Probably a typo in the paper

[ZifengDing]

Hello,

I was checking your paper. In table 4, hits3 of ICEWS14 on RE-Net seems not very reasonable (it might be higher?). And could you please also provide the hits1 performance?

Best, Zifeng

[fangyiming]

Hello,

I was checking your paper. In table 4, hits3 of ICEWS14 on RE-Net seems not very reasonable (it might be higher?). And could you please also provide the hits1 performance?

Best, Zifeng

HELLO,I'm checking too.Can you run his code? Or could you please send your code to me? Thank you, my email [email protected]

[ZifengDing]

Hello, I was checking your paper. In table 4, hits3 of ICEWS14 on RE-Net seems not very reasonable (it might be higher?). And could you please also provide the hits1 performance? Best, Zifeng

HELLO,I'm checking too.Can you run his code? Or could you please send your code to me? Thank you, my email [email protected]

Hello, I ran it and got 26.495 as Hits@3 for ICEWS14 raw result. The code provided by author is runable, just a bit slow.

[fangyiming]

Thank you for your reply, I run the code a lot of problems, can you send your code and configuration information to me, what I use is python3.7, torch1.1, cuda11.0, thank you very much

------------------ 原始邮件 ------------------ 发件人: "INK-USC/RE-Net" <[email protected]>; 发送时间: 2020年12月15日(星期二) 上午7:41 收件人: "INK-USC/RE-Net"<[email protected]>; 抄送: "房意明"<[email protected]>;"Comment"<[email protected]>; 主题: Re: [INK-USC/RE-Net] Probably a typo in the paper (#37)

Hello, I was checking your paper. In table 4, hits3 of ICEWS14 on RE-Net seems not very reasonable (it might be higher?). And could you please also provide the hits1 performance? Best, Zifeng

HELLO,I'm checking too.Can you run his code? Or could you please send your code to me? Thank you, my email [email protected]

Hello, I ran it and got 26.495 as Hits@3 for ICEWS14 raw result. The code provided by author is runable, just a bit slow.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

[woojeongjin]

Hi! Thanks for letting me know. The result on the paper is Hit@1. It should be 26.56.

[ZhengJialin1000]

Thank you for your reply, I run the code a lot of problems, can you send your code and configuration information to me, what I use is python3.7, torch1.1, cuda11.0, thank you very much … ------------------ 原始邮件 ------------------ 发件人: "INK-USC/RE-Net" <[email protected]>; 发送时间: 2020年12月15日(星期二) 上午7:41 收件人: "INK-USC/RE-Net"<[email protected]>; 抄送: "房意明"<[email protected]>;"Comment"<[email protected]>; 主题: Re: [INK-USC/RE-Net] Probably a typo in the paper (#37) Hello, I was checking your paper. In table 4, hits3 of ICEWS14 on RE-Net seems not very reasonable (it might be higher?). And could you please also provide the hits1 performance? Best, Zifeng HELLO,I'm checking too.Can you run his code? Or could you please send your code to me? Thank you, my email [email protected] Hello, I ran it and got 26.495 as Hits@3 for ICEWS14 raw result. The code provided by author is runable, just a bit slow. — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

HELLO,I'm checking too.Could you run his code now? I met the environment problem cause dgl updated.My device is the same to you.Could you send a version of yours.Thank you, my email is [email protected]