Opus Mt En Sla

🚀 eng-sla Translation Model

This model is designed for translating English to Slavic languages. It provides a reliable solution for cross - language communication within the English - Slavic language context.

🚀 Quick Start

The eng-sla model is a transformer - based translation model. To use it, you need to follow the pre - processing steps and download the necessary files.

Prerequisites

• A sentence initial language token in the form of >>id<< (id = valid target language ID) is required.
• Pre - processing involves normalization + SentencePiece (spm32k,spm32k).

Download Resources

• Download original weights: opus2m-2020-08-01.zip
• Access test set translations: opus2m-2020-08-01.test.txt
• Check test set scores: opus2m-2020-08-01.eval.txt

✨ Features

• Multilingual Targets: Supports translation from English to multiple Slavic languages, including Belarusian, Bosnian, Bulgarian, Czech, Croatian, Macedonian, Polish, Russian, Serbian, Slovak, Slovene, and Ukrainian.
• Transformer Architecture: Utilizes the powerful transformer model for high - quality translations.
• Pre - processing: Applies normalization and SentencePiece for better performance.

📦 Installation

No specific installation steps are provided in the original document.

📚 Documentation

Model Information

System Info

• hf_name: eng - sla
• source_languages: eng
• target_languages: sla
• opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/eng-sla/README.md
• original_repo: Tatoeba - Challenge
• tags: ['translation']
• languages: ['en', 'be', 'hr', 'mk', 'cs', 'ru', 'pl', 'bg', 'uk', 'sl', 'sla']
• src_constituents: {'eng'}
• tgt_constituents: {'bel', 'hrv', 'orv_Cyrl', 'mkd', 'bel_Latn', 'srp_Latn', 'bul_Latn', 'ces', 'bos_Latn', 'csb_Latn', 'dsb', 'hsb', 'rus', 'srp_Cyrl', 'pol', 'rue', 'bul', 'ukr', 'slv'}
• src_multilingual: False
• tgt_multilingual: True
• prepro: normalization + SentencePiece (spm32k,spm32k)
• url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/eng-sla/opus2m-2020-08-01.zip
• url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/eng-sla/opus2m-2020-08-01.test.txt
• src_alpha3: eng
• tgt_alpha3: sla
• short_pair: en - sla
• chrF2_score: 0.61
• bleu: 41.0
• brevity_penalty: 0.976
• ref_len: 64809.0
• src_name: English
• tgt_name: Slavic languages
• train_date: 2020 - 08 - 01
• src_alpha2: en
• tgt_alpha2: sla
• prefer_old: False
• long_pair: eng - sla
• helsinki_git_sha: 480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535
• transformers_git_sha: 2207e5d8cb224e954a7cba69fa4ac2309e9ff30b
• port_machine: brutasse
• port_time: 2020 - 08 - 21 - 14:41

🔧 Technical Details

No specific technical details are provided in the original document.

📄 License

The model is licensed under the Apache 2.0 license.

Benchmarks