Difference between revisions of "NLP Status Report 2017-5-22"

From cslt Wiki
Jump to: navigation, search
Line 7: Line 7:
 
|-
 
|-
 
|Aodong LI ||
 
|Aodong LI ||
 
+
* bleu of baseline = 43.87
 +
* 2nd translator uses as training data the concat(Chinese, machine translated English):
 +
  hidden_size, emb_size, lr = 500, 310, 0.001 bleu = 43.53 (best)
 +
  hidden_size, emb_size, lr = 700, 510, 0.001 bleu = 45.21 (best) but most results are under 43.1
 +
  hidden_size, emb_size, lr = 700, 510, 0.0005 bleu = 42.19 (best)
 +
* double-decoder model with joint loss (final loss = 1st decoder's loss + 2nd decoder's loss):
 +
  bleu = 40.11 (best)
 +
  The 1st decoder's output is generally better than 2nd decoder's output.
 +
* The training process of double-decoder model '''without''' joint loss is problematic.
 
||
 
||
 +
* Replace the force teaching mechanism in training process with beam search mechanism.
 
|-
 
|-
 
|Shiyue Zhang ||  
 
|Shiyue Zhang ||  

Revision as of 04:27, 24 May 2017

Date People Last Week This Week
2017/5/22 Jiyuan Zhang
Aodong LI
  • bleu of baseline = 43.87
  • 2nd translator uses as training data the concat(Chinese, machine translated English):
 hidden_size, emb_size, lr = 500, 310, 0.001 bleu = 43.53 (best)
 hidden_size, emb_size, lr = 700, 510, 0.001 bleu = 45.21 (best) but most results are under 43.1
 hidden_size, emb_size, lr = 700, 510, 0.0005 bleu = 42.19 (best)
  • double-decoder model with joint loss (final loss = 1st decoder's loss + 2nd decoder's loss):
 bleu = 40.11 (best)
 The 1st decoder's output is generally better than 2nd decoder's output.
  • The training process of double-decoder model without joint loss is problematic.
  • Replace the force teaching mechanism in training process with beam search mechanism.
Shiyue Zhang
  • tried to not train embedding but use external word vectors
  • most results of my attempts are bad, only 3-layer rnn + no dropout model got 25.54 bleu which about 2 points worse than original baseline
  • trained original baseline on new data ( the data fixed the reverse sentence problem), got bleu=27.88; moses bleu=32.47
  • try more models to get similar results as original baseline on new data
  • m-nmt model on new data
Shipan Ren
  • learn the implement of seq2seq model
  • read tf_translate code
  • understand the meaning of main code
  • start writing documents