xu-song's picture
update
428b731
raw
history blame
330 Bytes
from zhon.hanzi import punctuation as zh_punc
"""
这些zh_punc都是oov
['"', '#', '$', '%', '&', ''', '*', '+', '-', '/', '<', '=', '>', '@', '[', '\', ']',
'^', '_', '`', '{', '}', '⦅', '⦆', '\u3000', '〘', '〙', '〚', '〛', '〟', '〰', '〾', '〿', '‛', '‟']
"""