Add comma deletion to data noising

This commit is contained in:
Matthew Honnibal 2018-12-01 13:42:18 +00:00
parent 21ee1c7a17
commit 9536ee787c
1 changed files with 1 additions and 1 deletions

View File

@ -265,7 +265,7 @@ def _corrupt(c, noise_level):
return '\n' return '\n'
elif c == '\n': elif c == '\n':
return ' ' return ' '
elif c in ['.', "'", "!", "?"]: elif c in ['.', "'", "!", "?", ',']:
return '' return ''
else: else:
return c.lower() return c.lower()