- dataset: opusTCv20210807
- model: transformer-align
- source language(s): rus ukr
- target language(s): lav lit ltg
- raw source language(s): rus ukr
- raw target language(s): lav lit ltg
- model: transformer-align
- pre-processing: normalization + SentencePiece (spm32k,spm32k)
- a sentence initial language token is required in the form of
>>id<<(id = valid target language ID) - valid language labels:
- download: opusTCv20210807_transformer-align_2022-03-14.zip
- test set translations: opusTCv20210807_transformer-align_2022-03-14.test.txt
- test set scores: opusTCv20210807_transformer-align_2022-03-14.eval.txt
| testset | BLEU | chr-F | #sent | #words | BP |
|---|---|---|---|---|---|
| Tatoeba-test-v2021-08-07.multi-multi | 47.8 | 0.70999 | 3876 | 22206 | 0.964 |
| Tatoeba-test-v2021-08-07.rus-lav | 55.7 | 0.74105 | 274 | 1516 | 0.989 |
| Tatoeba-test-v2021-08-07.rus-lit | 46.8 | 0.70382 | 3598 | 20605 | 0.960 |
| Tatoeba-test-v2021-08-07.ukr-lav | 2.0 | 0.13943 | 1 | 6 | 1.000 |
| Tatoeba-test-v2021-08-07.ukr-lit | 68.0 | 0.84108 | 3 | 20 | 1.000 |
- dataset: opusTCv20210807
- model: transformer-align
- source language(s): rus ukr
- target language(s): lav lit ltg
- raw source language(s): rus ukr
- raw target language(s): lav lit ltg
- model: transformer-align
- pre-processing: normalization + SentencePiece (spm32k,spm32k)
- a sentence initial language token is required in the form of
>>id<<(id = valid target language ID) - valid language labels:
- download: opusTCv20210807_transformer-align_2022-03-17.zip
- test set translations: opusTCv20210807_transformer-align_2022-03-17.test.txt
- test set scores: opusTCv20210807_transformer-align_2022-03-17.eval.txt
| testset | BLEU | chr-F | #sent | #words | BP |
|---|---|---|---|---|---|
| Tatoeba-test-v2021-08-07.multi-multi | 48.0 | 0.70964 | 3876 | 22206 | 0.963 |
| Tatoeba-test-v2021-08-07.rus-lav | 55.7 | 0.74105 | 274 | 1516 | 0.989 |
| Tatoeba-test-v2021-08-07.rus-lit | 46.8 | 0.70382 | 3598 | 20605 | 0.960 |
| Tatoeba-test-v2021-08-07.ukr-lav | 2.0 | 0.13943 | 1 | 6 | 1.000 |
| Tatoeba-test-v2021-08-07.ukr-lit | 68.0 | 0.84108 | 3 | 20 | 1.000 |