xiehou
xiehou
maybe your socre is too big, you can make your score smaller in yolo.py.
> it will work。And I did it Can you open source the Windows version of the code? I currently have no way to run this code successfully on Windows.
ltp.py里面的模型路径没改好
> 那个relation.txt估计是哪里的公开数据集或者人工标注的吧,一般那个relation可以用实体关系抽取模型来做到,不过要标注数据,如果有公开数据集就比较简单了,直接先命名实体识别在关系抽取就会形成”实体-关系-实体“三元组,也就是那个relation.txt的一行5个值
Did you run it successfully? I have the same problem.
ok,thanks for your reply. I want to seek other solutions. ------------------ 原始邮件 ------------------ 发件人: "aboulch/ConvPoint" ***@***.***>; 发送时间: 2022年6月2日(星期四) 下午2:19 ***@***.***>; ***@***.******@***.***>; 主题: Re: [aboulch/ConvPoint] try run in windows. (Issue #45) Hello, I am sorry for...
Did you run it successfully in Windows 10? I have the same problem.
can you solve this problem? I have the same question?
install transformers==4.8.1 can solve this problem
你好,我想问一下原始项目是不是又不开源了?网上看到几个好像是fork原始代码的仓库(https://github.com/China-ChallengeHub/OneRel),特别注意到,这个头实体或者尾实体是单个字符的情况,那是不是在原始代码的情况下添加你说的这两种解码方式就行啊,就是单独考虑单个字符的情况。 你那还有没有之前开源的代码?方便分享一份吗?