[glossary] token classification-词元分类

#33
by AIBoy1993 - opened

出处:The Hugging Face Hub is home to over 100,000 public models for different tasks such as next-word prediction, mask filling, token classification, sequence classification, and so on.
示例:Hugging Face Hub拥有超过10万个用于不同任务的公共模型,例如下一词预测、掩码填充、词元分类、序列分类等。
参考:https://github.com/jiqizhixin/Artificial-Intelligence-Terminology-Database/blob/master/data/T.md

Hugging Face China Community org

口语化的翻译我建议翻译成:符号分类或者词符分类
如果考虑汉语语言学的语境,我建议翻译成:形符分类
形符,也称作意符,指汉字形声字的构成中,类属形体及指明意义类别的部分。token在英语中的定义跟这个比较接近。

Hugging Face China Community org

“词元”在语言学里对应了另一个概念lemma,当一个词有多种不同的形式时,词元(lemma)是主要的词条,其定义、词源、变格等都应放在这里。该词的所有其他形式称为非词元形式(non-lemma forms),这类的词条一般只包含与词元相连的连结。
就感觉token翻译成词元是一个错误翻译。

Sign up or log in to comment