Модели загрузки трубопроводов и токенизаторы для вопросов и ответов

Question

Привет, я пытаюсь использовать 'fmikaelian / flaubert-base-uncased-squad' для ответа на вопрос. Я понимаю, что я должен загрузить модель и токенизаторы. Я не уверен, как мне это сделать.

Мой код в основном далеко

from transformers import pipeline, BertTokenizer

nlp = pipeline('question-answering', \
model='fmikaelian/flaubert-base-uncased-squad', \
tokenizer='fmikaelian/flaubert-base-uncased-squad')

Скорее всего, это можно решить с помощью двух строк.

Большое спасибо

РЕДАКТИРОВАТЬ

Я также пытался использовать автомодели, но, похоже, их там нет:

OSError: Model name 'flaubert-base-uncased-squad' was not found in model name list (bert-base-uncased, bert-large-uncased, bert-base-cased, bert-large-cased, bert-base-multilingual-uncased, bert-base-multilingual-cased, bert-base-chinese, bert-base-german-cased, bert-large-uncased-whole-word-masking, bert-large-cased-whole-word-masking, bert-large-uncased-whole-word-masking-finetuned-squad, bert-large-cased-whole-word-masking-finetuned-squad, bert-base-cased-finetuned-mrpc, bert-base-german-dbmdz-cased, bert-base-german-dbmdz-uncased). We assumed 'flaubert-base-uncased-squad' was a path or url to a configuration file named config.json or a directory containing such a file but couldn't find any such file at this path or url.

РЕДАКТИРОВАТЬ II Я попытался следовать предложенному подходу со следующим кодом, который загружает модели, которые были сохранены из S3:

tokenizer_ = FlaubertTokenizer.from_pretrained(MODELS)
model_ = FlaubertModel.from_pretrained(MODELS)


p = transformers.QuestionAnsweringPipeline(
    model=transformers.AutoModel.from_pretrained(MODELS), 
    tokenizer=transformers.AutoTokenizer.from_pretrained(MODELS)
)

question_="Quel est le montant de la garantie?"
language_="French"
context_="le montant de la garantie est € 1000"

output=p({'question':question_, 'context': context_})
print(output)

К сожалению, я получаю следующую ошибку:

Traceback (most recent call last):
  File "<string>", line 1, in <module>
  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\multiprocessing\spawn.py", line 105, in spawn_main
    exitcode = _main(fd)
Traceback (most recent call last):
  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\multiprocessing\spawn.py", line 114, in _main
  File "question_extraction.py", line 61, in <module>
        prepare(preparation_data)
output=p({'question':question_, 'context': context_})  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\multiprocessing\spawn.py", line 225, in prepare

      File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\site-packages\transformers\pipelines.py", line 802, in __call__
_fixup_main_from_path(data['init_main_from_path'])
  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\multiprocessing\spawn.py", line 277, in _fixup_main_from_path
    run_name="__mp_main__")
  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\runpy.py", line 263, in run_path
    pkg_name=pkg_name, script_name=fname)
  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\runpy.py", line 96, in _run_module_code
    mod_name, mod_spec, pkg_name, script_name)
  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "C:\Users\... ...\Box Sync\nlp - 2...\NLP\src\question_extraction.py", line 61, in <module>
    output=p({'question':question_, 'context': context_})
  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\site-packages\transformers\pipelines.py", line 802, in __call__
    for example in examples
      File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\site-packages\transformers\pipelines.py", line 802, in <listcomp>
for example in examples
for example in examples  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\site-packages\transformers\pipelines.py", line 802, in <listcomp>

      File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\site-packages\transformers\data\processors\squad.py", line 304, in squad_convert_examples_to_features
for example in examples
  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\site-packages\transformers\data\processors\squad.py", line 304, in squad_convert_examples_to_features
        with Pool(threads, initializer=squad_convert_example_to_features_init, initargs=(tokenizer,)) as p:with Pool(threads, initializer=squad_convert_example_to_features_init, initargs=(tokenizer,)) as p:

  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\multiprocessing\context.py", line 119, in Pool
  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\multiprocessing\context.py", line 119, in Pool
        context=self.get_context())context=self.get_context())

  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\multiprocessing\pool.py", line 174, in __init__
  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\multiprocessing\pool.py", line 174, in __init__
        self._repopulate_pool()self._repopulate_pool()

  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\multiprocessing\pool.py", line 239, in _repopulate_pool
  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\multiprocessing\pool.py", line 239, in _repopulate_pool
    w.start()
    w.start()
  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\multiprocessing\process.py", line 105, in start
  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\multiprocessing\process.py", line 105, in start
    self._popen = self._Popen(self)
      File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\multiprocessing\context.py", line 322, in _Popen
self._popen = self._Popen(self)
      File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\multiprocessing\context.py", line 322, in _Popen
return Popen(process_obj)
return Popen(process_obj)  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\multiprocessing\popen_spawn_win32.py", line 65, in __init__

  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\multiprocessing\popen_spawn_win32.py", line 33, in __init__
        prep_data = spawn.get_preparation_data(process_obj._name)reduction.dump(process_obj, to_child)

  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\multiprocessing\spawn.py", line 143, in get_preparation_data
  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\multiprocessing\reduction.py", line 60, in dump
    _check_not_importing_main()
  File "C:\Users\... ...\AppData\Local\Continuum\Anaconda3\envs\nlp_nlp\lib\multiprocessing\spawn.py", line 136, in _check_not_importing_main
    is not going to be frozen to produce an executable.''')
RuntimeError:
        An attempt has been made to start a new process before the
        current process has finished its bootstrapping phase.

        This probably means that you are not using fork to start your
        child processes and you have forgotten to use the proper idiom
        in the main module:

            if __name__ == '__main__':
                freeze_support()
                ...

        The "freeze_support()" line can be omitted if the program
        is not going to be frozen to produce an executable.
    ForkingPickler(file, protocol).dump(obj)
BrokenPipeError: [Errno 32] Broken pipe

* РЕДАКТИРОВАТЬ IV *

Я решил предыдущую ошибку РЕДАКТИРОВАНИЯ, поместив функции внутри " main ". К сожалению, когда я запускаю следующий код:

tokenizer_ = FlaubertTokenizer.from_pretrained(MODELS)
model_ = FlaubertModel.from_pretrained(MODELS)

def question_extraction(text, question, model, tokenizer, language="French", verbose=False):

    if language=="French":
        nlp = pipeline('question-answering', \
        model=model, \
        tokenizer=tokenizer)
    else:
        nlp=pipeline('question-answering')

    output=nlp({'question':question, 'context': text})

    answer, score = output.answer, output.score 

    if verbose==True:
        print("Q: ", question ,"\n",\
              "A:", answer,"\n", \
              "Confidence (%):", "{0:.2f}".format(str(score*100) )
              )

    return answer, score

if __name__=="__main__":
    question_="Quel est le montant de la garantie?"
    language_="French"
    text="le montant de la garantie est € 1000"

    answer, score=question_extraction(text, question_, model_, tokenizer_, language_, verbose= True)

я получаю следующую ошибку:

C:\...\NLP\src>python question_extraction.py
OK
OK
convert squad examples to features: 100%|████████████████████████████████████████████████| 1/1 [00:00<00:00,  4.66it/s]
add example index and unique id: 100%|███████████████████████████████████████████████████████████| 1/1 [00:00<?, ?it/s]
Traceback (most recent call last):
  File "question_extraction.py", line 77, in <module>
    answer, score=question_extraction(text, question_, model_, tokenizer_, language_, verbose= True)
  File "question_extraction.py", line 60, in question_extraction
    output=nlp({'question':question, 'context': text})
  File "C:\...\transformers\pipelines.py", line 818, in __call__
    start, end = self.model(**fw_args)
ValueError: not enough values to unpack (expected 2, got 1)

Answer 1

Как указано в источнике , существует спецификация c QuestionAnsweringPipeline. Ниже приведен пример того, что я использовал для успешной загрузки модели Флобера.

import transformers as trf
p = trf.QuestionAnsweringPipeline(model=trf.AutoModel.from_pretrained("fmikaelian/flaubert-base-uncased-squad"), tokenizer=trf.AutoTokenizer.from_pretrained("fmikaelian/flaubert-base-uncased-squad"))

Конечно, есть и альтернатива для использования предварительно обученной модели FlaubertForQuestionAnswering, поскольку pipeline s только что вышел с последней версией и может быть изменен.