opus-2020-10-04.zip dataset: opus model: transformer source language(s): eng kan mal tam tam_Deva tel target language(s): eng kan mal tam tam_Deva tel model: transformer pre-processing: normalization + SentencePiece (spm32k,spm32k) a sentence initial language token is required in the form of >>id<< (id = valid target language ID) download: opus-2020-10-04.zip test set translations: opus-2020-10-04.test.txt test set scores: opus-2020-10-04.eval.txt Benchmarks testset BLEU chr-F Tatoeba-test.eng-kan.eng.kan 7.4 0.376 Tatoeba-test.eng-mal.eng.mal 13.2 0.499 Tatoeba-test.eng-tam.eng.tam 9.2 0.434 Tatoeba-test.eng-tel.eng.tel 7.1 0.364 Tatoeba-test.kan-eng.kan.eng 22.6 0.435 Tatoeba-test.mal-eng.mal.eng 43.2 0.600 Tatoeba-test.multi.multi 26.8 0.486 Tatoeba-test.tam-eng.tam.eng 31.1 0.466 Tatoeba-test.tel-eng.tel.eng 24.2 0.430