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.

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",
}

Model info

Release: 2022-03-07
source language(s): rus ukr
target language(s): fin
model: transformer-big
data: opusTCv20210807+bt (source)
tokenization: SentencePiece (spm32k,spm32k)
original model: opusTCv20210807+bt_transformer-big_2022-03-07.zip
more information released models: OPUS-MT zle-fin README

Usage

A short example code:

python

from transformers import MarianMTModel, MarianTokenizer

src_text = [
    "Мы уже проголосовали.",
    "Один, два, три, чотири, п'ять, шість, сім, вісім, дев'ять, десять."
]

model_name = "pytorch-models/opus-mt-tc-big-zle-fi"
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:
#     Olemme jo äänestäneet.
#     Yksi, kaksi, kolme, neljä, viisi, kuusi, seitsemän, kahdeksan, yhdeksän, kymmenen.

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-zle-fi")
print(pipe("Мы уже проголосовали."))

# expected output: Olemme jo äänestäneet.

Benchmarks

test set translations: opusTCv20210807+bt_transformer-big_2022-03-07.test.txt
test set scores: opusTCv20210807+bt_transformer-big_2022-03-07.eval.txt
benchmark results: benchmark_results.txt
benchmark output: benchmark_translations.zip

langpair	testset	chr-F	BLEU	#sent	#words
rus-fin	tatoeba-test-v2021-08-07	0.66334	42.2	3643	19319
rus-fin	flores101-devtest	0.52577	17.4	1012	18781
ukr-fin	flores101-devtest	0.53440	18.0	1012	18781

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: 42126b6
port time: Thu Mar 24 09:28:52 EET 2022
port machine: LM0-400-22516.local