|
|
| 第58行： |
第58行： |
| | * refine the models tried before and give the result report | | * refine the models tried before and give the result report |
| | * finish the code of dynamic memory model | | * finish the code of dynamic memory model |
| − | || || | + | || |
| | + | * implemented on MKL successfully |
| | + | * reran original model |
| | + | * reran the models tried before, but still running |
| | + | * finished the code of dynamic memory model, ran and got a result |
| | + | * tried another structure of memory |
| | + | || |
| | + | * 90% |
| | |- | | |- |
| | | 2016/11/21 | | | 2016/11/21 |
| | | | | | |
| − | * modify model | + | * try more different models |
| − | * try to prove the positive function of dynamic memory
| + | # rerun original model, if the same result |
| | + | # run more dynamic memory models, hope to see better results |
| | + | # run models with another structure of memory, if it is better than previous structure |
| | + | * summary experiments' results and give report |
| | + | * publish the TRPs of RNNG |
| | || || | | || || |
| | |- | | |- |
2016年11月21日 (一) 06:55的最后版本
Main Idea
People
Yang Feng, Shiyue Zhang, Andi Zhang
Time Table
| Date |
Work Plan |
Work Done |
Completion Rate
|
| 2016/10/31
|
- implement rnng+static memory discriminative model
- fix the unexpected action
- rerun the original discriminative model
- rerun the centred memory rnng model
- get wrong instances of original trained model, and get statistics
- run the wrong memory rnng model
- run the sampled memory rnng model
- update experiment report
|
- implementation is done, but result is not satisfied.
- fixed the unexpected action
- reran the original discriminative model
- reran the centred memory rnng model
- got wrong instances of original trained model, and get statistics
- ran the sampled memory rnng model
- ran the wrong memory rnng model
- updated experiment report [1]
|
|
| 2016/11/7
|
- modify discriminative model, try to prove the positive function of static memory
- change train set
- add parameter before cos
- modify model structure
- read the code of Teacher Feng
- understand and run
- learn to use it
|
- modify discriminative model, try to prove the positive function of static memory
- changed train set
- added parameter before cos
- modified model structure
- read the code of Teacher Feng
- understood but did't run
- the result is rnng cannot run fast on GPU
|
|
| 2016/11/14
|
- try to run rnng on multi cpu cores
- refine the models tried before and give the result report
- finish the code of dynamic memory model
|
- implemented on MKL successfully
- reran original model
- reran the models tried before, but still running
- finished the code of dynamic memory model, ran and got a result
- tried another structure of memory
|
|
| 2016/11/21
|
- try more different models
- rerun original model, if the same result
- run more dynamic memory models, hope to see better results
- run models with another structure of memory, if it is better than previous structure
- summary experiments' results and give report
- publish the TRPs of RNNG
|
|
|
| 2016/11/28
|
- modify model
- try to prove the positive function of dynamic memory
|
|
|
| 2016/12/5
|
- get the first final rnng+mm discriminative model
|
|
|
| 2016/12/12
|
- give a plan to transfer to generative model
|
|
|
| 2016/12/19
|
- implement rnng+mm generative model
|
|
|
| 2016/12/26
|
- modify model
- try to prove the positive function of memory
|
|
|
Progress
