Skip to content

Latest commit

 

History

History
 
 

eng-cel

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
README.md
README.md
 
 
opus-2020-06-28.yml
opus-2020-06-28.yml
 
 
opus-2020-07-26.yml
opus-2020-07-26.yml
 
 
opus2m-2020-08-01.yml
opus2m-2020-08-01.yml
 
 

README.md

opus-2020-06-28.zip

  • dataset: opus
  • model: transformer
  • source language(s): eng
  • target language(s): bre cor cym gla gle glv
  • 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: opus-2020-06-28.zip
  • test set translations: opus-2020-06-28.test.txt
  • test set scores: opus-2020-06-28.eval.txt

Benchmarks

testset BLEU chr-F
Tatoeba-test.eng-bre.eng.bre 10.8 0.326
Tatoeba-test.eng-cor.eng.cor 0.2 0.092
Tatoeba-test.eng-cym.eng.cym 29.7 0.536
Tatoeba-test.eng-gla.eng.gla 5.6 0.283
Tatoeba-test.eng-gle.eng.gle 34.5 0.574
Tatoeba-test.eng-glv.eng.glv 7.2 0.326
Tatoeba-test.eng.multi 15.8 0.324

opus-2020-07-26.zip

  • dataset: opus
  • model: transformer
  • source language(s): eng
  • target language(s): bre cor cym gla gle glv
  • 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: opus-2020-07-26.zip
  • test set translations: opus-2020-07-26.test.txt
  • test set scores: opus-2020-07-26.eval.txt

Benchmarks

testset BLEU chr-F
Tatoeba-test.eng-bre.eng.bre 10.4 0.334
Tatoeba-test.eng-cor.eng.cor 0.2 0.094
Tatoeba-test.eng-cym.eng.cym 29.4 0.539
Tatoeba-test.eng-gla.eng.gla 6.6 0.295
Tatoeba-test.eng-gle.eng.gle 35.0 0.578
Tatoeba-test.eng-glv.eng.glv 7.1 0.333
Tatoeba-test.eng.multi 16.5 0.329

opus2m-2020-08-01.zip

  • dataset: opus2m
  • model: transformer
  • source language(s): eng
  • target language(s): bre cor cym gla gle glv
  • 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: 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-bre.eng.bre 11.5 0.338
Tatoeba-test.eng-cor.eng.cor 0.3 0.095
Tatoeba-test.eng-cym.eng.cym 31.0 0.549
Tatoeba-test.eng-gla.eng.gla 7.6 0.317
Tatoeba-test.eng-gle.eng.gle 35.9 0.582
Tatoeba-test.eng-glv.eng.glv 9.9 0.454
Tatoeba-test.eng.multi 18.0 0.342