Views
No views yet
1>>> from transformers import pipeline
2>>> did = pipeline('text-classification', model='CAMeL-Lab/bert-base-arabic-camelbert-msa-did-nadi')
3>>> sentences = ['عامل ايه ؟', 'شلونك ؟ شخبارك ؟']
4>>> did(sentences)
5[{'label': 'Egypt', 'score': 0.9242768287658691},
6{'label': 'Saudi_Arabia', 'score': 0.3400847613811493}]transformers>=3.5.0.
Otherwise, you could download the models manually.1@inproceedings{inoue-etal-2021-interplay,
2 title = "The Interplay of Variant, Size, and Task Type in {A}rabic Pre-trained Language Models",
3 author = "Inoue, Go and
4 Alhafni, Bashar and
5 Baimukan, Nurpeiis and
6 Bouamor, Houda and
7 Habash, Nizar",
8 booktitle = "Proceedings of the Sixth Arabic Natural Language Processing Workshop",
9 month = apr,
10 year = "2021",
11 address = "Kyiv, Ukraine (Online)",
12 publisher = "Association for Computational Linguistics",
13 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.",
14}