HIT-SCIR / ltp

Language Technology Platform
http://ltp.ai
4.96k stars 1.04k forks source link

分词报错 #693

Open dedey opened 6 months ago

dedey commented 6 months ago

句子:视觉Transformers通过将图像区域表示为转换后的tokens并通过注意力权重整合它们来提取视觉信息。 报错信息: File "/venv3/lib/python3.10/site-packages/ltp/nerual.py", line 293, in _cws_post entities[-1].append((0, length2index[e])) KeyError: 13

@AlongWY 麻烦,帮忙看看

AlongWY commented 6 months ago

无法复现

dedey commented 6 months ago

您在处理时,有没有加载自己的字典?@AlongWY