- dataset: opus
- model: transformer
- source language(s): eus
- target language(s): eng
- model: transformer
- pre-processing: normalization + SentencePiece (spm12k,spm12k)
- download: opus-2020-08-12.zip
- test set translations: opus-2020-08-12.test.txt
- test set scores: opus-2020-08-12.eval.txt
testset | BLEU | chr-F |
---|---|---|
Tatoeba-test.eus.eng | 41.5 | 0.594 |
Tatoeba-test.eus-eng.eus.eng | 41.5 | 0.594 |