- dataset: opusTCv20210807
- model: transformer-big
- source language(s): bel bel_Latn rus ukr
- target language(s): pob por
- raw source language(s): bel rus ukr
- raw target language(s): pob por
- model: transformer-big
- 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-big_2022-03-14.zip
- test set translations: opusTCv20210807_transformer-big_2022-03-14.test.txt
- test set scores: opusTCv20210807_transformer-big_2022-03-14.eval.txt
| testset | BLEU | chr-F | #sent | #words | BP |
|---|---|---|---|---|---|
| Tatoeba-test-v2021-08-07.bel-por | 88.4 | 0.89898 | 3 | 21 | 1.000 |
| Tatoeba-test-v2021-08-07.multi-multi | 43.3 | 0.64145 | 10000 | 71519 | 0.971 |
| Tatoeba-test-v2021-08-07.rus-por | 41.8 | 0.63281 | 10000 | 74705 | 0.965 |
| Tatoeba-test-v2021-08-07.ukr-por | 43.9 | 0.64404 | 3372 | 21301 | 0.986 |
- dataset: opusTCv20210807
- model: transformer-big
- source language(s): bel bel_Latn rus ukr
- target language(s): pob por
- raw source language(s): bel rus ukr
- raw target language(s): pob por
- model: transformer-big
- 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-big_2022-03-19.zip
- test set translations: opusTCv20210807_transformer-big_2022-03-19.test.txt
- test set scores: opusTCv20210807_transformer-big_2022-03-19.eval.txt
| testset | BLEU | chr-F | #sent | #words | BP |
|---|---|---|---|---|---|
| Tatoeba-test-v2021-08-07.bel-por | 65.6 | 0.74422 | 3 | 21 | 1.000 |
| Tatoeba-test-v2021-08-07.multi-multi | 43.0 | 0.63923 | 10000 | 71519 | 0.968 |
| Tatoeba-test-v2021-08-07.rus-por | 42.1 | 0.63462 | 10000 | 74705 | 0.961 |
| Tatoeba-test-v2021-08-07.ukr-por | 44.4 | 0.64658 | 3372 | 21301 | 0.988 |
- dataset: opusTCv20210807
- model: transformer-big
- source language(s): bel bel_Latn rus ukr
- target language(s): pob por
- raw source language(s): bel rus ukr
- raw target language(s): pob por
- model: transformer-big
- 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-big_2022-03-23.zip
- test set translations: opusTCv20210807_transformer-big_2022-03-23.test.txt
- test set scores: opusTCv20210807_transformer-big_2022-03-23.eval.txt
| testset | BLEU | chr-F | #sent | #words | BP |
|---|---|---|---|---|---|
| Tatoeba-test-v2021-08-07.bel-por | 65.6 | 0.74422 | 3 | 21 | 1.000 |
| Tatoeba-test-v2021-08-07.multi-multi | 43.1 | 0.63939 | 10000 | 71519 | 0.965 |
| Tatoeba-test-v2021-08-07.rus-por | 42.3 | 0.63500 | 10000 | 74705 | 0.960 |
| Tatoeba-test-v2021-08-07.ukr-por | 44.6 | 0.65006 | 3372 | 21301 | 0.986 |