opus-mt-tc-big-gmq-zlw

Table of Contents

Model Details

Neural machine translation model for translating from North Germanic languages (gmq) to West Slavic languages (zlw).

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. >>ces<<

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 = [
    ">>ces<< Normalt er jeg hjemme hele weekenden.",
    ">>pol<< Lev ditt liv."
]

model_name = "pytorch-models/opus-mt-tc-big-gmq-zlw"
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:
#     Vฤ›tลกinou jsem doma celรฝ vรญkend.
#     ลปyj swoim ลผyciem.

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-zlw")
print(pipe(">>ces<< Normalt er jeg hjemme hele weekenden."))

# expected output: Vฤ›tลกinou jsem doma celรฝ vรญkend.

Training

Evaluation

langpair testset chr-F BLEU #sent #words
swe-pol tatoeba-test-v2021-08-07 0.66326 46.2 1392 8157
dan-ces flores101-devtest 0.54065 26.7 1012 22101
dan-pol flores101-devtest 0.48389 18.8 1012 22520
isl-ces flores101-devtest 0.43582 17.7 1012 22101
isl-pol flores101-devtest 0.41929 13.9 1012 22520
nob-ces flores101-devtest 0.50336 22.3 1012 22101
nob-pol flores101-devtest 0.46130 16.3 1012 22520
swe-ces flores101-devtest 0.53188 25.7 1012 22101
swe-pol flores101-devtest 0.48163 18.6 1012 22520

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 00:02:29 EEST 2022
  • port machine: LM0-400-22516.local
Downloads last month
9
Safetensors
Model size
236M params
Tensor type
FP16
ยท
Inference Examples
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-zlw 7

Evaluation results