无法加载变压器型号

Question

我在加载变压器模型时遇到以下问题。奇怪的是它可以在google colab上运行，甚至当我在另一台计算机上尝试时，它似乎是版本/缓存问题但我没有找到它。

from sentence_transformers import SentenceTransformer
from sentence_transformers.util import cos_sim
model = SentenceTransformer('sentence-transformers/paraphrase-xlm-r-multilingual-v1')

---------------------------------------------------------------------------
Exception                                 Traceback (most recent call last)
<ipython-input-8-0b8b6a3eea75> in <module>
      1 from sentence_transformers import SentenceTransformer
      2 from sentence_transformers.util import cos_sim
----> 3 model = SentenceTransformer('sentence-transformers/paraphrase-xlm-r-multilingual-v1')
      4 

~\AppData\Local\Programs\Python\Python39\lib\site-packages\sentence_transformers\SentenceTransformer.py in __init__(self, model_name_or_path, modules, device, cache_folder)
     88 
     89             if os.path.exists(os.path.join(model_path, 'modules.json')):    #Load as SentenceTransformer model
---> 90                 modules = self._load_sbert_model(model_path)
     91             else:   #Load with AutoModel
     92                 modules = self._load_auto_model(model_path)

~\AppData\Local\Programs\Python\Python39\lib\site-packages\sentence_transformers\SentenceTransformer.py in _load_sbert_model(self, model_path)
    820         for module_config in modules_config:
    821             module_class = import_from_string(module_config['type'])
--> 822             module = module_class.load(os.path.join(model_path, module_config['path']))
    823             modules[module_config['name']] = module
    824 

~\AppData\Local\Programs\Python\Python39\lib\site-packages\sentence_transformers\models\Transformer.py in load(input_path)
    122         with open(sbert_config_path) as fIn:
    123             config = json.load(fIn)
--> 124         return Transformer(model_name_or_path=input_path, **config)
    125 
    126 

~\AppData\Local\Programs\Python\Python39\lib\site-packages\sentence_transformers\models\Transformer.py in __init__(self, model_name_or_path, max_seq_length, model_args, cache_dir, tokenizer_args, do_lower_case, tokenizer_name_or_path)
     28         config = AutoConfig.from_pretrained(model_name_or_path, **model_args, cache_dir=cache_dir)
     29         self.auto_model = AutoModel.from_pretrained(model_name_or_path, config=config, cache_dir=cache_dir)
---> 30         self.tokenizer = AutoTokenizer.from_pretrained(tokenizer_name_or_path if tokenizer_name_or_path is not None else model_name_or_path, cache_dir=cache_dir, **tokenizer_args)
     31 
     32         #No max_seq_length set. Try to infer from model

~\AppData\Local\Programs\Python\Python39\lib\site-packages\transformers\models\auto\tokenization_auto.py in from_pretrained(cls, pretrained_model_name_or_path, *inputs, **kwargs)
    566             tokenizer_class_py, tokenizer_class_fast = TOKENIZER_MAPPING[type(config)]
    567             if tokenizer_class_fast and (use_fast or tokenizer_class_py is None):
--> 568                 return tokenizer_class_fast.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
    569             else:
    570                 if tokenizer_class_py is not None:

~\AppData\Local\Programs\Python\Python39\lib\site-packages\transformers\tokenization_utils_base.py in from_pretrained(cls, pretrained_model_name_or_path, *init_inputs, **kwargs)
   1730                 logger.info(f"loading file {file_path} from cache at {resolved_vocab_files[file_id]}")
   1731 
-> 1732         return cls._from_pretrained(
   1733             resolved_vocab_files, pretrained_model_name_or_path, init_configuration, *init_inputs, **kwargs
   1734         )

~\AppData\Local\Programs\Python\Python39\lib\site-packages\transformers\tokenization_utils_base.py in _from_pretrained(cls, resolved_vocab_files, pretrained_model_name_or_path, init_configuration, *init_inputs, **kwargs)
   1848         # Instantiate tokenizer.
   1849         try:
-> 1850             tokenizer = cls(*init_inputs, **init_kwargs)
   1851         except OSError:
   1852             raise OSError(

~\AppData\Local\Programs\Python\Python39\lib\site-packages\transformers\models\xlm_roberta\tokenization_xlm_roberta_fast.py in __init__(self, vocab_file, tokenizer_file, bos_token, eos_token, sep_token, cls_token, unk_token, pad_token, mask_token, **kwargs)
    132         mask_token = AddedToken(mask_token, lstrip=True, rstrip=False) if isinstance(mask_token, str) else mask_token
    133 
--> 134         super().__init__(
    135             vocab_file,
    136             tokenizer_file=tokenizer_file,

~\AppData\Local\Programs\Python\Python39\lib\site-packages\transformers\tokenization_utils_fast.py in __init__(self, *args, **kwargs)
    105         elif fast_tokenizer_file is not None and not from_slow:
    106             # We have a serialization from tokenizers which let us directly build the backend
--> 107             fast_tokenizer = TokenizerFast.from_file(fast_tokenizer_file)
    108         elif slow_tokenizer is not None:
    109             # We need to convert a slow tokenizer to build the backend

Exception: data did not match any variant of untagged enum PyPreTokenizerTypeWrapper at line 1 column 317584

为了提供更多详细信息，我还遇到了另一个问题，仅在这台计算机上使用另一个型号：

model = SentenceTransformer('etalab-ia/dpr-question_encoder-fr_qa-camembert')

ValueError: unable to parse C:\Users\david.rouyre/.cache\torch\sentence_transformers\etalab-ia_dpr-question_encoder-fr_qa-camembert\tokenizer_config.json as a URL or as a local path

所以我检查了缓存路径，没有tokenizer_config.json，只有tokenizer.json（通过重命名它工作的文件）

包：（colab中的相同版本）

Name: sentence-transformers
Version: 2.0.0
Summary: Sentence Embeddings using BERT / RoBERTa / XLM-R
Home-page: https://github.com/UKPLab/sentence-transformers
Author: Nils Reimers
Author-email: info@nils-reimers.de
License: Apache License 2.0
Location: c:\users\david.rouyre\appdata\local\programs\python\python39\lib\site-packages
Requires: transformers, tqdm, torch, torchvision, numpy, scikit-learn, scipy, nltk, sentencepiece, huggingface-hub
Required-by:

我尝试清除缓存并使用 pip 卸载所有依赖项（transformers、tqdm、torch、torchvision、numpy、scikit-learn、scipy、nltk、sentencepiece、huggingface-hub），卸载 sentence-transformers 并重新安装它。

Answer 1

您安装了哪个标记器版本？对我来说，它有助于升级标记器：

pip3 install tokenizers==0.10.3