XLMR-base-MaCoCu-is / README.md
RVN's picture
Update README.md
9a60a43
metadata
license: cc0-1.0
language:
  - is
tags:
  - MaCoCu

Model description

XLMR-base-MaCoCu-is is a large pre-trained language model trained on Icelandic texts. It was created by continuing training from the XLM-RoBERTa-base model. It was developed as part of the MaCoCu project and only uses data that was crawled during the project. The main developer is Jaume Zaragoza-Bernabeu from Prompsit Language Engineering.

XLMR-base-MaCoCu-is was trained on 4.4GB of Icelandic text, which is equal to 688M tokens. It was trained for 40,000 steps with a batch size of 256. It uses the same vocabulary as the original XLMR-base model.

The training and fine-tuning procedures are described in detail on our Github repo.

Warning

This model has not been fully trained because it was intended for use as base of Bicleaner AI Icelandic model. If you need better performance, please use XLMR-MaCoCu-is.

How to use

from transformers import AutoTokenizer, AutoModel, TFAutoModel

tokenizer = AutoTokenizer.from_pretrained("MaCoCu/XLMR-base-MaCoCu-is")
model = AutoModel.from_pretrained("MaCoCu/XLMR-base-MaCoCu-is") # PyTorch
model = TFAutoModel.from_pretrained("MaCoCu/XLMR-base-MaCoCu-is") # Tensorflow

Data

For training, we used all Icelandic data that was present in the monolingual Icelandic MaCoCu corpus. After de-duplicating the data, we were left with a total of 4.4 GB of text, which equals 688M tokens.

Acknowledgements

Research supported with Cloud TPUs from Google's TPU Research Cloud (TRC). The authors received funding from the European Union’s Connecting Europe Facility 2014- 2020 - CEF Telecom, under Grant Agreement No.INEA/CEF/ICT/A2020/2278341 (MaCoCu).

Citation

If you use this model, please cite the following paper:

@inproceedings{non-etal-2022-macocu,
    title = "{M}a{C}o{C}u: Massive collection and curation of monolingual and bilingual data: focus on under-resourced languages",
    author = "Ba{\~n}{\'o}n, Marta  and
      Espl{\`a}-Gomis, Miquel  and
      Forcada, Mikel L.  and
      Garc{\'\i}a-Romero, Cristian  and
      Kuzman, Taja  and
      Ljube{\v{s}}i{\'c}, Nikola  and
      van Noord, Rik  and
      Sempere, Leopoldo Pla  and
      Ram{\'\i}rez-S{\'a}nchez, Gema  and
      Rupnik, Peter  and
      Suchomel, V{\'\i}t  and
      Toral, Antonio  and
      van der Werff, Tobias  and
      Zaragoza, Jaume",
    booktitle = "Proceedings of the 23rd Annual Conference of the European Association for Machine Translation",
    month = jun,
    year = "2022",
    address = "Ghent, Belgium",
    publisher = "European Association for Machine Translation",
    url = "https://aclanthology.org/2022.eamt-1.41",
    pages = "303--304"
}