Name		Name	Last commit message	Last commit date
parent directory ..
README.md		README.md
opus+bt-2021-04-30.yml		opus+bt-2021-04-30.yml
opus+bt.yml		opus+bt.yml
opus-2021-02-19.yml		opus-2021-02-19.yml
opus-2021-02-22.yml		opus-2021-02-22.yml
opus.yml		opus.yml
opusTCv20210807+bt_transformer-big.yml		opusTCv20210807+bt_transformer-big.yml
opusTCv20210807+bt_transformer-big_2022-03-09.yml		opusTCv20210807+bt_transformer-big_2022-03-09.yml

README.md

opus-2021-02-16.zip

dataset: opus
model: transformer-align
source language(s): fra
target language(s): eng
model: transformer-align
pre-processing: normalization + SentencePiece (spm32k,spm32k)
download: opus-2021-02-16.zip
test set translations: opus-2021-02-16.test.txt
test set scores: opus-2021-02-16.eval.txt

Benchmarks

testset	BLEU	chr-F
newsdiscussdev2015-enfr-fraeng.fra.eng	33.9	0.585
newsdiscusstest2015-enfr-fraeng.fra.eng	38.9	0.617
newssyscomb2009-fraeng.fra.eng	30.2	0.571
news-test2008-fraeng.fra.eng	26.2	0.542
newstest2009-fraeng.fra.eng	30.1	0.570
newstest2010-fraeng.fra.eng	32.1	0.590
newstest2011-fraeng.fra.eng	32.9	0.597
newstest2012-fraeng.fra.eng	32.9	0.592
newstest2013-fraeng.fra.eng	34.0	0.592
newstest2014-fren-fraeng.fra.eng	37.8	0.632
Tatoeba-test.fra.eng	57.7	0.722
Tatoeba-test.fra-eng.fra.eng	57.7	0.722

opus-2021-02-17.zip

dataset: opus
model: transformer-align
source language(s): fra
target language(s): eng
model: transformer-align
pre-processing: normalization + SentencePiece (spm32k,spm32k)
download: opus-2021-02-17.zip
test set translations: opus-2021-02-17.test.txt
test set scores: opus-2021-02-17.eval.txt

Benchmarks

testset	BLEU	chr-F	#sent	#words	BP
Tatoeba-test.fra-eng	57.9	0.723	10000	77174	0.982

opus-2021-02-19.zip

dataset: opus
model: transformer-align
source language(s): fra
target language(s): eng
model: transformer-align
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
newsdiscussdev2015-enfr.fra-eng	33.9	0.585	1500	27759	0.958
newsdiscusstest2015-enfr.fra-eng	39.3	0.620	1500	26995	0.992
newssyscomb2009.fra-eng	30.7	0.573	502	11821	0.996
news-test2008.fra-eng	26.3	0.543	2051	49380	1.000
newstest2009.fra-eng	30.5	0.572	2525	65402	1.000
newstest2010.fra-eng	32.3	0.591	2489	61724	1.000
newstest2011.fra-eng	33.2	0.599	3003	74681	1.000
newstest2012.fra-eng	33.0	0.592	3003	72812	1.000
newstest2013.fra-eng	34.0	0.592	3000	64505	1.000
newstest2014-fren.fra-eng	38.0	0.633	3003	70708	1.000
Tatoeba-test.fra-eng	57.9	0.724	10000	77174	0.981

opus-2021-02-22.zip

dataset: opus
model: transformer-align
source language(s): fra
target language(s): eng
model: transformer-align
pre-processing: normalization + SentencePiece (spm32k,spm32k)
download: opus-2021-02-22.zip
test set translations: opus-2021-02-22.test.txt
test set scores: opus-2021-02-22.eval.txt

Benchmarks

testset	BLEU	chr-F	#sent	#words	BP
newsdiscussdev2015-enfr.fra-eng	34.4	0.587	1500	27759	0.960
newsdiscusstest2015-enfr.fra-eng	39.7	0.621	1500	26995	0.993
newssyscomb2009.fra-eng	31.1	0.575	502	11821	0.999
news-test2008.fra-eng	26.5	0.544	2051	49380	1.000
newstest2009.fra-eng	30.5	0.572	2525	65402	1.000
newstest2010.fra-eng	32.7	0.593	2489	61724	1.000
newstest2011.fra-eng	33.1	0.598	3003	74681	1.000
newstest2012.fra-eng	33.2	0.593	3003	72812	1.000
newstest2013.fra-eng	34.0	0.592	3000	64505	1.000
newstest2014-fren.fra-eng	38.4	0.636	3003	70708	1.000
Tatoeba-test.fra-eng	57.8	0.723	10000	77174	0.983

opus+bt-2021-04-30.zip

dataset: opus+bt
model: transformer-align
source language(s): fra
target language(s): eng
model: transformer-align
pre-processing: normalization + SentencePiece (spm32k,spm32k)
download: opus+bt-2021-04-30.zip
test set translations: opus+bt-2021-04-30.test.txt
test set scores: opus+bt-2021-04-30.eval.txt

Benchmarks

testset	BLEU	chr-F	#sent	#words	BP
newsdiscussdev2015-enfr.fra-eng	33.9	0.584	1500	27759	0.959
newsdiscusstest2015-enfr.fra-eng	38.8	0.618	1500	26995	0.987
newssyscomb2009.fra-eng	30.6	0.572	502	11821	0.995
news-test2008.fra-eng	26.2	0.542	2051	49380	1.000
newstest2009.fra-eng	30.2	0.569	2525	65402	1.000
newstest2010.fra-eng	32.5	0.591	2489	61724	1.000
newstest2011.fra-eng	33.0	0.597	3003	74681	1.000
newstest2012.fra-eng	32.9	0.591	3003	72812	1.000
newstest2013.fra-eng	33.9	0.591	3000	64505	1.000
newstest2014-fren.fra-eng	38.2	0.635	3003	70708	1.000
Tatoeba-test.fra-eng	57.6	0.721	10000	77174	0.983
tico19-test.fra-eng	40.2	0.633	2100	56347	1.000

opusTCv20210807+bt_transformer-big_2022-03-09.zip

dataset: opusTCv20210807+bt
model: transformer-big
source language(s): fra
target language(s): eng
raw source language(s): fra
raw target language(s): eng
model: transformer-big
pre-processing: normalization + SentencePiece (spm32k,spm32k)
download: opusTCv20210807+bt_transformer-big_2022-03-09.zip
test set translations: opusTCv20210807+bt_transformer-big_2022-03-09.test.txt
test set scores: opusTCv20210807+bt_transformer-big_2022-03-09.eval.txt

Benchmarks

testset	BLEU	chr-F	#sent	#words	BP
newsdiscussdev2015-enfr.fra-eng	34.4	0.59003	1500	27759	0.951
newsdiscusstest2015-enfr.fra-eng	40.2	0.62593	1500	26995	0.980
newssyscomb2009.fra-eng	31.1	0.57486	502	11821	0.989
news-test2008.fra-eng	26.5	0.54317	2051	49380	0.996
newstest2009.fra-eng	30.4	0.56959	2525	65402	0.992
newstest2010.fra-eng	33.4	0.59559	2489	61724	1.000
newstest2011.fra-eng	33.8	0.60271	3003	74681	1.000
newstest2012.fra-eng	33.6	0.59515	3003	72812	0.999
newstest2013.fra-eng	34.8	0.59691	3000	64505	1.000
newstest2014-fren.fra-eng	39.4	0.64533	3003	70708	0.991
Tatoeba-test-v2021-08-07.fra-eng	60.0	0.74090	10000	80205	0.990
tico19-test.fra-eng	41.2	0.63315	2100	56347	0.979