“NLP Status Report 2017-5-15”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"|2017/4/5 |Jiyuan Zhang || || |- |Aodong LI || || |- |Shiyue Zhang || || |- |Shipan...”为内容创建页面)
 
第11行: 第11行:
 
|-
 
|-
 
|Shiyue Zhang ||  
 
|Shiyue Zhang ||  
 +
* got result of M-NMT 28.92 ( +2.2, baseline=26.73)
 +
* trained word2vec on big zh,uy data
 +
* tested NMT baseline when there is UNK in ref, bleu=34.10 (better than MOSES=33.10), which means UNK is the biggest problem in NMT
 +
* found a problem in dataset, some sentences are reversed
 
||
 
||
 +
* test embedding untrained model
 +
* fix reversed sentences problem and rerun on MOSES, NMT, M-NMT
 +
* implement UNK model
 
|-
 
|-
 
|Shipan Ren ||
 
|Shipan Ren ||

2017年5月15日 (一) 01:38的版本

Date People Last Week This Week
2017/4/5 Jiyuan Zhang
Aodong LI
Shiyue Zhang
  • got result of M-NMT 28.92 ( +2.2, baseline=26.73)
  • trained word2vec on big zh,uy data
  • tested NMT baseline when there is UNK in ref, bleu=34.10 (better than MOSES=33.10), which means UNK is the biggest problem in NMT
  • found a problem in dataset, some sentences are reversed
  • test embedding untrained model
  • fix reversed sentences problem and rerun on MOSES, NMT, M-NMT
  • implement UNK model
Shipan Ren