模型:
Helsinki-NLP/opus-mt-en-ar
任务:
翻译许可:
apache-2.0源语言组: English
目标语言组: Arabic
OPUS自述文件: eng-ara
模型: transformer
源语言(们): eng
目标语言(们): acm afb apc apc_Latn ara ara_Latn arq arq_Latn ary arz
模型: transformer
预处理方法: normalization + SentencePiece (spm32k,spm32k)
需要一个句子开头的语言标记,格式为 >>id<< (id = 有效的目标语言ID)
下载原始权重: opus-2020-07-03.zip
测试集翻译结果: opus-2020-07-03.test.txt
测试集分数: opus-2020-07-03.eval.txt
testset | BLEU | chr-F |
---|---|---|
Tatoeba-test.eng.ara | 14.0 | 0.437 |
hf_name: eng-ara
源语言: eng
目标语言: ara
OPUS自述文件URL: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/eng-ara/README.md
原始仓库: Tatoeba-Challenge
标签: ['translation']
语言: ['en', 'ar']
源构成部分: {'eng'}
目标构成部分: {'apc', 'ara', 'arq_Latn', 'arq', 'afb', 'ara_Latn', 'apc_Latn', 'arz'}
源多语言: False
目标多语言: False
预处理: normalization + SentencePiece (spm32k,spm32k)
模型URL: https://object.pouta.csc.fi/Tatoeba-MT-models/eng-ara/opus-2020-07-03.zip
测试集URL: https://object.pouta.csc.fi/Tatoeba-MT-models/eng-ara/opus-2020-07-03.test.txt
源alpha3代码: eng
目标alpha3代码: ara
短对: en-ar
chrF2分数: 0.43700000000000006
BLEU分数: 14.0
brevity_penalty: 1.0
参考文本长度: 58935.0
源语言名称: English
目标语言名称: Arabic
训练日期: 2020-07-03
源alpha2代码: en
目标alpha2代码: ar
首选旧模型: False
长对: eng-ara
helsinki_git_sha: 480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535
transformers_git_sha: 2207e5d8cb224e954a7cba69fa4ac2309e9ff30b
port_machine: brutasse
port_time: 2020-08-21-14:41