zjunlp / DeepKE

[EMNLP 2022] An Open Toolkit for Knowledge Graph Extraction and Construction
http://deepke.zjukg.cn/
MIT License
3.42k stars 675 forks source link

这个问题该如何解决? #414

Closed jsh-hit closed 6 months ago

jsh-hit commented 6 months ago

Describe the question

A clear and concise description of what the question is. 使用预训练好的模型,对readme中的例子进行测试,但是报错

QQ截图20240301223803

Environment (please complete the following information):

Screenshots

If applicable, add screenshots to help explain your problem.

Additional context

Add any other context about the problem here.

zxlzr commented 6 months ago

您好,请问您使用的是哪个具体的代码,我们好帮您定位

jsh-hit commented 6 months ago

我是按照DeepKE/example/triple/cnschema/README_CN.md的教程,nerfp修改为ner模型文件夹地址 DD71A7FF7AB9A04EDB742FF49A7A5F0C 模型文件夹中文件包括 AB50F73DCC5378024D12826479A32406 具体报错在cnschema/predict.py的第60行 111 感谢您的回答!

BeasterYong commented 6 months ago

请问您使用的是哪个模型呢

BeasterYong commented 6 months ago

您有下载对应的re模型嘛?re模型也是需要下载的,然后更改refp

jsh-hit commented 6 months ago

我下载的BERT-wwm, Chinese模型,请问只进行实体识别任务也需要下载re模型吗

BeasterYong commented 6 months ago

您好,您可以参考一下readme的内容,re是中间的一个步骤,因此推荐您下载对应的re模型

jsh-hit commented 6 months ago

我下载了re模型,但是还是之前那个报错

zxlzr commented 6 months ago

请麻烦贴出报错的信息,好帮您排查

BeasterYong commented 6 months ago

您好,经我们排查后使用您对应的模型并不会出现上述问题,推荐您使用Linux系统,并且在虚拟环境中安装requirements.txt中所需要的python包