chunyuany / dialogue Goto Github PK

你好
我想請問要如何從文字檔產生vocab_and_embeddings.pkl

你好，请问下数据是怎么处理的，能否提供一下？我们打算用在项目上，谢谢

It seems that the score file in your google drive folder could not give the same result as reported in the article on Ubuntu Corpus ?
The results computed by Metric.py are

R1: 0.6459
R2: 0.79898
R5: 0.95414

Hi, I cannot reproduce the exactly same results with your pertained models.
on some metrics, I got high scores than your report numbers, while others are lower.
I suspect this may be due to the environment difference.
could you please give more specific information about your work environment? like the version of dependencies.

thanks!

Thanks for your great work ! I'd like to reproduce your result on Douban corpus and incorporate your result in our manuscript. However, the script for preprecessing the data is not provided in this repository. Could you please provide the data processing script or the processed Douban corpus ? It is important to reproduce your result. Thanks a lot !