+    Lowercase and strips accents from a piece of text based on
+    https://github.com/facebookresearch/XLM/blob/master/tools/lowercase_and_remove_accent.py
+    """
+    text = text.lower()
+    text = unicodedata.normalize("NFD", text)
+    output = []
+    for char in text:
+        cat = unicodedata.category(char)
+        if cat == "Mn":
+            continue
+        output.append(char)
+    return "".join(output).lower()
+
+
+def replace_unicode_punct(text):
+    '''
+    Port of https://github.com/moses-smt/mosesdecoder/blob/master/scripts/tokenizer/replace-unicode-punctuation.perl
+    '''
+    text = text.replace('，', ',')
+    text = text.replace('。 *', '. ')
+    text = text.replace('、', ',')
+    text = text.replace('”', '"')
+    text = text.replace('“', '"')
+    text = text.replace('∶', ':')
+    text = text.replace('：', ':')
+    text = text.replace('？', '?')
+    text = text.replace('《', '"')
+    text = text.replace('》', '"')
+    text = text.replace('）', ')')
+    text = text.replace('！', '!')
+    text = text.replace('（', '(')
+    text = text.replace('；', ';')
+    text = text.replace('１', '"')
+    text = text.replace('」', '"')
+    text = text.replace('「', '"')
+    text = text.replace('０', '0')
+    text = text.replace('３', '3')
+    text = text.replace('２', '2')
+    text = text.replace('５', '5')
+    text = text.replace('６', '6')
+    text = text.replace('９', '9')
+    text = text.replace('７', '7')
+    text = text.replace('８', '8')
+    text = text.replace('４', '4')
+    text = re.sub(r'．\s*', '. ', text)
+    text = text.replace('～', '~')
+    text = text.replace('’', '\'')
+    text = text.replace('…', '...')
+    text = text.replace('━', '-')
+    text = text.replace('〈', '<')
+    text = text.replace('〉', '>')
+    text = text.replace('【', '[')
+    text = text.replace('】', ']')
+    text = text.replace('％', '%')
+    return text
+
+
+def remove_non_printing_char(text):
+    '''
+    Port of https://github.com/moses-smt/mosesdecoder/blob/master/scripts/tokenizer/remove-non-printing-char.perl
+    '''
+    output = []
+    for char in text:
+        cat = unicodedata.category(char)
+        if cat.startswith('C'):
+            continue
+        output.append(char)
+    return "".join(output)
+
+
+def romanian_preprocessing(text):
+    '''Sennrich's WMT16 scripts for Romanian preprocessing, used by model `xlm-mlm-enro-1024`'''
+    # https://github.com/rsennrich/wmt16-scripts/blob/master/preprocess/normalise-romanian.py
+    text = text.replace("\u015e", "\u0218").replace("\u015f", "\u0219")
+    text = text.replace("\u0162", "\u021a").replace("\u0163", "\u021b")
+    # https://github.com/rsennrich/wmt16-scripts/blob/master/preprocess/remove-diacritics.py
+    text = text.replace("\u0218", "S").replace("\u0219", "s") #s-comma
+    text = text.replace("\u021a", "T").replace("\u021b", "t") #t-comma
+    text = text.replace("\u0102", "A").replace("\u0103", "a")
+    text = text.replace("\u00C2", "A").replace("\u00E2", "a")
+    text = text.replace("\u00CE", "I").replace("\u00EE", "i")
+    return text
+
+
 class XLMTokenizer(PreTrainedTokenizer):
     """
     BPE tokenizer for XLM, adapted from OpenAI BPE tokenizer. Peculiarities:
@@ -122,16 +212,14 @@ class XLMTokenizer(PreTrainedTokenizer):
                                            cls_token=cls_token, mask_token=mask_token,
                                            additional_special_tokens=additional_special_tokens,
                                            **kwargs)
-        try:
+
-            import ftfy
+        # cache of sm.MosesPunctNormalizer instance
-            from spacy.lang.en import English
+        self.cache_moses_punct_normalizer = dict()
-            _nlp = English()
+        # cache of sm.MosesTokenizer instance
-            self.nlp = _nlp.Defaults.create_tokenizer(_nlp)
+        self.cache_moses_tokenizer = dict()
-            self.fix_text = ftfy.fix_text
+        self.lang_with_custom_tokenizer = set(['zh', 'th', 'ja'])
-        except ImportError:
+        # True for current supported model (v1.2.0), False for XLM-17 & 100
-            logger.warning("ftfy or spacy is not installed using BERT BasicTokenizer instead of SpaCy & ftfy.")
+        self.do_lowercase_and_remove_accent = True
-            self.nlp = BasicTokenizer(do_lower_case=True)
-            self.fix_text = None
 
         self.encoder = json.load(open(vocab_file, encoding="utf-8"))
         self.decoder = {v:k for k,v in self.encoder.items()}
@@ -140,6 +228,28 @@ class XLMTokenizer(PreTrainedTokenizer):
         self.bpe_ranks = dict(zip(merges, range(len(merges))))
         self.cache = {}
 
+    def moses_punct_norm(self, text, lang):
+        if lang not in self.cache_moses_punct_normalizer:
+            punct_normalizer = sm.MosesPunctNormalizer(lang=lang)
+            self.cache_moses_punct_normalizer[lang] = punct_normalizer
+        else:
+            punct_normalizer = self.cache_moses_punct_normalizer[lang]
+        return punct_normalizer.normalize(text)
+
+    def moses_tokenize(self, text, lang):
+        if lang not in self.cache_moses_tokenizer:
+            moses_tokenizer = sm.MosesTokenizer(lang=lang)
+            self.cache_moses_tokenizer[lang] = moses_tokenizer
+        else:
+            moses_tokenizer = self.cache_moses_tokenizer[lang]
+        return moses_tokenizer.tokenize(text, return_str=False, escape=False)
+
+    def moses_pipeline(self, text, lang):
+        text = replace_unicode_punct(text)
+        text = self.moses_punct_norm(text, lang)
+        text = remove_non_printing_char(text)
+        return text
+
     @property
     def vocab_size(self):
         return len(self.encoder)
@@ -187,19 +297,21 @@ class XLMTokenizer(PreTrainedTokenizer):
         self.cache[token] = word
         return word
 
-    def _tokenize(self, text):
+    def _tokenize(self, text, lang='en'):
         """ Tokenize a string. """
         split_tokens = []
-        if self.fix_text is None:
+        if self.do_lowercase_and_remove_accent:
-            # Using BERT's BasicTokenizer
+            text = lowercase_and_remove_accent(text)
-            text = self.nlp.tokenize(text)
+        if lang not in self.lang_with_custom_tokenizer:
+            text = self.moses_pipeline(text, lang=lang)
+            # TODO: make sure we are using `xlm-mlm-enro-1024`, since XLM-100 doesn't have this step
+            if lang == 'ro':
+                text = romanian_preprocessing(text)
+            text = self.moses_tokenize(text, lang=lang)
             for token in text:
                 split_tokens.extend([t for t in self.bpe(token).split(' ')])
         else:
-            # Using SpaCy & ftfy (original tokenization process of OpenAI GPT)
+            raise ValueError
-            text = self.nlp(text_standardize(self.fix_text(text)))
-            for token in text:
-                split_tokens.extend([t for t in self.bpe(token.text.lower()).split(' ')])
         return split_tokens
 
     def _convert_token_to_id(self, token):

requirements.txt

@@ -10,3 +10,5 @@ requests
 regex
 # For XLNet
 sentencepiece
+# For XLM
+sacremoses

setup.py

@@ -55,7 +55,8 @@ setup(
                       'requests',
                       'tqdm',
                       'regex',
-                      'sentencepiece'],
+                      'sentencepiece',
+                      'sacremoses'],
     entry_points={
       'console_scripts': [
         "pytorch_transformers=pytorch_transformers.__main__:main",