README.md

opusTCv20210807_transformer-big_2022-09-15.zip

dataset: opusTCv20210807
model: transformer-big
source language(s): bel bel_Latn bos_Cyrl bos_Latn bul ces dsb hbs hbs_Cyrl hrv hsb mkd orv_Cyrl pol rus slv srp_Cyrl srp_Latn ukr
target language(s): fra
raw source language(s): bel bos bul ces dsb hbs hrv hsb mkd orv pol rus slv srp ukr
raw target language(s): fra
model: transformer-big
pre-processing: normalization + SentencePiece (spm32k,spm32k)
download: opusTCv20210807_transformer-big_2022-09-15.zip
test set translations: opusTCv20210807_transformer-big_2022-09-15.test.txt
test set scores: opusTCv20210807_transformer-big_2022-09-15.eval.txt

testset	BLEU	chr-F	#sent	#words	BP
Tatoeba-test-v2021-08-07.bel-fra	52.1	0.68739	283	2005	1.000
Tatoeba-test-v2021-08-07.bos_Latn-fra	51.6	0.74849	42	211	0.931
Tatoeba-test-v2021-08-07.bul-fra	53.7	0.69997	446	3669	0.970
Tatoeba-test-v2021-08-07.ces-fra	55.2	0.71144	438	3345	0.970
Tatoeba-test-v2021-08-07.dsb-fra	46.6	0.59848	4	22	0.905
Tatoeba-test-v2021-08-07.hbs-fra	49.6	0.68413	474	3368	0.959
Tatoeba-test-v2021-08-07.hrv-fra	44.9	0.64588	258	1943	0.946
Tatoeba-test-v2021-08-07.hsb-fra	42.4	0.54063	6	127	0.968
Tatoeba-test-v2021-08-07.mkd-fra	36.6	0.62266	26	171	1.000
Tatoeba-test-v2021-08-07.multi-fra	52.1	0.68104	10000	68650	0.975
Tatoeba-test-v2021-08-07.orv-fra	8.4	0.27583	37	290	0.921
Tatoeba-test-v2021-08-07.pol-fra	48.9	0.66709	3087	24254	0.970
Tatoeba-test-v2021-08-07.rus-fra	51.3	0.68161	11490	80573	0.977
Tatoeba-test-v2021-08-07.slv-fra	36.8	0.56198	448	3791	0.953
Tatoeba-test-v2021-08-07.srp_Cyrl-fra	53.3	0.71509	130	895	0.977
Tatoeba-test-v2021-08-07.srp_Latn-fra	65.6	0.79022	44	319	1.000
Tatoeba-test-v2021-08-07.ukr-fra	51.3	0.68178	10035	63222	0.987