Multi dialect Arabic BERT

Multi-dialect-arabic-Bert

Este é um repositório do modelo Bert árabe multi-dialet.

Por mawdoo3-ai.

Sobre o nosso modelo multi-dialet-arabic-Bert

Em vez de treinar o modelo Bert árabe multi-dialetto a partir do zero, inicializamos os pesos do modelo usando o árabe-bert e o treinamos em tweets de 10 milhões de árabe a partir dos dados que não são da tarefa compartilhada de identificação de dialetos árabes (NADI).

Para citar este trabalho

 @misc{talafha2020multidialect,
    title={Multi-Dialect Arabic BERT for Country-Level Dialect Identification},
    author={Bashar Talafha and Mohammad Ali and Muhy Eddin Za'ter and Haitham Seelawi and Ibraheem Tuffaha and Mostafa Samir and Wael Farhan and Hussein T. Al-Natsheh},
    year={2020},
    eprint={2007.05612},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}

Uso

Os pesos do modelo podem ser carregados usando a biblioteca transformers pelo HuggingFace.

 from transformers import AutoTokenizer , AutoModel

tokenizer = AutoTokenizer . from_pretrained ( "bashar-talafha/multi-dialect-bert-base-arabic" )
model = AutoModel . from_pretrained ( "bashar-talafha/multi-dialect-bert-base-arabic" )

Exemplo usando pipeline :

 from transformers import pipeline

fill_mask = pipeline (
    "fill-mask" ,
    model = "bashar-talafha/multi-dialect-bert-base-arabic " ,
    tokenizer = "bashar-talafha/multi-dialect-bert-base-arabic "
)

fill_mask ( " سافر الرحالة من مطار [MASK] " )

 [{'sequence': '[CLS] سافر الرحالة من مطار الكويت [SEP]', 'score': 0.08296813815832138, 'token': 3226},
 {'sequence': '[CLS] سافر الرحالة من مطار دبي [SEP]', 'score': 0.05123933032155037, 'token': 4747},
 {'sequence': '[CLS] سافر الرحالة من مطار مسقط [SEP]', 'score': 0.046838656067848206, 'token': 13205},
 {'sequence': '[CLS] سافر الرحالة من مطار القاهرة [SEP]', 'score': 0.03234650194644928, 'token': 4003},
 {'sequence': '[CLS] سافر الرحالة من مطار الرياض [SEP]', 'score': 0.02606341242790222, 'token': 2200}]

Parâmetros do modelo

Parâmetro	Valor
arquitetura	Bertformaskedlm
hidden_size	768
max_position_embeddings	512
NUM_ATTENTENTEN_HEADS	12
num_hidden_layers	12
vocab_size	32000
hidden_size	768
Número total de parâmetros	110m