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: 2022-08-09
License: CC-BY-4.0
Language(s):
- Source Language(s): cat fra glg ita lat_Latn por ron spa
- Target Language(s): ara
- Language Pair(s): cat-ara fra-ara glg-ara ita-ara por-ara ron-ara spa-ara
- Valid Target Language Labels: >>ajp<< >>apc<< >>ara<< >>arq<< >>ary<< >>arz<<
Original Model: opusTCv20210807_transformer-big_2022-08-09.zip
Resources for more information:
- OPUS-MT-train GitHub Repo
- More information about released models for this language pair: OPUS-MT itc-ara README
- More information about MarianNMT models in the transformers library
- [Tatoeba Translation Challenge](https://github.com/Helsinki-NLP/Tatoeba-Challenge/

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. >>ara<<

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 = [
    ">>ary<< Entiendo.",
    ">>arq<< Por favor entiende mi posición."
]

model_name = "pytorch-models/opus-mt-tc-big-itc-ar"
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) )

# expected output:
#     فهمتك
#     من فضلك افهم موقفي.

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-big-itc-ar")
print(pipe(">>ary<< Entiendo."))

# expected output: فهمتك

Training

Data: opusTCv20210807 (source)
Pre-processing: SentencePiece (spm32k,spm32k)
Model Type: transformer-big
Original MarianNMT Model: opusTCv20210807_transformer-big_2022-08-09.zip
Training Scripts: GitHub Repo

Evaluation

test set translations: opusTCv20210807_transformer-big_2022-08-09.test.txt
test set scores: opusTCv20210807_transformer-big_2022-08-09.eval.txt
benchmark results: benchmark_results.txt
benchmark output: benchmark_translations.zip

langpair	testset	chr-F	BLEU	#sent	#words
fra-ara	tatoeba-test-v2021-08-07	0.46463	18.9	1569	7956
ita-ara	tatoeba-test-v2021-08-07	0.53797	25.7	235	1161
spa-ara	tatoeba-test-v2021-08-07	0.55520	26.6	1511	7547
cat-ara	flores101-devtest	0.52029	18.9	1012	21357
fra-ara	flores101-devtest	0.52573	19.5	1012	21357
glg-ara	flores101-devtest	0.51181	19.2	1012	21357
ita-ara	flores101-devtest	0.49401	15.0	1012	21357
por-ara	flores101-devtest	0.53356	20.2	1012	21357
ron-ara	flores101-devtest	0.51849	18.4	1012	21357
spa-ara	flores101-devtest	0.47872	14.3	1012	21357

Citation Information

Publications: 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.)

@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 European Language Grid as pilot project 2866, by the FoTran project, funded by the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No 771113), and the MeMAD project, funded by the European Union’s Horizon 2020 Research and Innovation Programme under grant agreement No 780069. We are also grateful for the generous computational resources and IT infrastructure provided by CSC -- IT Center for Science, Finland.

Model conversion info

transformers version: 4.16.2
OPUS-MT git hash: 8b9f0b0
port time: Sat Aug 13 00:00:31 EEST 2022
port machine: LM0-400-22516.local

Helsinki-NLP/opus-mt-tc-big-itc-ar

opus-mt-tc-big-itc-ar

Table of Contents

Model Details