xyjigsaw / cenet Goto Github PK

Hello, a wonderful job. I want to ask that the hyperparameters settings are the same for all datasets or not.

您好，请问这个Oracle Training 就是单独的指第二阶段的对比训练吗？这个最后要表达的是什么呢

您好，请问这个错误是因为读取数据文件时的原因吗？

同学您好，由于计算指标的方式不同，请问CENET论文中汇报的RE-GCN结果是如何得到的呢？非常感谢您的帮助

May I know the purpose of the following code in model.forward()?

s_history_tag[s_history_tag != 0] = self.args.lambdax
o_history_tag[o_history_tag != 0] = self.args.lambdax

s_non_history_tag[s_history_tag == 1] = -self.args.lambdax
s_non_history_tag[s_history_tag == 0] = self.args.lambdax

o_non_history_tag[o_history_tag == 1] = -self.args.lambdax
o_non_history_tag[o_history_tag == 0] = self.args.lambdax

s_history_tag[s_history_tag == 0] = -self.args.lambdax
o_history_tag[o_history_tag == 0] = -self.args.lambdax

您好，请问代码中Oracle这个单词的命名指代的是对比模块的意思吗还是？谢谢解答

你好，请问一下get_history_graph.py的get_history_target方法中，对于s_history_label_true的处理是只要过去有相同（subject，？，object）的集合就能设置为true，为什么这么设置呢？根据论文描述应该是还要约束relation的信息，但这段代码不要求relation相同。

此外，cenet_model.py中关于λ的处理：
s_history_tag[s_history_tag != 0] = self.args.lambdax
o_history_tag[o_history_tag != 0] = self.args.lambdax

    s_non_history_tag[s_history_tag == 1] = -self.args.lambdax
    s_non_history_tag[s_history_tag == 0] = self.args.lambdax

    o_non_history_tag[o_history_tag == 1] = -self.args.lambdax
    o_non_history_tag[o_history_tag == 0] = self.args.lambdax

    s_history_tag[s_history_tag == 0] = -self.args.lambdax
    o_history_tag[o_history_tag == 0] = -self.args.lambdax

第三行中为什么约束s_history_tag==1呢？以上已经把s_history_tag中不等0的部分全部设置为λ了，那么如果λ不等于1，此行就不会起作用，有时间的话解决一下疑问，谢谢。

Firstly, congrats your paper has been accepted in AAAI 2023, and appreciate your elegant work on the task of knowledge graph reasoning!

I'm trying to reproduce your result according to your repo, but I found some key files are missing, e.g., main.py.

It would be greatly appreciated if you could provide the full code.

hi! it's a wonderful job in this mission, but i want to ask some question about your code performance. as your instructions, i use the hyper parameter noticed in the md, but i cannot get the same performance as your paper says. the all test hits@ is a bit lower than the paper performance, like 0.8%. could you give me your seed or some example of hyperparameter setting, or maybe it's some performance fluctuation? it really bothers me. thank your project! i like your idea!

I am running your code but meeting the above problem for dataset "GDELT" when I try to load the frequency matrix which is a novelty in your paper.
Can you tell me how to handle loading the frequency data for dataset "GDELT", because it needs 99GB to load it? I wonder how you can handle this dataset in your paper.
Thanks!

Hello! Based on the previous answer, I still have discrepancies between the experimental results on five datasets and the results mentioned in the paper. Could you please give me the parameter settings used for conducting experiments on these five datasets?

Thanks for your work!
However I couldn't get the data successfully.

Here's the device I use

What should I do?

你好！请问这个项目对环境配置有要求吗？需要几张显卡呢？

Excellent work!
I have some questions about the code. The validation set is not used, right?