提交 · 8f2486ca16c0c18bf2eb3cfec9b7c74952b2cea7 · PaddlePaddle / Paddle

10 7月, 2018 2 次提交
- C
  
  change the copyright year form 2016 to 2018 · d552b900
  由 chenweihang 提交于 7月 10, 2018
  
  d552b900
- Q
  Make IfElse operator works and fix unit testing. (#11972) · ef4895df
  由 qingqing01 提交于 7月 10, 2018
```
1. Fix bug when only true or false branch works.
2. Fix bug in unit testing.
```
  ef4895df
09 7月, 2018 1 次提交
- C
  
  simplify construct function · 1721613f
  由 chenweihang 提交于 7月 09, 2018
  
  1721613f
06 7月, 2018 1 次提交
- Y
  
  Fix data balance on single GPU · 8e86721f
  由 yuyang18 提交于 7月 06, 2018
  
  8e86721f
05 7月, 2018 2 次提交
- D
  Move fluid::framework::InitDevices into fluid::platform (#11757) · 4ed0b624
  由 dzhwinter 提交于 7月 05, 2018
```
* move to platform

* "move init from framework to platform"

* "remove used init"

* "fix ci"

* "fix ci"

* "fix generic"

* "fix ci"

* "fix ci"

* "fix ci"

* "disable fragile test"
```
  4ed0b624
- D
  
  "remove lapack" (#11966) · 99a99ec7
  由 dzhwinter 提交于 7月 05, 2018
  
  99a99ec7
03 7月, 2018 5 次提交
- X
  
  fix · d70a38d8
  由 Xin Pan 提交于 7月 03, 2018
  
  d70a38d8
- Y
  Use reshape_op inside squeeze_op · 1854814d
  由 yuyang18 提交于 7月 03, 2018
```
* also convert tab to space
```
  1854814d
- X
  
  hide utils to legacy · 94cb59ad
  由 Xin Pan 提交于 7月 03, 2018
  
  94cb59ad
- C
  
  adjust the dims range to [1,6] and fix some problem · 0cef33a4
  由 chenweihang 提交于 7月 03, 2018
  
  0cef33a4
- F
  
  Add EOFException to represent EOF in C++ reader · 3fab4f65
  由 fengjiayi 提交于 7月 03, 2018
  
  3fab4f65
02 7月, 2018 9 次提交
- Y
  
  Use single file than multiple files · 550ab8d7
  由 yuyang18 提交于 7月 02, 2018
  
  550ab8d7
- P
  
  typos fix · ac323343
  由 Paweł Żelazko 提交于 7月 02, 2018
  
  ac323343
- Y
  
  Fix fc mkldnn op · 6038a631
  由 yuyang18 提交于 7月 02, 2018
  
  6038a631
- Y
  Add register kernel functor and shrink reshape op · 82866d4a
  由 yuyang18 提交于 7月 02, 2018
```
* Shrink reshape_op library size
* User can register a standard C++ functor as a op kernel
```
  82866d4a
- Y
  
  Polish reshape op · 1ce478f1
  由 yuyang18 提交于 7月 02, 2018
  
  1ce478f1
- Y
  
  update by comment · 37410a0c
  由 Yancey1989 提交于 7月 02, 2018
  
  37410a0c
- C
  
  Adjust squeeze op and code the unittest, test passed · 9ca88fa8
  由 chenweihang 提交于 6月 29, 2018
  
  9ca88fa8
- S
  
  fix type comparation bugs · dd70fb43
  由 sneaxiy 提交于 7月 02, 2018
  
  dd70fb43
- X
  FasterRCNN Anchor Generator Op (#11218) · 5056d3ec
  由 Xingyuan Bu 提交于 7月 02, 2018
```
* Add anchor generator operator for Faster-RCNN.
* Add unittest testing.
* Add Python API.
```
  5056d3ec
01 7月, 2018 4 次提交
- Y
  
  update · 029425a5
  由 Yancey1989 提交于 7月 01, 2018
  
  029425a5
- Y
  
  add dist pass barrier · 1366832a
  由 Yancey1989 提交于 7月 01, 2018
  
  1366832a
- X
  
  also move a few other dir to legacy/ · a9086bf3
  由 Xin Pan 提交于 7月 01, 2018
  
  a9086bf3
- G
  
  Improve brpccmake (#11842) · 66c91911
  由 gongweibao 提交于 6月 30, 2018
  
  66c91911
30 6月, 2018 1 次提交
- F
  
  Add unittests and fix a few bugs · 4b950951
  由 fengjiayi 提交于 6月 29, 2018
  
  4b950951
29 6月, 2018 2 次提交
- G
  Revert "Extend fill_zeros_like_op for zero-filling an LoDTensorArray (#11496)" · 6a358991
  由 guochaorong 提交于 6月 29, 2018
```
This reverts commit bc28cf61.
```
  6a358991
- C
  
  add squeeze op c++ part, compile success · 298e74da
  由 chenweihang 提交于 6月 28, 2018
  
  298e74da
28 6月, 2018 6 次提交
- F
  
  Add DataBalanceOpHandle to MultiDeviceSSAGragh · b6dc3a59
  由 fengjiayi 提交于 6月 28, 2018
  
  b6dc3a59
- M
  
  Duplicated code was moved to common function · b8a04c2f
  由 mozga-intel 提交于 6月 26, 2018
  
  b8a04c2f
- M
  
  The mkldnn batch norm supports other data format · 3b128337
  由 mozga-intel 提交于 6月 22, 2018
  
  3b128337
- X
  small AverageOptimizer enhance. (#11761) · 2ecc5622
  由 Xin Pan 提交于 6月 28, 2018
```
* small AverageOptimizer enhance.

* clean

* clean
```
  2ecc5622
- Y
  
  feature/analysis to support sub-graph for TRT engine (#11538) · 5082642b
  由 Yan Chunwei 提交于 6月 28, 2018
  
  5082642b
- H
  Extend fill_zeros_like_op for zero-filling an LoDTensorArray (#11496) · bc28cf61
  由 Haichao Zhang 提交于 6月 27, 2018
```
* Add fill_zeros_array op. This op is used for zero-filling an LoDTensorArray.

* merge fill_zeros_array_op with fill_zeros_like_op

* add unit_test for fill_zeros_like for array
```
  bc28cf61
27 6月, 2018 7 次提交
- Q
  
  adam op handle grad.rows().size == 0 condition · 20fae681
  由 qiaolongfei 提交于 6月 27, 2018
  
  20fae681
- P
  bnorm+relu fuse for mkldnn (inference) (#11434) · 9a15c923
  由 pzelazko-intel 提交于 6月 27, 2018
```
* bnorm+relu fuse for mkldnn

* separate fuse_relu function

* bug fix

* proper while range in inference_transpiler

* description fix

* review fix

* review fix

* unit test for fwd batch norm+relu MKLDNN fuse
```
  9a15c923
- B
  Optimize bipartite_match_op in large scale input (#11730) · 778b71fc
  由 baiyf 提交于 6月 27, 2018
```
* optimize bipartite_match_op in large scale input
```
  778b71fc
- Q
  
  fix adam op for selected rows · df7a266a
  由 qiaolongfei 提交于 6月 27, 2018
  
  df7a266a
- T
  
  move SetNumThreads to platform · e3a96300
  由 tensor-tang 提交于 6月 27, 2018
  
  e3a96300
- Q
  Speed depthwise transposed conv2d. (#11740) · b756063c
  由 qingqing01 提交于 6月 27, 2018
```
* Speed depthwise transposed conv2d.
```
  b756063c
- Q
  Fix sequence expand op (#11618) · 8630ba2e
  由 Qingsheng Li 提交于 6月 27, 2018
```
* Set zero outside functor
```
  8630ba2e

PaddlePaddle / Paddle 1 年多 前同步成功

PaddlePaddle / Paddle
1 年多前同步成功