“NLP Status Report 2016-10-31”版本间的差异

来自cslt Wiki
跳转至： 导航、 搜索
（以“{| class="wikitable" !Date !! People !! Last Week !! This Week |- | rowspan="6"|2016/10/24 |Yang Feng || *surveyed NN with C and wrote the code to add Memory network...”为内容创建页面）
 
 
（4位用户的10个中间修订版本未显示）
第2行： 第2行：
 
!Date !! People !! Last Week !! This Week
 
!Date !! People !! Last Week !! This Week
 
|-
 
|-
−
| rowspan="6"|2016/10/24
+
| rowspan="6"|2016/10/31
 
|Yang Feng ||
 
|Yang Feng ||
−
*surveyed NN with C and wrote the code to add Memory network to RNNG
+
*almost finish the coding of rnng with memory network and can transfer to Shiyue today or tomorrow.
 
||
 
||
−
*finish and debug the code
+
*start the work of sequence-to-sequence translation.
 
|-
 
|-
 
|Jiyuan Zhang ||
 
|Jiyuan Zhang ||
−
*fixed the code of poetry generation
+
*attemped to solve the bug that costs which are calculated by running poetry model weren't exactly the same(partially solved)
−
*understood the paper of MemNN
+
*coded the memory model,bebugging the model
−
*checked the code about MemNN
+
 
||
 
||
−
*integrate MemNN into the poetry generation
+
*run and improve memory model
 
|-
 
|-
 
|Aodong Li ||  
 
|Aodong Li ||  
第21行： 第20行：
 
|-
 
|-
 
|Andi Zhang ||
 
|Andi Zhang ||
−
*read source code of seq2seq of tensorflow
+
*ran machine translation model
−
*then shifted to original version in theano (deprecated package)
+
*prepare the Chinese training set
 
||
 
||
−
*install deps for theano
+
*try to run machine translation model on chinese QA dataset
−
*understand source code
+
 
|-
 
|-
 
|Shiyao Li ||
 
|Shiyao Li ||
第33行： 第31行：
 
|-
 
|-
 
|Shiyue Zhang ||  
 
|Shiyue Zhang ||  
−
* run, read and understand the code of MemNN
+
* finished writing the document of rnng and cnn [[http://cslt.riit.tsinghua.edu.cn/mediawiki/images/e/e6/RNNG_Document.pdf rnng doc]]
−
* discuss the code with Teacher Feng
+
* tried a simple way to add memory into rnng [[http://cslt.riit.tsinghua.edu.cn/mediawiki/images/2/2f/RNNG%2Bmm%E5%AE%9E%E9%AA%8C%E6%8A%A5%E5%91%8A.pdf RNNG+MM实验报告]]
−
* simply revise Rnng Code Use Guide [[http://cslt.riit.tsinghua.edu.cn/mediawiki/images/2/2b/RNNG_Code_Use_Guide_simplified.pdf Rnng Code Use Guide]]
+
 
||
 
||
−
* write a doc of RNNG code
+
* try better way to add memory into rnng
 
|-
 
|-
 
|}
 
|}

2016年11月2日 (三) 01:18的最后版本

Date People Last Week This Week
2016/10/31 Yang Feng
  • almost finish the coding of rnng with memory network and can transfer to Shiyue today or tomorrow.
  • start the work of sequence-to-sequence translation.
Jiyuan Zhang
  • attemped to solve the bug that costs which are calculated by running poetry model weren't exactly the same(partially solved)
  • coded the memory model,bebugging the model
  • run and improve memory model
Aodong Li
Andi Zhang
  • ran machine translation model
  • prepare the Chinese training set
  • try to run machine translation model on chinese QA dataset
Shiyao Li
Shiyue Zhang
  • try better way to add memory into rnng
取自“http://index.cslt.org/mediawiki/index.php?title=NLP_Status_Report_2016-10-31&oldid=23332”