lighteternal/SSE-TUC-mt-el-en-cased

Using the Fairseq framework, transformer_iwslt_de_en architecture . BPE segmentation (20k codes). Mixed-case model .Tokenizer tokenizer, FSMTForConditionalGeneration (FSMT) tokenizer.tokenizer.from_pretrained(FSMTTokenizer)tokenizer .model.model: transformer(fairseq)pre-processing: tokenization + BPE segmentsation. BPE.model. model. model . Model description: "Trained using the Fairsequ framework, transformers. from Opus, CCmatrix. model." Text: "Greek to English NMT. The Hellenic Army Academy",