--- tags: - adapterhub:dialect/arabic - text-classification - adapter-transformers - bert license: "apache-2.0" --- # Adapter `bert-base-arabert-ar-dialect` for aubmindlab/bert-base-arabert **Note: This adapter was not trained by the AdapterHub team, but by these author(s): Nick Doiron. See author details below.** Adapter for AraBERT (aubmindlab/bert-base-arabert) trained to classify Arabic by dialect {0=Egyptian, 1=Gulf, 2=Levantine, 3=Maghrebi, 4=MSA} Trained for 3 epochs on 85k samples (+ 28k test set) from University of British Columbia and John Hopkins University. **This adapter was created for usage with the [Adapters](https://github.com/Adapter-Hub/adapters) library.** ## Usage First, install `adapters`: ``` pip install -U adapters ``` Now, the adapter can be loaded and activated like this: ```python from adapters import AutoAdapterModel model = AutoAdapterModel.from_pretrained("aubmindlab/bert-base-arabert") adapter_name = model.load_adapter("AdapterHub/bert-base-arabert-ar-dialect") model.set_active_adapters(adapter_name) ``` ## Architecture & Training - Adapter architecture: pfeiffer - Prediction head: classification - Dataset: [ArabicDialect](https://adapterhub.ml/explore/dialect/arabic/) ## Author Information - Author name(s): Nick Doiron - Author email: ndoiron@mapmeld.com - Author links: [Website](https://github.com/MonsoonNLP/sanaa-dialect), [GitHub](https://github.com/mapmeld), [Twitter](https://twitter.com/@mapmeld) ## Versions - `1` - `2` **(main)** ## Citation ```bibtex ``` *This adapter has been auto-imported from https://github.com/Adapter-Hub/Hub/blob/master/adapters/mapmeld/bert-base-arabert-ar-dialect.yaml*.