antoniocappiello/bert-base-italian-uncased-squad-it

The model is the uncased base version of the italian BERT (which you may find at dbmdz/bert-base-italian-uncased) trained on the question answering task . It has been trained using SQuAD-it, derived from the original dataset and obtained through the semi-automatic translation of the dataset in Italian . The training procedure is described as 'predict_file' and 'train-v1.1' by python . The results have been evaluated with the help of transformers and pipeline . The model was trained on a question-answering task using the original SQuad-it dataset . The data was derived from,