Hugging Face's logo

Helsinki-NLP
/
opus-mt-tc-bible-big-deu_eng_fra_por_spa-pqw

Translation
Transformers
PyTorch
Safetensors
marian
text2text-generation
opus-mt-tc-bible
Eval Results
Inference Endpoints
Model card Files Community
opus-mt-tc-bible-big-deu_eng_fra_por_spa-pqw / README.md
tiedeman's picture
tiedeman
Initial commit
d92c0ca
preview code
|
raw
history blame contribute delete
10.9 kB
metadata
library_name: transformers
language:
  - ace
  - akl
  - ban
  - bcl
  - bik
  - btd
  - bth
  - bto
  - bts
  - btx
  - bug
  - ceb
  - cgc
  - ch
  - de
  - dtp
  - en
  - es
  - fil
  - fr
  - gor
  - hil
  - iba
  - id
  - ify
  - ilo
  - jv
  - krj
  - ljp
  - mad
  - mak
  - mbb
  - mbt
  - mg
  - mhy
  - mog
  - mrw
  - ms
  - msm
  - mta
  - mwv
  - nia
  - nij
  - obo
  - pag
  - pam
  - pau
  - pt
  - rej
  - sas
  - sda
  - sml
  - su
  - sxn
  - tbl
  - war
tags:
  - translation
  - opus-mt-tc-bible
license: apache-2.0
model-index:
  - name: opus-mt-tc-bible-big-deu_eng_fra_por_spa-pqw
    results:
      - task:
          name: Translation multi-multi
          type: translation
          args: multi-multi
        dataset:
          name: tatoeba-test-v2020-07-28-v2023-09-26
          type: tatoeba_mt
          args: multi-multi
        metrics:
          - name: BLEU
            type: bleu
            value: 20.1
          - name: chr-F
            type: chrf
            value: 0.46165

opus-mt-tc-bible-big-deu_eng_fra_por_spa-pqw

Table of Contents

Model Details

Neural machine translation model for translating from unknown (deu+eng+fra+por+spa) to Western Malayo-Polynesian languages (pqw).

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: 2024-05-30
  • License: Apache-2.0
  • Language(s):
    • Source Language(s): deu eng fra por spa
    • Target Language(s): ace akl ban bcl bik btd bth bto bts btx bug ceb cgc cha dtp fil gor hil iba ify ilo ind jak jav krj ljp mad mak max mbb mbt mhy mlg mog mrw msa msm mta mwv nia nij obo pag pam pau plt rej sas sda sml sun sxn tbl tmw war zlm zsm
    • Valid Target Language Labels: >>abl<< >>abs<< >>abx<< >>ace<< >>agk<< >>agz<< >>akb<< >>akl<< >>akl_Latn<< >>atd<< >>atl<< >>ban<< >>bbc<< >>bcl<< >>bdg<< >>bdl<< >>bdr<< >>beg<< >>bew<< >>bgs<< >>bik<< >>bkd<< >>bkz<< >>blf<< >>bno<< >>bnq<< >>bsu<< >>btd<< >>bth<< >>btm<< >>bto<< >>bts<< >>btx<< >>btz<< >>buc<< >>bug<< >>ceb<< >>cgc<< >>cha<< >>cia<< >>cja<< >>cje<< >>cjm<< >>cps<< >>dbj<< >>drg<< >>dtp<< >>dtr<< >>dun<< >>duq<< >>duw<< >>eno<< >>fil<< >>gay<< >>goq<< >>gor<< >>hil<< >>hro<< >>huq<< >>iba<< >>ibg<< >>ibh<< >>ibl<< >>ify<< >>ilk<< >>ilo<< >>ind<< >>jak_Latn<< >>jav<< >>jav_Java<< >>jra<< >>kak<< >>kaw<< >>kge<< >>kjc<< >>kjk<< >>kqr<< >>krj<< >>ktq<< >>kyi<< >>kyj<< >>kyk<< >>kys<< >>lbw<< >>lbx<< >>ley<< >>ljp<< >>llk<< >>loc<< >>lra<< >>mad<< >>mak<< >>max_Latn<< >>mba<< >>mbb<< >>mbd<< >>mbi<< >>mbs<< >>mbt<< >>mdh<< >>mdr<< >>mhy<< >>mkm<< >>mkx<< >>mlg<< >>mog<< >>mqk<< >>mqn<< >>mrw<< >>msa<< >>msa_Arab<< >>msa_Latn<< >>msb<< >>msm<< >>mta<< >>mtd<< >>mwv<< >>mxr<< >>myl<< >>mzq<< >>nia<< >>nij<< >>nrm<< >>obo<< >>otd<< >>pag<< >>pam<< >>pau<< >>pdo<< >>pku<< >>plt<< >>rad<< >>raz<< >>ree<< >>rej<< >>rgs<< >>roc<< >>rog<< >>rth<< >>sas<< >>sda<< >>sdo<< >>sgd<< >>sjm<< >>skh<< >>slm<< >>sml<< >>sml_Latn<< >>smr<< >>smw<< >>sne<< >>snl<< >>snv<< >>ssb<< >>sse<< >>sun<< >>sxn<< >>sya<< >>tbl<< >>tdi<< >>tdn<< >>tjg<< >>tlk<< >>tmw_Latn<< >>tnt<< >>tnw<< >>tom<< >>twy<< >>txs<< >>ulu<< >>vkl<< >>vko<< >>war<< >>wow<< >>wru<< >>xkq<< >>xmz<< >>yka<< >>zbc<< >>zbe<< >>zbw<< >>zlm_Arab<< >>zlm_Latn<< >>zsm_Arab<< >>zsm_Latn<<
  • Original Model: opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-30.zip
  • Resources for more information:

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. >>ace<<

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:

from transformers import MarianMTModel, MarianTokenizer

src_text = [
    ">>ace<< Replace this with text in an accepted source language.",
    ">>zsm<< This is the second sentence."
]

model_name = "pytorch-models/opus-mt-tc-bible-big-deu_eng_fra_por_spa-pqw"
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) )

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-bible-big-deu_eng_fra_por_spa-pqw")
print(pipe(">>ace<< Replace this with text in an accepted source language."))

Training

Evaluation

langpair testset chr-F BLEU #sent #words
multi-multi tatoeba-test-v2020-07-28-v2023-09-26 0.46165 20.1 10000 66909

Citation Information 

@article{tiedemann2023democratizing,
  title={Democratizing neural machine translation with {OPUS-MT}},
  author={Tiedemann, J{\"o}rg and Aulamo, Mikko and Bakshandaeva, Daria and Boggia, Michele and Gr{\"o}nroos, Stig-Arne and Nieminen, Tommi and Raganato, Alessandro and Scherrer, Yves and Vazquez, Raul and Virpioja, Sami},
  journal={Language Resources and Evaluation},
  number={58},
  pages={713--755},
  year={2023},
  publisher={Springer Nature},
  issn={1574-0218},
  doi={10.1007/s10579-023-09704-w}
}

@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 HPLT project, funded by the European Union’s Horizon Europe research and innovation programme under grant agreement No 101070350. We are also grateful for the generous computational resources and IT infrastructure provided by CSC -- IT Center for Science, Finland, and the EuroHPC supercomputer LUMI.

Model conversion info

  • transformers version: 4.45.1
  • OPUS-MT git hash: 0882077
  • port time: Tue Oct 8 10:24:47 EEST 2024
  • port machine: LM0-400-22516.local