language:
- es
license: cc-by-4.0
tags:
- national library of spain
- spanish
- bne
datasets:
- bne
metrics:
- ppl
widget:
- text: Este año las campanadas de La Sexta las presentará <mask>.
- text: David Broncano es un presentador de La <mask>.
- text: >-
Gracias a los datos de la BNE se ha podido <mask> este modelo del
lenguaje.
- text: Hay base legal dentro del marco <mask> actual.
RoBERTa base trained with data from National Library of Spain (BNE)
Model Description
RoBERTa-base-bne is a transformer-based masked language model for the Spanish language. It is based on the RoBERTa base model and has been pre-trained using the largest Spanish corpus known to date, with a total of 570GB of clean and deduplicated text processed for this work, compiled from the web crawlings performed by the National Library of Spain from 2009 to 2019.
Training corpora and preprocessing
We cleaned 59TB of WARC files and we deduplicated them at computing node level. This resulted into 2TB of Spanish clean corpus. After that, we performed a global deduplication resulting into 570GB of text.
Some of the statistics of the corpus:
| Corpora | Number of documents | Number of tokens | Size (GB) |
|---|---|---|---|
| BNE | 201,080,084 | 135,733,450,668 | 570GB |
Tokenization and pre-training
We trained a BBPE tokenizer with a size of 50,262 tokens. We used 10,000 documents for validation and we trained the model for 48 hours into 16 computing nodes with 4 Nvidia V100 GPUs per node.
Evaluation and results
For evaluation details visit our GitHub repository.
Citing
Check out our paper for all the details: https://arxiv.org/abs/2107.07253
@misc{gutierrezfandino2021spanish,
title={Spanish Language Models},
author={Asier Gutiérrez-Fandiño and Jordi Armengol-Estapé and Marc Pàmies and Joan Llop-Palao and Joaquín Silveira-Ocampo and Casimiro Pio Carrino and Aitor Gonzalez-Agirre and Carme Armentano-Oller and Carlos Rodriguez-Penagos and Marta Villegas},
year={2021},
eprint={2107.07253},
archivePrefix={arXiv},
primaryClass={cs.CL}
}