How can I reproduce the Paper result?

[r06921078]

No description provided.

[r06921078]

How can I reproduce the Paper result?

[todpole3]

Have you try following the instructions in the README?

If you met any specific problems, please post them here.

[r06921078]

Hi: Yes, I clone this repo and follow the "Run Experiments" steps,(I didn't modify any hyper parameters in this repo) but I got the result in dev is not reach the Accuracy in paper, Can you help me find out the question? Very thank you. Kang Victoria X Lin <notifications@github.com> 於 2019年4月1日 週一 上午11:43寫道：

…

Have you try following the instructions in the README? If you met any specific problems, please post them here. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#15 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/Ad7hAEwgZyYF7r9SWQObRkctdBVMtt-bks5vcYBJgaJpZM4cUP9y> .

[todpole3]

Ah, please check if it is the automatic evaluation metrics that does not match the published results or it is the manual evaluation metrics.

If only the manual evaluation metrics does not match those in the paper, you might be experiencing the same issue as #6.