提交 · fcd74e06b8f8ed1e7cd13a0255f207f25e638992 · s920243400 / PaddleDetection

24 10月, 2017 2 次提交

add book04.word2vec train test (#5002) · fcd74e06

由 QI JUN 提交于 10月 23, 2017

* init

* ensure ids in lookup table op must be a column vector

* add book4 configuration in test_layers

* debug test_book4

* add test_word2vec

* follow comments

* follow comments

fcd74e06

Q
ensure ids in lookup table op must be a column vector (#4987) · 40e7caf6
由 QI JUN 提交于 10月 23, 2017
```
* ensure ids in lookup table op must be a column vector

* follow comments
```
40e7caf6

23 10月, 2017 16 次提交
- C
  Merge pull request #5015 from ranqiu92/develop · 7d653c41
  由 Cao Ying 提交于 10月 23, 2017
```
Fix a bug of dot-product attention.
```
  7d653c41
- R
  
  Fix bugs of dot-product attention · 4ad12a0b
  由 ranqiu 提交于 10月 23, 2017
  
  4ad12a0b
- Q
  Merge pull request #4929 from qingqing01/lstm · 3f1062d7
  由 qingqing01 提交于 10月 23, 2017
```
Forward implementation for LSTM operator.
```
  3f1062d7
- D
  
  update to the develop branch. · cf2608e3
  由 dangqingqing 提交于 10月 23, 2017
  
  cf2608e3
- D
  
  Update lstm comments and fix bug. · 64fe9bcc
  由 dangqingqing 提交于 10月 23, 2017
  
  64fe9bcc
- 武
  Merge pull request #4972 from typhoonzero/fix_v2_optimizer_order · 154e1d04
  由武毅提交于 10月 23, 2017
```
Fix v2 optimizer define order
```
  154e1d04
- T
  
  revert local changes · 6942eb2c
  由 typhoonzero 提交于 10月 23, 2017
  
  6942eb2c
- T
  
  add ut, follow comments · 023166a8
  由 typhoonzero 提交于 10月 23, 2017
  
  023166a8
- T
  Merge pull request #5010 from tensor-tang/merge · 48173e85
  由 Tao Luo 提交于 10月 23, 2017
```
fix crash on ds2
```
  48173e85
- T
  
  fix crash on ds2 · 386bb1fe
  由 tensor-tang 提交于 10月 23, 2017
  
  386bb1fe
- C
  Merge pull request #4968 from luotao1/sparse_vector · 43d69818
  由 Cao Ying 提交于 10月 23, 2017
```
rename sparse_vector to sparse_float_vector, and fix typo.
```
  43d69818
- L
  
  rename sparse_vector to sparse_float_vector in tests · 70d9d953
  由 Luo Tao 提交于 10月 23, 2017
  
  70d9d953
- L
  
  Merge branch 'develop' into sparse_vector · c2feab7f
  由 Luo Tao 提交于 10月 23, 2017
  
  c2feab7f
- T
  Merge pull request #4953 from tensor-tang/merge_grad_gtest · abce9eb7
  由 Tao Luo 提交于 10月 23, 2017
```
refine the mkldnn logic
```
  abce9eb7
- Q
  Merge pull request #3 from reyoung/pr/4929 · 34aac18c
  由 qingqing01 提交于 10月 22, 2017
```
Several Enhancement
```
  34aac18c
- Q
  CompileTime InferShape should find var recursively in stack of blocks (#4998) · c91de280
  由 Qiao Longfei 提交于 10月 22, 2017
```
* recursive find var in BlockDesc

* add HasVarRecursive and FindVarRecursive to BlockDesc

* fix FindVarRecursive
```
  c91de280
22 10月, 2017 2 次提交
- Q
  
  use context to get attribute (#4997) · 54ffafa1
  由 Qiao Longfei 提交于 10月 21, 2017
  
  54ffafa1
- Q
  
  fix InferShapeContext Has interface (#4994) · e7f62703
  由 Qiao Longfei 提交于 10月 21, 2017
  
  e7f62703
21 10月, 2017 16 次提交
- G
  Merge pull request #4496 from guoshengCS/add-NormOp · d0cfbba4
  由 Guo Sheng 提交于 10月 21, 2017
```
Add norm_op
```
  d0cfbba4
- K
  Merge pull request #4977 from kexinzhao/python_adagrad · 5fd4bee2
  由 kexinzhao 提交于 10月 20, 2017
```
Adding interface for the adagrad optimizer
```
  5fd4bee2
- Y
  Merge pull request #4990 from tonyyang-svail/add_optional_in_op_proto · fa050941
  由 Yang Yang(Tony) 提交于 10月 20, 2017
```
add optional in op proto
```
  fa050941
- Y
  
  add optional in op proto · c030589d
  由 Yang Yang 提交于 10月 20, 2017
  
  c030589d
- D
  Merge pull request #4985 from reyoung/add_nccl_to_docker_image · ac4f7598
  由 dzhwinter 提交于 10月 20, 2017
```
Add nccl to docker image
```
  ac4f7598
- Y
  
  Add nccl to docker image · 784fc32b
  由 Yu Yang 提交于 10月 21, 2017
  
  784fc32b
- Y
  
  Correct the dependencies (#4978) · e9e0d7d7
  由 Yu Yang 提交于 10月 20, 2017
  
  e9e0d7d7
- Y
  
  Global function, op_support_gpu (#4980) · 86437a8d
  由 Yu Yang 提交于 10月 20, 2017
  
  86437a8d
- K
  
  add default value to epsilon · 7256cbe1
  由 Kexin Zhao 提交于 10月 20, 2017
  
  7256cbe1
- Y
  
  Several Enhancement · 65906ef1
  由 Yu Yang 提交于 10月 20, 2017
  
  65906ef1
- Q
  Merge pull request #4973 from QiJune/clang-build-error · d2f3c8bb
  由 QI JUN 提交于 10月 20, 2017
```
fix clang build error
```
  d2f3c8bb
- Y
  
  change lod tensor to absolute offsets (#4952) · db7b1171
  由 Yan Chunwei 提交于 10月 20, 2017
  
  db7b1171
- Q
  
  fix clang build error · 7edc1d96
  由 qijun 提交于 10月 20, 2017
  
  7edc1d96
- Q
  Merge pull request #4956 from QiJune/test_book2 · 1680903d
  由 QI JUN 提交于 10月 20, 2017
```
add book02.recognize_digits mlp train test
```
  1680903d
- F
  Trainable conv net of MNIST (#4960) · 05ece848
  由 fengjiayi 提交于 10月 20, 2017
```
* Init file

* Update

* Update

* Complete conv net of MNIST
```
  05ece848
- Y
  
  feature/dynamic recurrent op forward and backward (#4799) · 07ea9ade
  由 Yan Chunwei 提交于 10月 20, 2017
  
  07ea9ade
20 10月, 2017 4 次提交
- K
  
  Adding Nesterov Momentum (#4948) · 5380a547
  由 kavyasrinet 提交于 10月 20, 2017
  
  5380a547
- T
  
  update · 71c2b296
  由 typhoonzero 提交于 10月 20, 2017
  
  71c2b296
- T
  remove unused code · 5c892db6
  由 tensor-tang 提交于 10月 20, 2017
```
refine comments and bias
fix typo and todo
```
  5c892db6
- G
  
  Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into add-NormOp · b12988f0
  由 guosheng 提交于 10月 20, 2017
  
  b12988f0

s920243400 / PaddleDetection 与 Fork 源项目一致

s920243400 / PaddleDetection
与 Fork 源项目一致