metadata

language:
  - en
  - hu
tags:
  - translation
  - opus-mt-tc
license: cc-by-4.0
model-index:
  - name: opus-mt-tc-big-hu-en
    results:
      - task:
          name: Translation hun-eng
          type: translation
          args: hun-eng
        dataset:
          name: flores101-devtest
          type: flores_101
          args: hun eng devtest
        metrics:
          - name: BLEU
            type: bleu
            value: 34.6
      - task:
          name: Translation hun-eng
          type: translation
          args: hun-eng
        dataset:
          name: tatoeba-test-v2021-08-07
          type: tatoeba_mt
          args: hun-eng
        metrics:
          - name: BLEU
            type: bleu
            value: 50.4
      - task:
          name: Translation hun-eng
          type: translation
          args: hun-eng
        dataset:
          name: newstest2009
          type: wmt-2009-news
          args: hun-eng
        metrics:
          - name: BLEU
            type: bleu
            value: 23.4

opus-mt-tc-big-hu-en

Neural machine translation model for translating from Hungarian (hu) 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: 2022-03-09
source language(s): hun
target language(s): eng
model: transformer-big
data: opusTCv20210807+bt (source)
tokenization: SentencePiece (spm32k,spm32k)
original model: opusTCv20210807+bt_transformer-big_2022-03-09.zip
more information released models: OPUS-MT hun-eng README

Usage

A short example code:

from transformers import MarianMTModel, MarianTokenizer

src_text = [
    "Bárcsak ne láttam volna ilyen borzalmas filmet!",
    "Iskolában van."
]

model_name = "pytorch-models/opus-mt-tc-big-hu-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:
#     I wish I hadn't seen such a terrible movie.
#     She's at school.

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-hu-en")
print(pipe("Bárcsak ne láttam volna ilyen borzalmas filmet!"))

# expected output: I wish I hadn't seen such a terrible movie.

Benchmarks

test set translations: opusTCv20210807+bt_transformer-big_2022-03-09.test.txt
test set scores: opusTCv20210807+bt_transformer-big_2022-03-09.eval.txt
benchmark results: benchmark_results.txt
benchmark output: benchmark_translations.zip

langpair	testset	chr-F	BLEU	#sent	#words
hun-eng	tatoeba-test-v2021-08-07	0.66644	50.4	13037	94699
hun-eng	flores101-devtest	0.61974	34.6	1012	24721
hun-eng	newssyscomb2009	0.52563	24.7	502	11818
hun-eng	newstest2009	0.51698	23.4	2525	65399

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 19:33:38 EEST 2022
port machine: LM0-400-22516.local