tiedeman's picture
Initial commit
416a057
|
raw
history blame
9.3 kB
metadata
language:
  - da
  - he
  - nb
  - sv
tags:
  - translation
  - opus-mt-tc
license: cc-by-4.0
model-index:
  - name: opus-mt-tc-big-he-gmq
    results:
      - task:
          name: Translation heb-dan
          type: translation
          args: heb-dan
        dataset:
          name: flores101-devtest
          type: flores_101
          args: heb dan devtest
        metrics:
          - name: BLEU
            type: bleu
            value: 31.4
          - name: chr-F
            type: chrf
            value: 0.58023
      - task:
          name: Translation heb-isl
          type: translation
          args: heb-isl
        dataset:
          name: flores101-devtest
          type: flores_101
          args: heb isl devtest
        metrics:
          - name: BLEU
            type: bleu
            value: 14
          - name: chr-F
            type: chrf
            value: 0.41998
      - task:
          name: Translation heb-nob
          type: translation
          args: heb-nob
        dataset:
          name: flores101-devtest
          type: flores_101
          args: heb nob devtest
        metrics:
          - name: BLEU
            type: bleu
            value: 23.7
          - name: chr-F
            type: chrf
            value: 0.53086
      - task:
          name: Translation heb-swe
          type: translation
          args: heb-swe
        dataset:
          name: flores101-devtest
          type: flores_101
          args: heb swe devtest
        metrics:
          - name: BLEU
            type: bleu
            value: 29.6
          - name: chr-F
            type: chrf
            value: 0.56881

opus-mt-tc-big-he-gmq

Table of Contents

Model Details

Neural machine translation model for translating from Hebrew (he) to North Germanic languages (gmq).

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:

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. >>dan<<

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 = [
    ">>dan<< ื›ืœ ืฉืœื•ืฉืช ื”ื™ืœื“ื™ื ืฉืœ ืืœื™ืขื–ืจ ืœื•ื“ื•ื•ื™ื’ ื–ืžื ื”ื•ืฃ ื ืจืฆื—ื• ื‘ืฉื•ืื”.",
    ">>swe<< ื”ืกืชื‘ืจ ืฉื˜ื•ื ื”ื™ื” ืžืจื’ืœ."
]

model_name = "pytorch-models/opus-mt-tc-big-he-gmq"
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:
#     Alle tre bรธrn af Eliezer Ludwig Zamenhof blev drรฆbt i Holocaust.
#     Det visade sig att Tom var en spion.

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-he-gmq")
print(pipe(">>dan<< ื›ืœ ืฉืœื•ืฉืช ื”ื™ืœื“ื™ื ืฉืœ ืืœื™ืขื–ืจ ืœื•ื“ื•ื•ื™ื’ ื–ืžื ื”ื•ืฃ ื ืจืฆื—ื• ื‘ืฉื•ืื”."))

# expected output: Alle tre bรธrn af Eliezer Ludwig Zamenhof blev drรฆbt i Holocaust.

Training

Evaluation

langpair testset chr-F BLEU #sent #words
heb-dan flores101-devtest 0.58023 31.4 1012 24638
heb-isl flores101-devtest 0.41998 14.0 1012 22834
heb-nob flores101-devtest 0.53086 23.7 1012 23873
heb-swe flores101-devtest 0.56881 29.6 1012 23121

Citation Information

@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 European Language Grid as pilot project 2866, by the FoTran project, funded by the European Research Council (ERC) under the European Unionโ€™s Horizon 2020 research and innovation programme (grant agreement No 771113), and the MeMAD project, funded by the European Unionโ€™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: 8b9f0b0
  • port time: Sat Aug 13 00:07:45 EEST 2022
  • port machine: LM0-400-22516.local