-
Notifications
You must be signed in to change notification settings - Fork 50
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
evaluate.py #2
Comments
我正在训练, 你这边训练好了么?? |
你好 你这边训练的模型能达到论文的精度吗? |
代码更新了。效果在README中有写。比起论文的精度要差一点。没有对超参数进行微调,并且解码的结果中还存在exposure bias的问题,我尝试加入schedule sampling解决,后续有进展会继续更新。 |
font{
line-height: 1.6;
}
ul,ol{
padding-left: 20px;
list-style-position: inside;
}
您好:
非常感谢您更新代码,然后我这边现在遇到一个问题就是我在用哈佛的脚本(https://github.com/harvardnlp/im2markup)处理标签的时候遇到很多问题,例如:原本的标签是f,\tilde{f}\to 1,\hspace{3mm} a,\tilde{a}\to 1\hspace{3mm} (r\to \infty). 对应的公式是
然后用哈佛的脚本正则化之后的标签是 f , \tilde { f } \to 1 , \hspace { 3 m m } a , \tilde { a } \to 1 \hspace { 3 m m } ( r \to \infty ) . 对应的公式却是也就是说哈佛的脚本处理标签还是会有很多问题,所以我想请问一下您这边是怎么处理标签以及图片数据的。谢谢。祝好
刘恒业
On 9/26/2019 19:15,luopeixiang<[email protected]> wrote:
代码更新了。效果在README中有写。比起论文的精度要差一点。没有对超参数进行微调,并且解码的结果中还存在exposure bias的问题,我尝试加入schedule sampling解决,后续有进展会继续更新。
—You are receiving this because you commented.Reply to this email directly, view it on GitHub, or mute the thread.
|
http://lstm.seas.harvard.edu/latex/data/ |
你好,能提供预训练模型和evaluate.py文件吗
The text was updated successfully, but these errors were encountered: