eng-spa

opus-2021-02-19.zip

• dataset: opus
• model: transformer
• source language(s): eng
• target language(s): spa
• model: transformer
• pre-processing: normalization + SentencePiece (spm32k,spm32k)
• download: opus-2021-02-19.zip
• test set translations: opus-2021-02-19.test.txt
• test set scores: opus-2021-02-19.eval.txt

Benchmarks

testset	BLEU	chr-F	#sent	#words	BP
newssyscomb2009.eng-spa	31.3	0.583	502	12506	0.990
news-test2008.eng-spa	29.6	0.564	2051	52596	1.000
newstest2009.eng-spa	30.2	0.578	2525	68114	1.000
newstest2010.eng-spa	36.9	0.620	2489	65522	1.000
newstest2011.eng-spa	38.3	0.620	3003	79476	0.984
newstest2012.eng-spa	39.1	0.626	3003	79006	0.969
newstest2013.eng-spa	35.1	0.598	3000	70528	0.960
Tatoeba-test.eng-spa	55.1	0.721	10000	77311	0.978
tico19-test.eng-spa	50.4	0.727	2100	66591	0.959

opus+bt-2021-04-10.zip

• dataset: opus+bt
• model: transformer-align
• source language(s): eng
• target language(s): spa
• model: transformer-align
• pre-processing: normalization + SentencePiece (spm32k,spm32k)
• download: opus+bt-2021-04-10.zip
• test set translations: opus+bt-2021-04-10.test.txt
• test set scores: opus+bt-2021-04-10.eval.txt

Benchmarks

testset	BLEU	chr-F	#sent	#words	BP
newssyscomb2009.eng-spa	30.8	0.581	502	12506	0.989
news-test2008.eng-spa	29.8	0.566	2051	52596	1.000
newstest2009.eng-spa	30.3	0.579	2525	68114	1.000
newstest2010.eng-spa	37.1	0.621	2489	65522	1.000
newstest2011.eng-spa	38.2	0.620	3003	79476	0.984
newstest2012.eng-spa	39.0	0.627	3003	79006	0.969
newstest2013.eng-spa	35.3	0.600	3000	70528	0.960
Tatoeba-test.eng-spa	54.9	0.721	10000	77311	0.979
tico19-test.eng-spa	50.5	0.729	2100	66591	0.961

opusTCv20210807+bt_transformer-big_2022-03-13.zip

• dataset: opusTCv20210807+bt
• model: transformer-big
• source language(s): eng
• target language(s): spa
• raw source language(s): eng
• raw target language(s): spa
• model: transformer-big
• pre-processing: normalization + SentencePiece (spm32k,spm32k)
• download: opusTCv20210807+bt_transformer-big_2022-03-13.zip
• test set translations: opusTCv20210807+bt_transformer-big_2022-03-13.test.txt
• test set scores: opusTCv20210807+bt_transformer-big_2022-03-13.eval.txt

Benchmarks

testset	BLEU	chr-F	#sent	#words	BP
newssyscomb2009.eng-spa	31.3	0.58391	502	12506	0.998
news-test2008.eng-spa	30.1	0.56595	2051	52596	0.999
newstest2009.eng-spa	30.3	0.57884	2525	68114	1.000
newstest2010.eng-spa	37.5	0.62334	2489	65522	1.000
newstest2011.eng-spa	39.0	0.62460	3003	79476	0.980
newstest2012.eng-spa	39.5	0.62941	3003	79006	0.966
newstest2013.eng-spa	35.9	0.60383	3000	70528	0.959
Tatoeba-test-v2021-08-07.eng-spa	56.9	0.73626	10000	80493	0.990
tico19-test.eng-spa	52.9	0.73528	2100	66591	0.941