“Qixin Wang 2016-04-25”版本间的差异

来自cslt Wiki
跳转至: 导航搜索
(以“ * Work done last week: ** New topic-many model with sequence LV model and constrain out YUN model ** Refactoring the code, pre-build model ** Accelerating the code,...”为内容创建页面)
 
Wqx讨论 | 贡献
 
(相同用户的3个中间修订版本未显示)
第10行: 第10行:
 
** some results (you can try it on WeChat now):
 
** some results (you can try it on WeChat now):
  
作诗5:花前留下
+
***作诗5:花前柳下
  
红裙带翠低,露湿锦帘西。长步摇条柳,垂杨巧向堤。
+
***红裙带翠低,露湿锦帘西。长步摇条柳,垂杨巧向堤。
  
玉梁映霓裳,美叶弄黄妆。翠盖飘香雾,披襟荡落光。
+
***玉梁映霓裳,美叶弄黄妆。翠盖飘香雾,披襟荡落光。
  
 
** Training the topic-many model for Song Iambic 15k
 
** Training the topic-many model for Song Iambic 15k
 +
 +
 +
* Plan on this week:
 +
** semantic space
 +
** New energy model

2016年4月25日 (一) 01:15的最后版本

  • Work done last week:
    • New topic-many model with sequence LV model and constrain out YUN model
    • Refactoring the code, pre-build model
    • Accelerating the code, return hidden layer, pre-calculate encoder, accelerate up to 10 times faster
    • Random sample function
    • Solved poly-phone problem, rebuilt the YUNLV list
    • Made code run in WeChat (bugs and YUN LV problem solved)
    • some results (you can try it on WeChat now):
      • 作诗5:花前柳下
      • 红裙带翠低,露湿锦帘西。长步摇条柳,垂杨巧向堤。
      • 玉梁映霓裳,美叶弄黄妆。翠盖飘香雾,披襟荡落光。
    • Training the topic-many model for Song Iambic 15k


  • Plan on this week:
    • semantic space
    • New energy model