神经机器翻译模型,用于将芬兰语(fi)翻译为乌克兰语(uk)。
该模型是 OPUS-MT project 的一部分,旨在使神经机器翻译模型广泛可用和适用于世界上许多语言。所有模型最初是使用令人惊叹的 Marian NMT 框架进行训练的,它是一个用纯C++编写的高效NMT实现。这些模型已经使用huggingface的transformers库转换为pyTorch。训练数据来自 OPUS ,训练流程使用 OPUS-MT-train 的过程。
@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",
}
简短的示例代码:
from transformers import MarianMTModel, MarianTokenizer
src_text = [
"Afrikka on ihmiskunnan kehto.",
"Yksi, kaksi, kolme, neljä, viisi, kuusi, seitsemän, kahdeksan, yhdeksän, kymmenen."
]
model_name = "pytorch-models/opus-mt-tc-base-fi-uk"
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:
# Африка є колискою людства.
# Один, два, три, чотири, п'ять, шість, сім, вісім, дев'ять, десять.
您还可以使用transformers的pipelines中的OPUS-MT模型,例如:
from transformers import pipeline
pipe = pipeline("translation", model="Helsinki-NLP/opus-mt-tc-base-fi-uk")
print(pipe("Afrikka on ihmiskunnan kehto."))
# expected output: Африка є колискою людства.
| langpair | testset | chr-F | BLEU | #sent | #words |
|---|---|---|---|---|---|
| fin-ukr | flores101-devtest | 0.49562 | 19.7 | 1012 | 22810 |
这项工作得到 European Language Grid 的支持,作为 pilot project 2866 ,由 FoTran project 资助,该资助项目由欧洲研究委员会 (ERC) 在欧盟的Horizon 2020研究和创新项目 (授予协议编号 771113) 和 MeMAD project 资助,该项目在欧盟的Horizon 2020研究和创新计划下供给资助,授予协议编号 780069。我们还感谢 CSC -- IT Center for Science 提供的慷慨计算资源和IT基础设施,芬兰。