神经机器翻译模型,用于将英语(en)翻译成匈牙利语(hu)。
此模型是 OPUS-MT project 的一部分,旨在使神经机器翻译模型在世界上许多语言中广泛可用和可访问。所有模型最初都是使用 Marian NMT 的出色框架进行训练的, Marian NMT 是纯C++编写的高效NMT实现。模型已使用huggingface的transformers库将其转换为pyTorch。训练数据来自 OPUS ,训练流程使用 OPUS-MT-train 的程序。
@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",
}
简短示例代码:
from transformers import MarianMTModel, MarianTokenizer
src_text = [
"I wish I hadn't seen such a horrible film.",
"She's at school."
]
model_name = "pytorch-models/opus-mt-tc-big-en-hu"
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:
# Bárcsak ne láttam volna ilyen szörnyű filmet.
# Iskolában van.
您还可以使用transformers pipelines来使用OPUS-MT模型,例如:
from transformers import pipeline
pipe = pipeline("translation", model="Helsinki-NLP/opus-mt-tc-big-en-hu")
print(pipe("I wish I hadn't seen such a horrible film."))
# expected output: Bárcsak ne láttam volna ilyen szörnyű filmet.
| langpair | testset | chr-F | BLEU | #sent | #words |
|---|---|---|---|---|---|
| eng-hun | tatoeba-test-v2021-08-07 | 0.62096 | 38.7 | 13037 | 79562 |
| eng-hun | flores101-devtest | 0.60159 | 29.6 | 1012 | 22183 |
| eng-hun | newssyscomb2009 | 0.51918 | 20.6 | 502 | 9733 |
| eng-hun | newstest2009 | 0.50973 | 20.3 | 2525 | 54965 |
该工作得到 European Language Grid 和 pilot project 2866 的支持,由 FoTran project 资助,由欧洲研究理事会(ERC)在欧洲联盟的Horizon 2020研究和创新计划(授予协议号771113)下资助,以及由 MeMAD project 以欧洲联盟的Horizon 2020研究和创新计划的资助协议号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。