Edit model card

opus-mt-tc-big-es-zle

Neural machine translation model for translating from Spanish (es) to East Slavic languages (zle).

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.

@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",
}

Model info

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. >>bel<<

Usage

A short example code:

from transformers import MarianMTModel, MarianTokenizer

src_text = [
    ">>rus<< Su novela se vendió bien.",
    ">>ukr<< Quiero ir a Corea del Norte."
]

model_name = "pytorch-models/opus-mt-tc-big-es-zle"
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:
#     Его роман хорошо продавался.
#     Я хочу поїхати до Північної Кореї.

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

from transformers import pipeline
pipe = pipeline("translation", model="Helsinki-NLP/opus-mt-tc-big-es-zle")
print(pipe(">>rus<< Su novela se vendió bien."))

# expected output: Его роман хорошо продавался.

Benchmarks

langpair testset chr-F BLEU #sent #words
spa-bel tatoeba-test-v2021-08-07 0.54506 27.5 205 1259
spa-rus tatoeba-test-v2021-08-07 0.68523 49.0 10506 69242
spa-ukr tatoeba-test-v2021-08-07 0.63502 42.3 10115 54544
spa-rus flores101-devtest 0.49913 20.2 1012 23295
spa-ukr flores101-devtest 0.47772 17.4 1012 22810
spa-rus newstest2012 0.52436 24.6 3003 64790
spa-rus newstest2013 0.54249 26.9 3000 58560

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: 1bdabf7
  • port time: Thu Mar 24 03:35:13 EET 2022
  • port machine: LM0-400-22516.local
Downloads last month
1
Hosted inference API
This model can be loaded on the Inference API on-demand.

Evaluation results