medalpaca

/

medalpaca-13b

like 80

Using the example: qa_pipeline = pipeline("question-answering", model="medalpaca/medalpaca-13b", tokenizer="medalpaca/medalpaca-13b")

I get the error:

RecursionError Traceback (most recent call last)
Cell In[4], line 1
----> 1 qa_pipeline = pipeline("question-answering", model="medalpaca/medalpaca-13b", tokenizer="medalpaca/medalpaca-13b")

File ~/hf/lib/python3.10/site-packages/transformers/pipelines/init.py:885, in pipeline(task, model, config, tokenizer, feature_extractor, image_processor, framework, revision, use_fast, use_auth_token, device, device_map, torch_dtype, trust_remote_code, model_kwargs, pipeline_class, **kwargs)
882 tokenizer_kwargs = model_kwargs.copy()
883 tokenizer_kwargs.pop("torch_dtype", None)
--> 885 tokenizer = AutoTokenizer.from_pretrained(
886 tokenizer_identifier, use_fast=use_fast, _from_pipeline=task, **hub_kwargs, **tokenizer_kwargs
887 )
889 if load_image_processor:
890 # Try to infer image processor from model or config name (if provided as str)
891 if image_processor is None:

File ~/hf/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:694, in AutoTokenizer.from_pretrained(cls, pretrained_model_name_or_path, *inputs, **kwargs)
690 if tokenizer_class is None:
691 raise ValueError(
692 f"Tokenizer class {tokenizer_class_candidate} does not exist or is not currently imported."
693 )
--> 694 return tokenizer_class.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
696 # Otherwise we have to be creative.
697 # if model is an encoder decoder, the encoder tokenizer class is used by default
698 if isinstance(config, EncoderDecoderConfig):

File ~/hf/lib/python3.10/site-packages/transformers/tokenization_utils_base.py:1820, in PreTrainedTokenizerBase.from_pretrained(cls, pretrained_model_name_or_path, *init_inputs, **kwargs)
1817 else:
1818 logger.info(f"loading file {file_path} from cache at {resolved_vocab_files[file_id]}")
-> 1820 return cls._from_pretrained(
1821 resolved_vocab_files,
1822 pretrained_model_name_or_path,
1823 init_configuration,
1824 *init_inputs,
1825 use_auth_token=use_auth_token,
1826 cache_dir=cache_dir,
1827 local_files_only=local_files_only,
1828 _commit_hash=commit_hash,
1829 _is_local=is_local,
1830 **kwargs,
1831 )

File ~/hf/lib/python3.10/site-packages/transformers/tokenization_utils_base.py:1983, in PreTrainedTokenizerBase._from_pretrained(cls, resolved_vocab_files, pretrained_model_name_or_path, init_configuration, use_auth_token, cache_dir, local_files_only, _commit_hash, _is_local, *init_inputs, **kwargs)
1981 # Instantiate tokenizer.
1982 try:
-> 1983 tokenizer = cls(*init_inputs, **init_kwargs)
1984 except OSError:
1985 raise OSError(
1986 "Unable to load vocabulary from file. "
1987 "Please check that the provided vocabulary is accessible and not corrupted."
1988 )

File ~/hf/lib/python3.10/site-packages/transformers/models/llama/tokenization_llama_fast.py:104, in LlamaTokenizerFast.init(self, vocab_file, tokenizer_file, clean_up_tokenization_spaces, unk_token, bos_token, eos_token, add_bos_token, add_eos_token, **kwargs)
102 self._add_bos_token = add_bos_token
103 self._add_eos_token = add_eos_token
--> 104 self.update_post_processor()
106 self.vocab_file = vocab_file
107 self.can_save_slow_tokenizer = False if not self.vocab_file else True

File ~/hf/lib/python3.10/site-packages/transformers/models/llama/tokenization_llama_fast.py:111, in LlamaTokenizerFast.update_post_processor(self)
109 def update_post_processor(self):
110 bos = self.bos_token
--> 111 bos_token_id = self.bos_token_id
113 eos = self.eos_token
114 eos_token_id = self.eos_token_id

File ~/hf/lib/python3.10/site-packages/transformers/tokenization_utils_base.py:1131, in SpecialTokensMixin.bos_token_id(self)
1129 if self._bos_token is None:
1130 return None
-> 1131 return self.convert_tokens_to_ids(self.bos_token)

File ~/hf/lib/python3.10/site-packages/transformers/tokenization_utils_fast.py:250, in PreTrainedTokenizerFast.convert_tokens_to_ids(self, tokens)
247 return None
249 if isinstance(tokens, str):
--> 250 return self._convert_token_to_id_with_added_voc(tokens)
252 return [self._convert_token_to_id_with_added_voc(token) for token in tokens]

File ~/hf/lib/python3.10/site-packages/transformers/tokenization_utils_fast.py:257, in PreTrainedTokenizerFast._convert_token_to_id_with_added_voc(self, token)
255 index = self._tokenizer.token_to_id(token)
256 if index is None:
--> 257 return self.unk_token_id
258 return index

File ~/hf/lib/python3.10/site-packages/transformers/tokenization_utils_base.py:1150, in SpecialTokensMixin.unk_token_id(self)
1148 if self._unk_token is None:
1149 return None
-> 1150 return self.convert_tokens_to_ids(self.unk_token)

File ~/hf/lib/python3.10/site-packages/transformers/tokenization_utils_fast.py:250, in PreTrainedTokenizerFast.convert_tokens_to_ids(self, tokens)
247 return None
249 if isinstance(tokens, str):
--> 250 return self._convert_token_to_id_with_added_voc(tokens)
252 return [self._convert_token_to_id_with_added_voc(token) for token in tokens]

File ~/hf/lib/python3.10/site-packages/transformers/tokenization_utils_fast.py:257, in PreTrainedTokenizerFast._convert_token_to_id_with_added_voc(self, token)
255 index = self._tokenizer.token_to_id(token)
256 if index is None:
--> 257 return self.unk_token_id
258 return index

File ~/hf/lib/python3.10/site-packages/transformers/tokenization_utils_base.py:1150, in SpecialTokensMixin.unk_token_id(self)
1148 if self._unk_token is None:
1149 return None
-> 1150 return self.convert_tokens_to_ids(self.unk_token)

[... skipping similar frames: PreTrainedTokenizerFast._convert_token_to_id_with_added_voc at line 257 (985 times), PreTrainedTokenizerFast.convert_tokens_to_ids at line 250 (985 times), SpecialTokensMixin.unk_token_id at line 1150 (985 times)]

File ~/hf/lib/python3.10/site-packages/transformers/tokenization_utils_fast.py:250, in PreTrainedTokenizerFast.convert_tokens_to_ids(self, tokens)
247 return None
249 if isinstance(tokens, str):
--> 250 return self._convert_token_to_id_with_added_voc(tokens)
252 return [self._convert_token_to_id_with_added_voc(token) for token in tokens]

File ~/hf/lib/python3.10/site-packages/transformers/tokenization_utils_fast.py:257, in PreTrainedTokenizerFast._convert_token_to_id_with_added_voc(self, token)
255 index = self._tokenizer.token_to_id(token)
256 if index is None:
--> 257 return self.unk_token_id
258 return index

File ~/hf/lib/python3.10/site-packages/transformers/tokenization_utils_base.py:1150, in SpecialTokensMixin.unk_token_id(self)
1148 if self._unk_token is None:
1149 return None
-> 1150 return self.convert_tokens_to_ids(self.unk_token)

File ~/hf/lib/python3.10/site-packages/transformers/tokenization_utils_base.py:1030, in SpecialTokensMixin.unk_token(self)
1028 logger.error("Using unk_token, but it is not set yet.")
1029 return None
-> 1030 return str(self._unk_token)

RecursionError: maximum recursion depth exceeded while calling a Python object