mirror of
https://github.com/Helsinki-NLP/OPUS-MT-train.git
synced 2024-11-30 22:14:14 +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 |
|---|---|---|
| Bible.en.ml | 50.4 | 0.598 |
opus+bt-2020-03-02.zip
- dataset: opus+bt
- model: transformer-align
- pre-processing: normalization + SentencePiece
- download: opus+bt-2020-03-02.zip
- test set translations: opus+bt-2020-03-02.test.txt
- test set scores: opus+bt-2020-03-02.eval.txt
Benchmarks
| testset | BLEU | chr-F |
|---|---|---|
| Tatoeba.en.ml | 17.0 | 0.507 |
opus-2020-04-20.zip
- dataset: opus
- model: transformer-align
- pre-processing: normalization + SentencePiece
- download: opus-2020-04-20.zip
- test set translations: opus-2020-04-20.test.txt
- test set scores: opus-2020-04-20.eval.txt
Benchmarks
| testset | BLEU | chr-F |
|---|---|---|
| Tatoeba.en.ml | 18.3 | 0.531 |
opus+bt+bt-2020-04-28.zip
- dataset: opus+bt+bt
- model: transformer-align
- pre-processing: normalization + SentencePiece
- download: opus+bt+bt-2020-04-28.zip
- test set translations: opus+bt+bt-2020-04-28.test.txt
- test set scores: opus+bt+bt-2020-04-28.eval.txt
Benchmarks
| testset | BLEU | chr-F |
|---|---|---|
| Tatoeba.en.ml | 19.1 | 0.536 |