opus-2021-02-19.zip dataset: opus model: transformer source language(s): spa target language(s): cjy cmn hsn lzh nan wuu yue 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) valid language labels: >>cmn_Hans<< >>cmn_Hant<< >>cmn<< >>yue_Hant<< >>yue_Hans<< >>nan<< 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 Tatoeba-test.spa-zho 38.8 0.324 2500 22762 0.878