bashar-talafha/multi-dialect-bert-base-arabic

This is a repository of Multi-dialect Arabic BERT model . Instead of training the model from scratch, we trained it on 10M arabic tweets from the unlabled data of The Nuanced Arabic Dialect Identification (NADI) shared task . The model weights can be loaded using transformers library by HuggingFace . To cite this work, the author of the work can be found at http://misc.com/talafha2020multidialect.org/ . The work was published in September 2007 and is published in Springer Springer Publishing House, Springer Springer, Springer Publishing, Springer Publishers, Springer Random House, and Springer Publishing Europe, Springer Books, The Netherlands,