Helsinki-NLP_opus-mt-tc-big-tr-en

---

language:

• en
• tr tags:
• translation
• opus-mt-tc license: cc-by-4.0 model-index:
• name: opus-mt-tc-big-tr-en results:
  • task: name: Translation tur-eng type: translation args: tur-eng dataset: name: flores101-devtest type: flores_101 args: tur eng devtest metrics:
    • name: BLEU type: bleu value: 37.6
  • task: name: Translation tur-eng type: translation args: tur-eng dataset: name: newsdev2016 type: newsdev2016 args: tur-eng metrics:
    • name: BLEU type: bleu value: 32.1
  • task: name: Translation tur-eng type: translation args: tur-eng dataset: name: tatoeba-test-v2021-08-07 type: tatoeba_mt args: tur-eng metrics:
    • name: BLEU type: bleu value: 57.6
  • task: name: Translation tur-eng type: translation args: tur-eng dataset: name: newstest2016 type: wmt-2016-news args: tur-eng metrics:
    • name: BLEU type: bleu value: 29.3
  • task: name: Translation tur-eng type: translation args: tur-eng dataset: name: newstest2017 type: wmt-2017-news args: tur-eng metrics:
    • name: BLEU type: bleu value: 29.7
  • task: name: Translation tur-eng type: translation args: tur-eng dataset: name: newstest2018 type: wmt-2018-news args: tur-eng metrics:
    • name: BLEU type: bleu value: 30.7

---

opus-mt-tc-big-tr-en

Neural machine translation model for translating from Turkish (tr) to English (en).

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: 2022-03-17
• source language(s): tur
• target language(s): eng
• model: transformer-big
• data: opusTCv20210807+bt (source)
• tokenization: SentencePiece (spm32k,spm32k)
• original model: opusTCv20210807+bt_transformer-big_2022-03-17.zip
• more information released models: OPUS-MT tur-eng README

Usage

A short example code:

from transformers import MarianMTModel, MarianTokenizer

src_text = [
    "Allahsızlığı Yayma Kürsüsü başkanıydı.",
    "Tom'a ne olduğunu öğrenin."
]

model_name = "pytorch-models/opus-mt-tc-big-tr-en"
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:
#     He was the president of the Curse of Spreading Godlessness.
#     Find out what happened to Tom.

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-tr-en")
print(pipe("Allahsızlığı Yayma Kürsüsü başkanıydı."))

# expected output: He was the president of the Curse of Spreading Godlessness.

Benchmarks

• test set translations: opusTCv20210807+bt_transformer-big_2022-03-17.test.txt
• test set scores: opusTCv20210807+bt_transformer-big_2022-03-17.eval.txt
• benchmark results: benchmark_results.txt
• benchmark output: benchmark_translations.zip

langpair	testset	chr-F	BLEU	#sent	#words
tur-eng	tatoeba-test-v2021-08-07	0.71895	57.6	13907	109231
tur-eng	flores101-devtest	0.64152	37.6	1012	24721
tur-eng	newsdev2016	0.58658	32.1	1001	21988
tur-eng	newstest2016	0.56960	29.3	3000	66175
tur-eng	newstest2017	0.57455	29.7	3007	67703
tur-eng	newstest2018	0.58488	30.7	3000	68725

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: 3405783
• port time: Wed Apr 13 20:02:48 EEST 2022
• port machine: LM0-400-22516.local