eng-lav

opus+bt-2021-04-14.zip

• dataset: opus+bt
• model: transformer-align
• source language(s): eng
• target language(s): lav ltg
• model: transformer-align
• pre-processing: normalization + SentencePiece (spm32k,spm32k)
• a sentence initial language token is required in the form of >>id<< (id = valid target language ID)
• valid language labels: >>lav<< >>ltg<<
• download: opus+bt-2021-04-14.zip
• test set translations: opus+bt-2021-04-14.test.txt
• test set scores: opus+bt-2021-04-14.eval.txt

Benchmarks

testset	BLEU	chr-F	#sent	#words	BP
newsdev2017-enlv.eng-lav	27.9	0.578	2003	41503	0.992
newstest2017-enlv.eng-lav	21.1	0.531	2001	39434	1.000
Tatoeba-test.eng-lav	46.8	0.683	1631	9932	0.986
Tatoeba-test.eng-ltg	10.7	0.392	1	4	1.000

opusTCv20210807+bt_transformer-big_2022-03-13.zip

• dataset: opusTCv20210807+bt
• model: transformer-big
• source language(s): eng
• target language(s): lav ltg
• raw source language(s): eng
• raw target language(s): lav ltg
• model: transformer-big
• pre-processing: normalization + SentencePiece (spm32k,spm32k)
• a sentence initial language token is required in the form of >>id<< (id = valid target language ID)
• valid language labels:
• 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
newsdev2017-enlv.eng-lav	28.6	0.57867	2003	41503	0.992
newstest2017-enlv.eng-lav	22.1	0.53124	2001	39434	1.000
Tatoeba-test-v2021-08-07.eng-ltg	21.4	0.63773	1	4	1.000
Tatoeba-test-v2021-08-07.eng-multi	44.0	0.66411	1631	9932	1.000