- dataset: opus1m+bt
- model: transformer-align
- source language(s): hye
- target language(s): eng
- model: transformer-align
- pre-processing: normalization + SentencePiece (spm32k,spm32k)
- download: opus1m+bt-2021-05-01.zip
- test set translations: opus1m+bt-2021-05-01.test.txt
- test set scores: opus1m+bt-2021-05-01.eval.txt
| testset | BLEU | chr-F | #sent | #words | BP |
|---|---|---|---|---|---|
| Tatoeba-test.hye-eng | 39.0 | 0.557 | 1121 | 5066 | 1.000 |
| Tatoeba-test.multi-eng | 39.2 | 0.558 | 1121 | 5066 | 1.000 |
- dataset: opus4m+btTCv20210807
- model: transformer
- source language(s): hye hyw xcl
- target language(s): eng
- model: transformer
- pre-processing: normalization + SentencePiece (spm32k,spm32k)
- download: opus4m+btTCv20210807-2021-09-30.zip
- test set translations: opus4m+btTCv20210807-2021-09-30.test.txt
- test set scores: opus4m+btTCv20210807-2021-09-30.eval.txt
| testset | BLEU | chr-F | #sent | #words | BP |
|---|---|---|---|---|---|
| Tatoeba-test-v2021-08-07.multi-eng | 35.3 | 0.527 | 1121 | 5066 | 1.000 |
| Tatoeba-test-v2021-08-07.multi-multi | 35.3 | 0.527 | 1121 | 5066 | 1.000 |