NLP Status Report 2017-4-5 - 版本历史

2017年4月5日 (三) 02:16 Fengyang

2017-04-05T02:16:12Z

2017年4月5日 (三) 02:15 Fengyang

2017-04-05T02:15:21Z

2017年4月5日 (三) 02:14 Fengyang

2017-04-05T02:14:59Z

2017年4月5日 (三) 02:14 Fengyang

2017-04-05T02:14:30Z

2017年4月5日 (三) 02:07 Zhangandy

2017-04-05T02:07:52Z

2017年4月5日 (三) 02:04 Zhangjy

2017-04-05T02:04:09Z

2017年4月5日 (三) 02:02 Zhangsy

2017-04-05T02:02:27Z

2017年4月5日 (三) 02:01 Zhangsy

2017-04-05T02:01:54Z

Fengyang：以“{| class="wikitable" !Date !! People !! Last Week !! This Week |- | rowspan="6"|2017/3/27 |Yang Feng || *tested for the baseline but cannot get the reasonable result...”为内容创建页面

2017-04-05T01:58:28Z

以“{| class="wikitable" !Date !! People !! Last Week !! This Week |- | rowspan="6"|2017/3/27 |Yang Feng || *tested for the baseline but cannot get the reasonable result...”为内容创建页面

新页面

{| class="wikitable"
!Date !! People !! Last Week !! This Week
|-
| rowspan="6"|2017/3/27
|Yang Feng ||
*tested for the baseline but cannot get the reasonable result.
*debug the baseline to try to reproduce the good result but failed.
*fixed the problem of nan in alpha-gamma method but the result is not good.
*changed the calculation of probability for alpha-gamma method but the result is neither good.
*ran Moses for cwmt zh-en translation, but the training data is case-sensitive, so need to rerun.
||
*rerun Moses for cwmt zh-en and cs-en
*decide to use tensorflow or theano
*run experiments based on the chosen platform
|-
|Jiyuan Zhang ||
* I did nothing (I found my ACL mark was at borderline, so I didn't have a mind to work)
||
*improve the effect of the qx's model
|-
|Andi Zhang ||
*fixed the bug, turns out it rises from the unfamiliarity with numpy.resize() function
*the demo model can deal with oov problem(both source word and target word are oov)
||
*some paper work about graduation design
*run some experiments using theano on old data set and new zh2en from lihang
|-
|Shiyue Zhang ||
* got good results from gnmt
* but haven't found the way to implement our model on gnmt
* trying to modify our code to make it work on big data
||
* go on trying to modify our code to make it work on big data
* go on looking into gnmt code
|-
|Peilun Xiao ||

||

|}

@@ 第5行： / 第5行： @@
 |Yang Feng ||
 * Got the sampled 100w good data and ran Moses (BLEU: 30.6)
-* Reimplemented the idea of ACL (added some optimization to the previous code) and check the performance in the following gradual steps: 1. use s\_i-1 as memory query; 2. use s\_i-1+c\_i as memory query; 3. use y as the memory states for attention; 4. use y + smt_attentions * h as memory states for attention.
+* Reimplemented the idea of ACL (added some optimization to the previous code) and check the performance in the following gradual steps: 1. use s_i-1 as memory query; 2. use s_i-1+c_i as memory query; 3. use y as the memory states for attention; 4. use y + smt_attentions * h as memory states for attention.
 * ran experiments for the above steps but the loss was inf. I am looking for reasons.
 ||

@@ 第5行： / 第5行： @@
 |Yang Feng ||
 * Got the sampled 100w good data and ran Moses (BLEU: 30.6)
-* Reimplemented the idea of ACL (added some optimization to the previous code) and check the performance in the following gradual steps: 1. use s\_i-1 as memory query; 2. use s\_i-1+c\_i
+* Reimplemented the idea of ACL (added some optimization to the previous code) and check the performance in the following gradual steps: 1. use s\_i-1 as memory query; 2. use s\_i-1+c\_i as memory query; 3. use y as the memory states for attention; 4. use y + smt_attentions * h as memory states for attention.
- as memory query; 3. use y as the memory states for attention; 4. use y + smt_attentions * h as memory states for attention.
 * ran experiments for the above steps but the loss was inf. I am looking for reasons.
 ||

@@ 第5行： / 第5行： @@
 |Yang Feng ||
 * Got the sampled 100w good data and ran Moses (BLEU: 30.6)
-* Reimplemented the idea of ACL (added some optimization to the previous code) and check the performance in the following gradual steps: 1. use s_i-1 as memory query; 2. use s_i-1+c_i
+* Reimplemented the idea of ACL (added some optimization to the previous code) and check the performance in the following gradual steps: 1. use s\_i-1 as memory query; 2. use s\_i-1+c\_i
   as memory query; 3. use y as the memory states for attention; 4. use y + smt_attentions * h as memory states for attention.
 * ran experiments for the above steps but the loss was inf. I am looking for reasons.

@@ 第29行： / 第29行： @@
 * got a reasonable baseline on big zhen data
 ||
-* implement mem model on this baseline, test on big data
+* implement mem model on this baseline, and test on big data
-* implement
 |-
 |Peilun Xiao ||

@@ 第2行： / 第2行： @@
 !Date !! People !! Last Week !! This Week
 |-
-| rowspan="6"|2017/3/27
+| rowspan="6"|2017/4/5
 |Yang Feng ||
-*tested for the baseline but cannot get the reasonable result.
+* Got the sampled 100w good data and ran Moses (BLEU: 30.6)
-*debug the baseline to try to reproduce the good result but failed.
+* Reimplemented the idea of ACL (added some optimization to the previous code) and check the performance in the following gradual steps: 1. use s_i-1 as memory query; 2. use s_i-1+c_i
-*fixed the problem of nan in alpha-gamma method but the result is not good.
+ as memory query; 3. use y as the memory states for attention; 4. use y + smt_attentions * h as memory states for attention.
-*changed the calculation of probability for alpha-gamma method but the result is neither good.
+* ran experiments for the above steps but the loss was inf. I am looking for reasons.
-*ran Moses for cwmt zh-en translation, but the training data is case-sensitive, so need to rerun.
 ||
-*rerun Moses for cwmt zh-en and cs-en
+*do experiments and write the paper
-*decide to use tensorflow or theano
-*run experiments based on the chosen platform
 |-
 |Jiyuan Zhang ||

@@ 第22行： / 第22行： @@
 |-
 |Andi Zhang ||
-*fixed the bug, turns out it rises from the unfamiliarity with numpy.resize() function
+*revise the original oov model so that it can automatically detect oov words and translate them
-*the demo model can deal with oov problem(both source word and target word are oov)
+*deal with the situation that source word is oov but target word is not oov first
 ||
-*some paper work about graduation design
+*make the model work as what we wanted
-*run some experiments using theano on old data set and new zh2en from lihang
+*deal with the situation that source word is oov and target word is also oov, then other situations
 |-
 |Shiyue Zhang ||

@@ 第15行： / 第15行： @@
 |-
 |Jiyuan Zhang ||
-* I did nothing (I found my ACL mark was at borderline, so I didn't have a mind to work)
+* I did keyword expansion on the qx's model
 ||
 *improve the effect of the qx's model

@@ 第27行： / 第27行： @@
 |-
 |Shiyue Zhang ||
-* got good results from gnmt
+* got a reasonable baseline on big zhen data
-* but haven't found the way to implement our model on gnmt
-* trying to modify our code to make it work on big data
 ||
-* go on trying to modify our code to make it work on big data
+* implement mem model on this baseline, test on big data
-* go on looking into gnmt code
+* implement
 |-
 |Peilun Xiao ||