opus-mt-he-ar (ONNX)

Usage with Transformers.js

heb-ara

• source group: Hebrew
• target group: Arabic
• OPUS readme: heb-ara
• model: transformer
• source language(s): heb
• target language(s): apc apc_Latn ara arq arz
• model: transformer
• pre-processing: normalization + SentencePiece (spm32k,spm32k)
• a sentence initial language token is required in the form of >>id<< (id = valid target language ID)
• download original weights: opus-2020-07-03.zip
• test set translations: opus-2020-07-03.test.txt
• test set scores: opus-2020-07-03.eval.txt

Benchmarks

System Info:

• hf_name: heb-ara
• source_languages: heb
• target_languages: ara
• opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/heb-ara/README.md
• original_repo: Tatoeba-Challenge
• tags: ['translation']
• languages: ['he', 'ar']
• src_constituents: {'heb'}
• tgt_constituents: {'apc', 'ara', 'arq_Latn', 'arq', 'afb', 'ara_Latn', 'apc_Latn', 'arz'}
• src_multilingual: False
• tgt_multilingual: False
• prepro: normalization + SentencePiece (spm32k,spm32k)
• url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/heb-ara/opus-2020-07-03.zip
• url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/heb-ara/opus-2020-07-03.test.txt
• src_alpha3: heb
• tgt_alpha3: ara
• short_pair: he-ar
• chrF2_score: 0.532
• bleu: 23.6
• brevity_penalty: 0.9259999999999999
• ref_len: 6372.0
• src_name: Hebrew
• tgt_name: Arabic
• train_date: 2020-07-03
• src_alpha2: he
• tgt_alpha2: ar
• prefer_old: False
• long_pair: heb-ara
• helsinki_git_sha: 480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535
• transformers_git_sha: 2207e5d8cb224e954a7cba69fa4ac2309e9ff30b
• port_machine: brutasse
• port_time: 2020-08-21-14:41