cel-eng

opus-2020-06-28.zip

• dataset: opus
• model: transformer
• source language(s): bre cor cym gla gle glv
• target language(s): eng
• model: transformer
• pre-processing: normalization + SentencePiece (spm32k,spm32k)
• 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.bre-eng.bre.eng	15.2	0.362
Tatoeba-test.cor-eng.cor.eng	3.1	0.175
Tatoeba-test.cym-eng.cym.eng	38.3	0.564
Tatoeba-test.gla-eng.gla.eng	14.7	0.318
Tatoeba-test.gle-eng.gle.eng	50.0	0.660
Tatoeba-test.glv-eng.glv.eng	30.0	0.466
Tatoeba-test.multi.eng	22.1	0.391

opus-2020-07-26.zip

• dataset: opus
• model: transformer
• source language(s): bre cor cym gla gle glv
• target language(s): eng
• model: transformer
• pre-processing: normalization + SentencePiece (spm32k,spm32k)
• 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.bre-eng.bre.eng	16.7	0.375
Tatoeba-test.cor-eng.cor.eng	2.7	0.175
Tatoeba-test.cym-eng.cym.eng	39.3	0.563
Tatoeba-test.gla-eng.gla.eng	13.7	0.312
Tatoeba-test.gle-eng.gle.eng	47.8	0.642
Tatoeba-test.glv-eng.glv.eng	45.2	0.614
Tatoeba-test.multi.eng	21.7	0.387

opus2m-2020-07-31.zip

• dataset: opus2m
• model: transformer
• source language(s): bre cor cym gla gle glv
• target language(s): eng
• model: transformer
• pre-processing: normalization + SentencePiece (spm32k,spm32k)
• download: opus2m-2020-07-31.zip
• test set translations: opus2m-2020-07-31.test.txt
• test set scores: opus2m-2020-07-31.eval.txt

Benchmarks

testset	BLEU	chr-F
Tatoeba-test.bre-eng.bre.eng	17.2	0.385
Tatoeba-test.cor-eng.cor.eng	3.0	0.172
Tatoeba-test.cym-eng.cym.eng	41.5	0.582
Tatoeba-test.gla-eng.gla.eng	15.4	0.330
Tatoeba-test.gle-eng.gle.eng	50.8	0.668
Tatoeba-test.glv-eng.glv.eng	11.0	0.297
Tatoeba-test.multi.eng	22.8	0.398

opus4m-2020-08-12.zip

• dataset: opus4m
• model: transformer
• source language(s): bre cor cym gla gle glv
• target language(s): eng
• model: transformer
• pre-processing: normalization + SentencePiece (spm32k,spm32k)
• download: opus4m-2020-08-12.zip
• test set translations: opus4m-2020-08-12.test.txt
• test set scores: opus4m-2020-08-12.eval.txt

Benchmarks

testset	BLEU	chr-F
Tatoeba-test.bre-eng.bre.eng	17.2	0.395
Tatoeba-test.cor-eng.cor.eng	3.2	0.177
Tatoeba-test.cym-eng.cym.eng	41.3	0.584
Tatoeba-test.gla-eng.gla.eng	16.3	0.337
Tatoeba-test.gle-eng.gle.eng	51.3	0.675
Tatoeba-test.glv-eng.glv.eng	17.0	0.525
Tatoeba-test.multi.eng	23.3	0.404