liaad/srl-pt_mbert-base

This model is the bert-base-multilingual-cased model fine-tuned on Portuguese semantic role labeling data . The model was trained on the PropBank.Br datasets, using 10-fold Cross-Validation . The 10 resulting models were tested on the folds as well as on a smaller opinion dataset "Buscapé" For more information, please see the accompanying article (See BibTeX entry and citation info below) and the project's GitHub page . The full SRL model (transformers portion + a decoding layer) can be used to train the transformers portion of this model . To use this model, refer to the model's training procedure or the full model .,