Model: BERT
Lang: IT
Model description
This is a BERT [1] model for the Italian language, obtained using mBERT (bert-base-multilingual-cased) as a starting point and focusing it on the Italian language by modifying the embedding layer (as in [2], computing document-level frequencies over the Wikipedia dataset)
The resulting model has 110M parameters, a vocabulary of 30.785 tokens, and a size of ~430 MB.
Quick usage
from transformers import BertTokenizerFast, BertModel
tokenizer = BertTokenizerFast.from_pretrained("osiria/bert-base-italian-cased")
model = BertModel.from_pretrained("osiria/bert-base-italian-cased")
References
[1] https://arxiv.org/abs/1810.04805
[2] https://arxiv.org/abs/2010.05609
License
The model is released under Apache-2.0 license