This model is part of the OPUS-MT project, an effort to make neural machine translation models widely available and accessible for many languages in the world. All models are originally trained using the amazing framework of Marian NMT, an efficient NMT implementation written in pure C++. The models have been converted to pyTorch using the transformers library by huggingface. Training data is taken from OPUS and training pipelines use the procedures of OPUS-MT-train. Model Description:

Developed by: Language Technology Research Group at the University of Helsinki
Model Type: Translation (transformer-big)
Release: 2024-05-30
License: Apache-2.0
Language(s):
- Source Language(s): lav lit prg sgs
- Target Language(s): deu eng fra por spa
- Valid Target Language Labels: >>deu<< >>eng<< >>fra<< >>por<< >>spa<< >>xxx<<
Original Model: opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-30.zip
Resources for more information:

This is a multilingual translation model with multiple target languages. A sentence initial language token is required in the form of >>id<< (id = valid target language ID), e.g. >>deu<<

Uses

This model can be used for translation and text-to-text generation.

Risks, Limitations and Biases

CONTENT WARNING: Readers should be aware that the model is trained on various public data sets that may contain content that is disturbing, offensive, and can propagate historical and current stereotypes.

Significant research has explored bias and fairness issues with language models (see, e.g., Sheng et al. (2021) and Bender et al. (2021)).

How to Get Started With the Model

A short example code:

python

from transformers import MarianMTModel, MarianTokenizer

src_text = [
    ">>deu<< Replace this with text in an accepted source language.",
    ">>spa<< This is the second sentence."
]

model_name = "pytorch-models/opus-mt-tc-bible-big-bat-deu_eng_fra_por_spa"
tokenizer = MarianTokenizer.from_pretrained(model_name)
model = MarianMTModel.from_pretrained(model_name)
translated = model.generate(**tokenizer(src_text, return_tensors="pt", padding=True))

for t in translated:
    print( tokenizer.decode(t, skip_special_tokens=True) )

You can also use OPUS-MT models with the transformers pipelines, for example:

python

from transformers import pipeline
pipe = pipeline("translation", model="Helsinki-NLP/opus-mt-tc-bible-big-bat-deu_eng_fra_por_spa")
print(pipe(">>deu<< Replace this with text in an accepted source language."))

Training

Data: opusTCv20230926max50+bt+jhubc (source)
Pre-processing: SentencePiece (spm32k,spm32k)
Model Type: transformer-big
Original MarianNMT Model: opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-30.zip
Training Scripts: GitHub Repo

Evaluation

Model scores at the OPUS-MT dashboard
test set translations: opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-29.test.txt
test set scores: opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-29.eval.txt
benchmark results: benchmark_results.txt
benchmark output: benchmark_translations.zip

langpair	testset	chr-F	BLEU	#sent	#words
lav-eng	tatoeba-test-v2021-08-07	0.63015	21.5	1631	11213
lit-deu	tatoeba-test-v2021-08-07	0.66527	47.5	1115	8531
lit-eng	tatoeba-test-v2021-08-07	0.72975	58.9	2528	17855
lit-spa	tatoeba-test-v2021-08-07	0.67956	49.9	454	2751
lav-deu	flores101-devtest	0.54001	23.8	1012	25094
lav-fra	flores101-devtest	0.57002	29.4	1012	28343
lav-por	flores101-devtest	0.55155	26.7	1012	26519
lav-spa	flores101-devtest	0.49259	20.8	1012	29199
lit-eng	flores101-devtest	0.59073	32.1	1012	24721
lit-por	flores101-devtest	0.55106	27.8	1012	26519
lit-deu	flores200-devtest	0.53223	23.7	1012	25094
lit-eng	flores200-devtest	0.59361	32.6	1012	24721
lit-fra	flores200-devtest	0.56786	30.0	1012	28343
lit-por	flores200-devtest	0.55393	28.2	1012	26519
lit-spa	flores200-devtest	0.49041	20.9	1012	29199
lav-eng	newstest2017	0.49729	22.0	2001	47511
lit-eng	newstest2019	0.59971	31.2	1000	25878
lav-deu	ntrex128	0.47317	18.5	1997	48761
lav-eng	ntrex128	0.53734	19.7	1997	47673
lav-fra	ntrex128	0.47843	19.6	1997	53481
lav-por	ntrex128	0.47027	19.3	1997	51631
lav-spa	ntrex128	0.49428	22.7	1997	54107
lit-deu	ntrex128	0.50279	19.4	1997	48761
lit-eng	ntrex128	0.56642	28.1	1997	47673
lit-fra	ntrex128	0.51276	22.6	1997	53481
lit-por	ntrex128	0.50864	22.6	1997	51631
lit-spa	ntrex128	0.53105	25.9	1997	54107

Citation Information

Publications: Democratizing neural machine translation with OPUS-MT and OPUS-MT – Building open translation services for the World and The Tatoeba Translation Challenge – Realistic Data Sets for Low Resource and Multilingual MT (Please, cite if you use this model.)

bibtex

@article{tiedemann2023democratizing,
  title={Democratizing neural machine translation with {OPUS-MT}},
  author={Tiedemann, J{\"o}rg and Aulamo, Mikko and Bakshandaeva, Daria and Boggia, Michele and Gr{\"o}nroos, Stig-Arne and Nieminen, Tommi and Raganato, Alessandro and Scherrer, Yves and Vazquez, Raul and Virpioja, Sami},
  journal={Language Resources and Evaluation},
  number={58},
  pages={713--755},
  year={2023},
  publisher={Springer Nature},
  issn={1574-0218},
  doi={10.1007/s10579-023-09704-w}
}

@inproceedings{tiedemann-thottingal-2020-opus,
    title = "{OPUS}-{MT} {--} Building open translation services for the World",
    author = {Tiedemann, J{\"o}rg  and Thottingal, Santhosh},
    booktitle = "Proceedings of the 22nd Annual Conference of the European Association for Machine Translation",
    month = nov,
    year = "2020",
    address = "Lisboa, Portugal",
    publisher = "European Association for Machine Translation",
    url = "https://aclanthology.org/2020.eamt-1.61",
    pages = "479--480",
}

@inproceedings{tiedemann-2020-tatoeba,
    title = "The Tatoeba Translation Challenge {--} Realistic Data Sets for Low Resource and Multilingual {MT}",
    author = {Tiedemann, J{\"o}rg},
    booktitle = "Proceedings of the Fifth Conference on Machine Translation",
    month = nov,
    year = "2020",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2020.wmt-1.139",
    pages = "1174--1182",
}

Acknowledgements

The work is supported by the HPLT project, funded by the European Union’s Horizon Europe research and innovation programme under grant agreement No 101070350. We are also grateful for the generous computational resources and IT infrastructure provided by CSC -- IT Center for Science, Finland, and the EuroHPC supercomputer LUMI.

Model conversion info

transformers version: 4.45.1
OPUS-MT git hash: a0ea3b3
port time: Mon Oct 7 17:27:51 EEST 2024
port machine: LM0-400-22516.local

WindstormLabs/origin-Helsinki-NLP-opus-mt-tc-bible-big-bat-deu_eng_fra_por_spa

opus-mt-tc-bible-big-bat-deu_eng_fra_por_spa

Table of Contents

Model Details