Why I can't get a good result

[linzhang92]

Hi, thank you for sharing the project code. After training with the code you provided, I got much worse prediction results than in the paper. Is the parameter setting in the code the same as that used in the paper? The epoch in the code is 5000, but when I trained 1000 epochs, the rmse no longer dropped, so I only trained 1000 epochs, I do n’t know if this is the cause of bad results. If you actually use other parameter values, can you provide them to me? thank you very much.

[hutchenky]

Hi, thank you for sharing the project code. After training with the code you provided, I got much worse prediction results than in the paper. Is the parameter setting in the code the same as that used in the paper? The epoch in the code is 5000, but when I trained 1000 epochs, the rmse no longer dropped, so I only trained 1000 epochs, I do n’t know if this is the cause of bad results. If you actually use other parameter values, can you provide them to me? thank you very much.

I can't run this code because part of the data set is missing. Can you help me? Thank you very much

[linzhang92]

The datasets I used are consistent with those provided in the code. 发件人: hutchenkymailto:[email protected] 发送时间: 2020年11月18日 19:33 收件人: LahiruJayasinghe/RUL-Netmailto:[email protected] 抄送: linzhang92mailto:[email protected]; Authormailto:[email protected] 主题: Re: [LahiruJayasinghe/RUL-Net] Why I can't get a good result (#4)

Hi, thank you for sharing the project code. After training with the code you provided, I got much worse prediction results than in the paper. Is the parameter setting in the code the same as that used in the paper? The epoch in the code is 5000, but when I trained 1000 epochs, the rmse no longer dropped, so I only trained 1000 epochs, I do n’t know if this is the cause of bad results. If you actually use other parameter values, can you provide them to me? thank you very much.

I can't run this code because part of the data set is missing. Can you help me? Thank you very much

― You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://github.com/LahiruJayasinghe/RUL-Net/issues/4#issuecomment-729620931, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALITGZTCGR77GSMWMSK72RDSQOWKDANCNFSM4MSOKV4Q.

[hutchenky]

The datasets I used are consistent with those provided in the code. 发件人: hutchenkymailto:[email protected] 发送时间: 2020年11月18日 19:33 收件人: LahiruJayasinghe/RUL-Netmailto:[email protected] 抄送: linzhang92mailto:[email protected]; Authormailto:[email protected] 主题: Re: [LahiruJayasinghe/RUL-Net] Why I can't get a good result (#4) Hi, thank you for sharing the project code. After training with the code you provided, I got much worse prediction results than in the paper. Is the parameter setting in the code the same as that used in the paper? The epoch in the code is 5000, but when I trained 1000 epochs, the rmse no longer dropped, so I only trained 1000 epochs, I do n’t know if this is the cause of bad results. If you actually use other parameter values, can you provide them to me? thank you very much. I can't run this code because part of the data set is missing. Can you help me? Thank you very much ― You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub<#4 (comment)>, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ALITGZTCGR77GSMWMSK72RDSQOWKDANCNFSM4MSOKV4Q.

没有懂，能交流下吗。谢谢