“NLP Status Report 2016-11-21”版本间的差异

来自cslt Wiki
跳转至: 导航搜索
第24行: 第24行:
 
*changed binding way of memory and atten models, can generate different style of poetry
 
*changed binding way of memory and atten models, can generate different style of poetry
 
*cleanned up our code
 
*cleanned up our code
*wrote a techreport about poemGen [[http://cslt.riit.tsinghua.edu.cn/mediawiki/images/c/c5/Atten-memory-poetry.pdf]
+
*wrote a techreport about poemGen [http://cslt.riit.tsinghua.edu.cn/mediawiki/images/c/c5/Atten-memory-poetry.pdf]
 
*submitted two databases about poemGen and musicGen
 
*submitted two databases about poemGen and musicGen
 
||  
 
||  

2016年11月21日 (一) 01:24的版本

Date People Last Week This Week
2016/11/21 Yang Feng
  • rnng+mn

1) ran experiments of rnng+mn [report] ; 2) used top-k for memory, under training

  • sequence-to-sequence + mn

1) wrote the proposal 2) discussed the details with Andy

  • intern interview
  • Huilan's work
  • rnng+mn

1) get the result of top-k; 2) try bigger memory;

  • sequence-to-sequence + mn

1)coding work

  • Huilan's work

1) try syntax-based TM

Jiyuan Zhang
  • ran decoder-memory model, but the effect is not obvious
  • changed binding way of memory and atten models, can generate different style of poetry
  • cleanned up our code
  • wrote a techreport about poemGen [1]
  • submitted two databases about poemGen and musicGen
  • explored a variety of binding ways of memory and atten model
Andi Zhang
  • prepare new data set for paraphrase, wiped out repetition & most of the noises
  • run NMT on fr-en data set and new paraphrase set
  • read through source code to find ways to modify it
  • helped Guli with running NMT on our server
  • decide to drop theano or not
  • start to work on codes
Shiyue Zhang
  • run rnng on MKL successfully, which can double or triple the speed.
  • rerun the original model and get the final result 92.32
  • rerun the wrong memory models, still running
  • implement the dynamic memory model and get the result 92.54 which is 0.22 better than baseline
  • try another structure of memory
  • try more different models and summary the results
  • publish the technical reports
Guli