Hugging Face's logo

Helsinki-NLP
/
opus-mt-tc-big-gmq-gmq

Translation
Transformers
PyTorch
TensorFlow
Safetensors
marian
text2text-generation
opus-mt-tc
Eval Results
Inference Endpoints
Model card Files Community
3
Edit model card

opus-mt-tc-big-gmq-gmq

Table of Contents

Model Details

Neural machine translation model for translating from North Germanic languages (gmq) 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 = [
    ">>fao<< Jeg er bange for kakerlakker.",
    ">>nob<< Vladivostok är en stad i Ryssland."
]

model_name = "pytorch-models/opus-mt-tc-big-gmq-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:
#     Tað eru uml.
#     Vladivostok er en by i Russland.

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-gmq-gmq")
print(pipe(">>fao<< Jeg er bange for kakerlakker."))

# expected output: Tað eru uml.

Training

Evaluation

langpair testset chr-F BLEU #sent #words
dan-nob tatoeba-test-v2021-08-07 0.87556 78.2 1299 9620
dan-swe tatoeba-test-v2021-08-07 0.83556 72.5 1549 10060
nno-nob tatoeba-test-v2021-08-07 0.88349 78.9 467 3129
nob-dan tatoeba-test-v2021-08-07 0.85345 73.9 1299 9794
nob-nno tatoeba-test-v2021-08-07 0.74571 55.2 466 3141
nob-swe tatoeba-test-v2021-08-07 0.84747 73.9 563 3698
swe-dan tatoeba-test-v2021-08-07 0.83392 72.6 1549 10239
swe-nob tatoeba-test-v2021-08-07 0.85815 76.3 563 3708
isl-swe europeana2021 0.45562 22.2 563 10293
nob-isl europeana2021 0.54171 29.7 538 9932
nob-swe europeana2021 0.73891 54.0 538 9885
dan-isl flores101-devtest 0.50227 22.2 1012 22834
dan-nob flores101-devtest 0.58445 28.6 1012 23873
dan-swe flores101-devtest 0.65000 38.5 1012 23121
isl-dan flores101-devtest 0.53630 27.2 1012 24638
isl-nob flores101-devtest 0.49434 20.5 1012 23873
isl-swe flores101-devtest 0.53373 26.0 1012 23121
nob-dan flores101-devtest 0.59657 31.7 1012 24638
nob-isl flores101-devtest 0.47432 18.9 1012 22834
nob-swe flores101-devtest 0.60030 31.3 1012 23121
swe-dan flores101-devtest 0.64340 39.0 1012 24638
swe-isl flores101-devtest 0.49590 21.7 1012 22834
swe-nob flores101-devtest 0.58336 28.9 1012 23873

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 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: 8b9f0b0
  • port time: Fri Aug 12 23:59:02 EEST 2022
  • port machine: LM0-400-22516.local
Downloads last month
14
Safetensors
Model size
212M params
Tensor type
FP16
·
Inference Examples
Translation
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Spaces using Helsinki-NLP/opus-mt-tc-big-gmq-gmq 7

Evaluation results

View on Papers With Code