đ JPN - POL Translation Model
This project focuses on Japanese - Polish translation, offering a practical solution for cross - language communication.
đ Quick Start
The JPN - POL model is designed for Japanese to Polish translation. You can download the original weights and test set translations from the provided links below to start using the model.
⨠Features
- Language Support: Supports translation from multiple Japanese variants (jpn, jpn_Bopo, jpn_Hani, jpn_Hira, jpn_Kana, jpn_Latn) to Polish.
- Pre - processing: Utilizes normalization and SentencePiece (spm32k, spm32k) for pre - processing.
đĻ Installation
There is no specific installation command provided in the original document. If you want to use this model, you can download the original weights from the following link:
[opus - 2020 - 06 - 17.zip](https://object.pouta.csc.fi/Tatoeba - MT - models/jpn - pol/opus - 2020 - 06 - 17.zip)
đ Documentation
Model Details
- Source Group: Japanese
- Target Group: Polish
- OPUS Readme: [jpn - pol](https://github.com/Helsinki - NLP/Tatoeba - Challenge/tree/master/models/jpn - pol/README.md)
- Model: transformer - align
- Source Languages: jpn, jpn_Bopo, jpn_Hani, jpn_Hira, jpn_Kana, jpn_Latn
- Target Languages: pol
- Pre - processing: normalization + SentencePiece (spm32k, spm32k)
Download Links
- Original Weights: [opus - 2020 - 06 - 17.zip](https://object.pouta.csc.fi/Tatoeba - MT - models/jpn - pol/opus - 2020 - 06 - 17.zip)
- Test Set Translations: [opus - 2020 - 06 - 17.test.txt](https://object.pouta.csc.fi/Tatoeba - MT - models/jpn - pol/opus - 2020 - 06 - 17.test.txt)
- Test Set Scores: [opus - 2020 - 06 - 17.eval.txt](https://object.pouta.csc.fi/Tatoeba - MT - models/jpn - pol/opus - 2020 - 06 - 17.eval.txt)
Benchmarks
testset |
BLEU |
chr - F |
Tatoeba - test.jpn.pol |
15.7 |
0.386 |
System Info
Property |
Details |
hf_name |
jpn - pol |
source_languages |
jpn |
target_languages |
pol |
opus_readme_url |
https://github.com/Helsinki - NLP/Tatoeba - Challenge/tree/master/models/jpn - pol/README.md |
original_repo |
Tatoeba - Challenge |
tags |
['translation'] |
languages |
['ja', 'pl'] |
src_constituents |
{'jpn_Hang', 'jpn', 'jpn_Yiii', 'jpn_Kana', 'jpn_Hani', 'jpn_Bopo', 'jpn_Latn', 'jpn_Hira'} |
tgt_constituents |
{'pol'} |
src_multilingual |
False |
tgt_multilingual |
False |
prepro |
normalization + SentencePiece (spm32k, spm32k) |
url_model |
https://object.pouta.csc.fi/Tatoeba - MT - models/jpn - pol/opus - 2020 - 06 - 17.zip |
url_test_set |
https://object.pouta.csc.fi/Tatoeba - MT - models/jpn - pol/opus - 2020 - 06 - 17.test.txt |
src_alpha3 |
jpn |
tgt_alpha3 |
pol |
short_pair |
ja - pl |
chrF2_score |
0.386 |
bleu |
15.7 |
brevity_penalty |
1.0 |
ref_len |
69904.0 |
src_name |
Japanese |
tgt_name |
Polish |
train_date |
2020 - 06 - 17 |
src_alpha2 |
ja |
tgt_alpha2 |
pl |
prefer_old |
False |
long_pair |
jpn - pol |
helsinki_git_sha |
480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535 |
transformers_git_sha |
2207e5d8cb224e954a7cba69fa4ac2309e9ff30b |
port_machine |
brutasse |
port_time |
2020 - 08 - 21 - 14:41 |
đ License
This project is licensed under the Apache - 2.0 license.