-
Notifications
You must be signed in to change notification settings - Fork 10
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
About PG Net #12
Comments
预训练有20轮,最终rouge值最高值的train-epoch为500 |
你好,我最近重新看了一下代码,发现在generator模块data.py里面的ImpressionDataset好像有点问题,具体问题见截图,大概就是target_input和target_output用的都是从target_inputs索引出来的。你看看这里是不是有点问题,谢谢。
…------------------ 原始邮件 ------------------
发件人: "LLluoling/PENS-Personalized-News-Headline-Generation" ***@***.***>;
发送时间: 2022年1月24日(星期一) 中午1:46
***@***.***>;
***@***.******@***.***>;
主题: Re: [LLluoling/PENS-Personalized-News-Headline-Generation] About PG Net (Issue #12)
这个评估方式有很严重的问题,我直接在预训练阶段训练200个epoch后不进行强化学习,直接去评估,rouge分数高的离谱
—
Reply to this email directly, view it on GitHub, or unsubscribe.
Triage notifications on the go with GitHub Mobile for iOS or Android.
You are receiving this because you authored the thread.Message ID: ***@***.***>
|
谢谢您的反馈 已修改 |
我自己在您所给代码的基础上,修改了小部分代码,使得Decoder的初始hidden state为encoder的最后一个hidden state而非user_embedding,并且embedding未被预训练,(即修改了论文中融入User_embedding的方式一)完成了PG+强化学习的代码。最后训练得的指针网络结果Rouge-1为28.1,Rouge-2为10.7,Rouge-L为22.7。Rouge版本为0.3.1,且测试集的标题并未参与到预训练过程中(因为数据加载相关代码使用的是您所给的代码)。请问这可能是什么问题导致的?
The text was updated successfully, but these errors were encountered: