Helsinki-NLP/opus-mt-tum-en

Opus-mt-tum-en dataset: opus-2020-01-21.zip . The dataset includes original weights, original translations and scores . The data set is based on languages: en, pre-processing, normalization + SentencePiece, translation, and scoring methods . The study was created from a database called Opus.mt-en.en.tum.en-en . The results were published on December 1, 2020, January 1, 2019 and January 2, 2019 . The language used is en-en and en-languages are en, en, and pre-processed and translated. The study is open to all languages and users,