“FIN Status Report 2017-3-23”版本间的差异

来自cslt Wiki
跳转至: 导航搜索
 
(2位用户的4个中间修订版本未显示)
第4行: 第4行:
 
|-
 
|-
 
| Yang Wang||
 
| Yang Wang||
train the model with different stock set, which consist of 10 stocks belonging to 10 industries. <br>  
+
refractor the code and build a toolkit bulltrader.<br>  
digitize the asset and set it as reward, which makes the training and the target more mathing<br>
+
work in haixia, learn some experience.<br>
clip the gradients, which stabilizes the training<br>
+
randomize the goal. Its impact remains to be seen<br>
+
  
 
||  
 
||  
do more test
+
tune our model and toolkit.
  
 
|-
 
|-
 
| Yuxin Zhang||  
 
| Yuxin Zhang||  
Try to use Pyalgotrading to test the algorithm.<br>
+
Did more experiments of ddpg.<br>
Start to develop algorithm for in-day market timing.<br>
+
Learn trading in the real world with Yansheng.
Did more experiments of ddpg.
+
 
||  
 
||  
Test the ddpg algorithm with stock selection algorithm.<br>
+
Develop algorithm for in-day market timing.<br>
Develop algorithm for in-day market timing.
+
Improve ddpg.
 
|-
 
|-
  
第28行: 第25行:
  
 
|-
 
|-
| Tongzheng Ren|| Review code.
+
| Tongzheng Ren||Help implement some of the toolkit.
  
|| Help do test.
+
||Help tune the toolkit.
  
 
|}
 
|}

2017年3月23日 (四) 03:56的最后版本

People Last Week This Week
Yang Wang

refractor the code and build a toolkit bulltrader.
work in haixia, learn some experience.

tune our model and toolkit.

Yuxin Zhang

Did more experiments of ddpg.
Learn trading in the real world with Yansheng.

Develop algorithm for in-day market timing.
Improve ddpg.

Yaodong Wang Update new wind data to databases and Write the markdown of code Continue writing the markdown and update data everyday
Tongzheng Ren Help implement some of the toolkit. Help tune the toolkit.