README.md

opusTCv20210807_transformer-big_2022-09-15.zip

dataset: opusTCv20210807
model: transformer-big
source language(s): pob por
target language(s): bel bel_Latn bos_Cyrl bos_Latn bul ces dsb hbs hbs_Cyrl hrv hsb mkd pol rus slv srp_Cyrl srp_Latn ukr
raw source language(s): pob por
raw target language(s): bel bos bul ces dsb hbs hrv hsb mkd pol rus slv srp ukr
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_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.multi-multi	45.4	0.66384	10000	62908	0.990
Tatoeba-test-v2021-08-07.por-bel	20.4	0.67669	3	16	0.936
Tatoeba-test-v2021-08-07.por-bul	51.7	0.65292	169	1036	0.994
Tatoeba-test-v2021-08-07.por-ces	45.1	0.67267	133	768	0.974
Tatoeba-test-v2021-08-07.por-dsb	35.4	0.81706	1	4	1.000
Tatoeba-test-v2021-08-07.por-hbs	53.9	0.69314	107	565	0.962
Tatoeba-test-v2021-08-07.por-hrv	52.7	0.68319	105	553	0.961
Tatoeba-test-v2021-08-07.por-hsb	19.0	0.44387	1	4	1.000
Tatoeba-test-v2021-08-07.por-mkd	51.7	0.76605	3	17	1.000
Tatoeba-test-v2021-08-07.por-pol	44.2	0.65811	705	4291	0.972
Tatoeba-test-v2021-08-07.por-rus	44.0	0.65455	10000	65172	0.993
Tatoeba-test-v2021-08-07.por-slv	3.4	0.25688	6	23	1.000
Tatoeba-test-v2021-08-07.por-srp_Cyrl	100.0	10.00000	2	12	1.000
Tatoeba-test-v2021-08-07.por-ukr	44.1	0.64626	3372	18863	0.980