jpn-hbs
source group: Japanese
target group: Serbo-Croatian
OPUS readme: jpn-hbs
model: transformer-align
source language(s): jpn jpn_Bopo jpn_Hani jpn_Hira jpn_Kana jpn_Latn
target language(s): bos_Latn hrv srp_Cyrl srp_Latn
model: transformer-align
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-2020-06-17.zip
test set translations: opus-2020-06-17.test.txt
test set scores: opus-2020-06-17.eval.txt
Benchmarks
testset | BLEU | chr-F |
---|---|---|
Tatoeba-test.jpn.hbs | 22.6 | 0.447 |
System Info:
hf_name: jpn-hbs
source_languages: jpn
target_languages: hbs
opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/jpn-hbs/README.md
original_repo: Tatoeba-Challenge
tags: ['translation']
languages: ['ja', 'sh']
src_constituents: {'jpn_Hang', 'jpn', 'jpn_Yiii', 'jpn_Kana', 'jpn_Hani', 'jpn_Bopo', 'jpn_Latn', 'jpn_Hira'}
tgt_constituents: {'hrv', 'srp_Cyrl', 'bos_Latn', 'srp_Latn'}
src_multilingual: False
tgt_multilingual: False
prepro: normalization + SentencePiece (spm32k,spm32k)
url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/jpn-hbs/opus-2020-06-17.zip
url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/jpn-hbs/opus-2020-06-17.test.txt
src_alpha3: jpn
tgt_alpha3: hbs
short_pair: ja-sh
chrF2_score: 0.447
bleu: 22.6
brevity_penalty: 0.9620000000000001
ref_len: 2525.0
src_name: Japanese
tgt_name: Serbo-Croatian
train_date: 2020-06-17
src_alpha2: ja
tgt_alpha2: sh
prefer_old: False
long_pair: jpn-hbs
helsinki_git_sha: 480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535
transformers_git_sha: 2207e5d8cb224e954a7cba69fa4ac2309e9ff30b
port_machine: brutasse
port_time: 2020-08-21-14:41
- Downloads last month
- 12