mirror of
https://github.com/Helsinki-NLP/OPUS-MT-train.git
synced 2024-11-30 12:32:24 +03:00
| .. | ||
| README.md | ||
opus-2019-12-04.zip
- dataset: opus
- model: transformer
- pre-processing: normalization + tokenization + BPE
- download: opus-2019-12-04.zip
- test set translations: opus-2019-12-04.test.txt
- test set scores: opus-2019-12-04.eval.txt
Benchmarks
| testset | BLEU | chr-F |
|---|---|---|
| newstest2014-csen.cs.en | 31.5 | 0.589 |
| newstest2015-encs.cs.en | 27.5 | 0.540 |
| newstest2016-encs.cs.en | 28.5 | 0.561 |
| newstest2017-encs.cs.en | 26.6 | 0.540 |
| newstest2018-encs.cs.en | 27.1 | 0.540 |
| Tatoeba.cs.en | 62.5 | 0.743 |
opus-2019-12-18.zip
- dataset: opus
- model: transformer-align
- pre-processing: normalization + SentencePiece
- download: opus-2019-12-18.zip
- test set translations: opus-2019-12-18.test.txt
- test set scores: opus-2019-12-18.eval.txt
Benchmarks
| testset | BLEU | chr-F |
|---|---|---|
| newstest2014-csen.cs.en | 34.1 | 0.612 |
| newstest2015-encs.cs.en | 30.4 | 0.565 |
| newstest2016-encs.cs.en | 31.8 | 0.584 |
| newstest2017-encs.cs.en | 28.7 | 0.556 |
| newstest2018-encs.cs.en | 30.3 | 0.566 |
| Tatoeba.cs.en | 58.0 | 0.721 |