Opus Mt Ine Ine

🚀 Indo - European Languages Translation Model

This project focuses on translation within the Indo - European language family. It provides a transformer - based model for translating between a wide range of Indo - European languages, with detailed benchmarks to evaluate its performance.

✨ Features

• Multilingual Support: Supports a vast array of Indo - European languages for translation tasks.
• Transformer Model: Utilizes a transformer model for efficient and accurate translations.
• Pre - processing: Applies normalization and SentencePiece for pre - processing.

📦 Installation

No installation steps are provided in the original document.

💻 Usage Examples

No code examples are provided in the original document.

📚 Documentation

Language Support

The model supports the following languages:

• ca, es, os, ro, fy, cy, sc, is, yi, lb, an, sq, fr, ht, rm, ps, af, uk, sl, lt, bg, be, gd, si, en, br, mk, or, mr, ru, fo, co, oc, pl, gl, nb, bn, id, hy, da, gv, nl, pt, hi, as, kw, ga, sv, gu, wa, lv, el, it, hr, ur, nn, de, cs, ine

ine - ine Model Details

• Source Group: Indo - European languages
• Target Group: Indo - European languages
• OPUS readme: [ine - ine](https://github.com/Helsinki - NLP/Tatoeba - Challenge/tree/master/models/ine - ine/README.md)
• Model: transformer
• Source Languages: afr afr_Arab aln ang_Latn arg asm ast awa bel bel_Latn ben bho bjn bos_Latn bre bul bul_Latn cat ces cor cos csb_Latn cym dan deu dsb egl ell eng enm_Latn ext fao fra frm_Latn frr fry gcf_Latn gla gle glg glv gom gos got_Goth grc_Grek gsw guj hat hif_Latn hin hrv hsb hye hye_Latn ind isl ita jdt_Cyrl ksh kur_Arab kur_Latn lad lad_Latn lat_Grek lat_Latn lav lij lit lld_Latn lmo ltg ltz mai mar max_Latn mfe min mkd mwl nds nld nno nob nob_Hebr non_Latn npi oci ori orv_Cyrl oss pan_Guru pap pcd pdc pes pes_Latn pes_Thaa pms pnb pol por prg_Latn pus roh rom ron rue rus rus_Latn san_Deva scn sco sgs sin slv snd_Arab spa sqi srd srp_Cyrl srp_Latn stq swe swg tgk_Cyrl tly_Latn tmw_Latn ukr urd vec wln yid zlm_Latn zsm_Latn zza
• Target Languages: afr afr_Arab aln ang_Latn arg asm ast awa bel bel_Latn ben bho bjn bos_Latn bre bul bul_Latn cat ces cor cos csb_Latn cym dan deu dsb egl ell eng enm_Latn ext fao fra frm_Latn frr fry gcf_Latn gla gle glg glv gom gos got_Goth grc_Grek gsw guj hat hif_Latn hin hrv hsb hye hye_Latn ind isl ita jdt_Cyrl ksh kur_Arab kur_Latn lad lad_Latn lat_Grek lat_Latn lav lij lit lld_Latn lmo ltg ltz mai mar max_Latn mfe min mkd mwl nds nld nno nob nob_Hebr non_Latn npi oci ori orv_Cyrl oss pan_Guru pap pcd pdc pes pes_Latn pes_Thaa pms pnb pol por prg_Latn pus roh rom ron rue rus rus_Latn san_Deva scn sco sgs sin slv snd_Arab spa sqi srd srp_Cyrl srp_Latn stq swe swg tgk_Cyrl tly_Latn tmw_Latn ukr urd vec wln yid zlm_Latn zsm_Latn zza
• Pre - processing: normalization + SentencePiece (spm32k, spm32k)
• Language Token Requirement: A sentence initial language token is required in the form of >>id<< (id = valid target language ID)
• Download Original Weights: [opus - 2020 - 07 - 27.zip](https://object.pouta.csc.fi/Tatoeba - MT - models/ine - ine/opus - 2020 - 07 - 27.zip)
• Test Set Translations: [opus - 2020 - 07 - 27.test.txt](https://object.pouta.csc.fi/Tatoeba - MT - models/ine - ine/opus - 2020 - 07 - 27.test.txt)
• Test Set Scores: [opus - 2020 - 07 - 27.eval.txt](https://object.pouta.csc.fi/Tatoeba - MT - models/ine - ine/opus - 2020 - 07 - 27.eval.txt)

Benchmarks

📄 License

This project is licensed under the Apache - 2.0 license.