CAMeLBERT-MSA POS-MSA 模型

模型描述

CAMeLBERT-MSA POS-MSA 模型是一个现代标准阿拉伯语（MSA）词性标注模型，是通过对 CAMeLBERT-MSA 模型进行微调而构建的。在微调过程中，我们使用了 PATB 数据集。有关我们使用的微调过程和超参数，请参阅我们的论文 " The Interplay of Variant, Size, and Task Type in Arabic Pre-trained Language Models "。我们的微调代码可以在 here 找到。

预期用途

您可以将 CAMeLBERT-MSA POS-MSA 模型作为 transformers 流水线的一部分使用。该模型将很快在 CAMeL Tools 中提供。

要在 transformers 流水线中使用该模型：

>>> from transformers import pipeline
>>> pos = pipeline('token-classification', model='CAMeL-Lab/bert-base-arabic-camelbert-msa-pos-msa')
>>> text = 'إمارة أبوظبي هي إحدى إمارات دولة الإمارات العربية المتحدة السبع'
>>> pos(text)
[{'entity': 'noun', 'score': 0.9999764, 'index': 1, 'word': 'إمارة', 'start': 0, 'end': 5}, {'entity': 'noun_prop', 'score': 0.99991846, 'index': 2, 'word': 'أبوظبي', 'start': 6, 'end': 12}, {'entity': 'pron', 'score': 0.9998356, 'index': 3, 'word': 'هي', 'start': 13, 'end': 15}, {'entity': 'noun', 'score': 0.99368894, 'index': 4, 'word': 'إحدى', 'start': 16, 'end': 20}, {'entity': 'noun', 'score': 0.9999426, 'index': 5, 'word': 'إما', 'start': 21, 'end': 24}, {'entity': 'noun', 'score': 0.9999339, 'index': 6, 'word': '##رات', 'start': 24, 'end': 27}, {'entity': 'noun', 'score': 0.99996775, 'index': 7, 'word': 'دولة', 'start': 28, 'end': 32}, {'entity': 'noun', 'score': 0.99996895, 'index': 8, 'word': 'الإمارات', 'start': 33, 'end': 41}, {'entity': 'adj', 'score': 0.99990183, 'index': 9, 'word': 'العربية', 'start': 42, 'end': 49}, {'entity': 'adj', 'score': 0.9999347, 'index': 10, 'word': 'المتحدة', 'start': 50, 'end': 57}, {'entity': 'noun_num', 'score': 0.99931145, 'index': 11, 'word': 'السبع', 'start': 58, 'end': 63}]

注意：要下载我们的模型，您需要 transformers>=3.5.0。否则，您可以手动下载模型。

引用

@inproceedings{inoue-etal-2021-interplay,
    title = "The Interplay of Variant, Size, and Task Type in {A}rabic Pre-trained Language Models",
    author = "Inoue, Go  and
      Alhafni, Bashar  and
      Baimukan, Nurpeiis  and
      Bouamor, Houda  and
      Habash, Nizar",
    booktitle = "Proceedings of the Sixth Arabic Natural Language Processing Workshop",
    month = apr,
    year = "2021",
    address = "Kyiv, Ukraine (Online)",
    publisher = "Association for Computational Linguistics",
    abstract = "In this paper, we explore the effects of language variants, data sizes, and fine-tuning task types in Arabic pre-trained language models. To do so, we build three pre-trained language models across three variants of Arabic: Modern Standard Arabic (MSA), dialectal Arabic, and classical Arabic, in addition to a fourth language model which is pre-trained on a mix of the three. We also examine the importance of pre-training data size by building additional models that are pre-trained on a scaled-down set of the MSA variant. We compare our different models to each other, as well as to eight publicly available models by fine-tuning them on five NLP tasks spanning 12 datasets. Our results suggest that the variant proximity of pre-training data to fine-tuning data is more important than the pre-training data size. We exploit this insight in defining an optimized system selection model for the studied tasks.",
}