opus-2021-02-19.zip dataset: opus model: transformer source language(s): spa target language(s): ces model: transformer pre-processing: normalization + SentencePiece (spm32k,spm32k) 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 newssyscomb2009.spa-ces 21.6 0.503 502 10032 0.984 news-test2008.spa-ces 19.9 0.481 2051 42484 0.986 newstest2009.spa-ces 20.7 0.489 2525 55533 0.983 newstest2010.spa-ces 22.5 0.506 2489 52958 0.990 newstest2011.spa-ces 21.4 0.490 3003 65653 0.971 newstest2012.spa-ces 20.6 0.479 3003 65456 0.969 newstest2013.spa-ces 23.7 0.506 3000 57250 0.994 Tatoeba-test.spa-ces 45.0 0.649 1807 11295 1.000