Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

训练过程和最后的评价过程使用的尾实体不一致 #1

Open
NaiveN00b opened this issue Feb 28, 2024 · 5 comments
Open

Comments

@NaiveN00b
Copy link

get_ranking中q@rhs中的rhs使用的是未加上时间的尾实体,如果按照forward中的方法进行ranking,实验结果非常差

@AAristotle
Copy link

@NaiveN00b 你好,请问如果不加时间的尾实体会导致什么错误吗?我关注到这篇论文的性能很高。

@NaiveN00b
Copy link
Author

@AAristotle 只能说get_ranking中scores和targets所得的分数不是用一个式子算出来的,那么MRR、hits@1等指标也就没意义了

@AAristotle
Copy link

@NaiveN00b 我去试了一下,的确如你所说。

@c18158977856
Copy link

请问在这种设置下 非常差是多少呢,例如在ICEWS14指标是多少呢 @NaiveN00b

@Yu-Chunmiao
Copy link

请问在这种设置下 非常差是多少呢,例如在ICEWS14指标是多少呢 @NaiveN00b

我自己在这个基础上重写了代码,训了50个epoch后mrr能达到41%,如果接着训应该能再高一点,但是绝对到不了89%

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants