eng-pqe

  • source group: English

  • target group: Eastern Malayo-Polynesian languages

  • OPUS readme: eng-pqe

  • model: transformer

  • source language(s): eng

  • target language(s): fij gil haw lkt mah mri nau niu rap smo tah ton tvl

  • model: transformer

  • pre-processing: normalization + SentencePiece (spm32k,spm32k)

  • a sentence initial language token is required in the form of >>id<< (id = valid target language ID)

  • download original weights: opus2m-2020-08-01.zip

  • test set translations: opus2m-2020-08-01.test.txt

  • test set scores: opus2m-2020-08-01.eval.txt

Benchmarks

testset BLEU chr-F
Tatoeba-test.eng-fij.eng.fij 22.1 0.396
Tatoeba-test.eng-gil.eng.gil 41.9 0.673
Tatoeba-test.eng-haw.eng.haw 0.6 0.114
Tatoeba-test.eng-lkt.eng.lkt 0.5 0.075
Tatoeba-test.eng-mah.eng.mah 9.7 0.386
Tatoeba-test.eng-mri.eng.mri 7.7 0.301
Tatoeba-test.eng.multi 11.3 0.306
Tatoeba-test.eng-nau.eng.nau 0.5 0.071
Tatoeba-test.eng-niu.eng.niu 42.5 0.560
Tatoeba-test.eng-rap.eng.rap 3.3 0.122
Tatoeba-test.eng-smo.eng.smo 27.0 0.462
Tatoeba-test.eng-tah.eng.tah 11.3 0.307
Tatoeba-test.eng-ton.eng.ton 27.0 0.528
Tatoeba-test.eng-tvl.eng.tvl 29.3 0.513

System Info:

Downloads last month
28
Hosted inference API
Translation
This model can be loaded on the Inference API on-demand.