提交 · 813c2ce539dcd1f69d81a42d711a0d46e1faaf40 · PaddlePaddle / PaddleDetection

28 12月, 2018 4 次提交
- P
  
  fix timer test=develop · 813c2ce5
  由 peizhilin 提交于 12月 28, 2018
  
  813c2ce5
- W
  Refactor distributed RPC (#15075) · a8bc05b5
  由 Wu Yi 提交于 12月 28, 2018
```
* wip

* wip

* refactor no.1 dir structure test=develop

* fix linking test=develop

* fix includes test=develop

* fix build test=develop

* fix build test=develop
```
  a8bc05b5
- B
  
  upgrade ngraph to v0.10.1 test=develop · 555fbc10
  由 baojun-nervana 提交于 12月 25, 2018
  
  555fbc10
- B
  
  simplify logic test=develop · c714c364
  由 baojun-nervana 提交于 12月 25, 2018
  
  c714c364
27 12月, 2018 13 次提交
- W
  refine batch merge pass (#14777) · e26cced7
  由 Wu Yi 提交于 12月 27, 2018
```
* refine batch merge pass

* refine batch merge pass test=develop
```
  e26cced7
- S
  fix py35 compile error · 9a3a246c
  由 sneaxiy 提交于 12月 27, 2018
```
test=develop
```
  9a3a246c
- X
  register float16 · f7294f8b
  由 Xin Pan 提交于 12月 27, 2018
```
test=develop
```
  f7294f8b
- M
  Add changes back · ef7d563d
  由 minqiyang 提交于 12月 27, 2018
```
test=develop
```
  ef7d563d
- Y
  Refine Dockerfile (#14908) · efa630ea
  由 Yu Yang 提交于 12月 27, 2018
```
* Refine Dockerfile

* Add tasks, cmake gen

* Fix code error

* Disable compile after paddle_build.sh

* Refine

* Skip on PY35 CI

* Change env

* Refine paddle_build.sh

* Expose gen_fluid_lib

* Refine mkldnn.cmake

* Refine mkldnn.cmake

* Refine mkldnnlib

* Skip unstable tests
```
  efa630ea
- C
  [WIP] Refine MultiDevSSAGraph (#15040) · fe8495a7
  由 chengduo 提交于 12月 26, 2018
```
* refine parallel_exe
test=develop

* rename shared_var_device

* code refine

* add test_weight_decay

* remove Sort
test=develop

* Add SortForReduce
test=develop

* code refine
test=develop

* follow comment
test=develop
```
  fe8495a7
- X
  address comments and resolve conflicts. · c132c790
  由 Xin Pan 提交于 12月 27, 2018
```
test=develop
```
  c132c790
- X
  clear operator changes · b91a7a9d
  由 Xin Pan 提交于 12月 27, 2018
```
test=develop
```
  b91a7a9d
- X
  
  call kernel · f52b514d
  由 Xin Pan 提交于 12月 27, 2018
  
  f52b514d
- X
  fix · 4e80e04f
  由 Xin Pan 提交于 12月 26, 2018
```
test=develop
```
  4e80e04f
- X
  make fill_constant kernel-based · 7b6bf9dd
  由 Xin Pan 提交于 12月 26, 2018
```
test=develop
```
  7b6bf9dd
- X
  clean · 61491ce2
  由 Xin Pan 提交于 12月 26, 2018
```
test=develop
```
  61491ce2
- X
  refactor to avoid scope. · ce7e503c
  由 Xin Pan 提交于 12月 25, 2018
```
test=develop
```
  ce7e503c
26 12月, 2018 18 次提交
- T
  simplify some data record in analyzer_tester · ecae157e
  由 Tao Luo 提交于 12月 26, 2018
```
test=develop
```
  ecae157e
- T
  simplify prepere_input in analyzer_test · 05f1b65d
  由 Tao Luo 提交于 12月 26, 2018
```
test=develop
```
  05f1b65d
- S
  try to fix py35 compile error · ee83ce75
  由 sneaxiy 提交于 12月 26, 2018
```
test=develop
```
  ee83ce75
- S
  modify API.spec · 10a6bc96
  由 sneaxiy 提交于 12月 26, 2018
```
test=develop
```
  10a6bc96
- S
  add scope_pool · 3e917a93
  由 sneaxiy 提交于 12月 26, 2018
```
add module cleanup
test=develop
```
  3e917a93
- N
  fix comments · 02e17396
  由 nhzlx 提交于 12月 26, 2018
```
test=develop
```
  02e17396
- P
  restore the huber_loss_op · e49276e7
  由 peizhilin 提交于 12月 26, 2018
```
test=develop
```
  e49276e7
- P
  Revert "cherry-pick the #12759" · 2388d0e7
  由 peizhilin 提交于 12月 26, 2018
```
test=develop

This reverts commit 7f6d8ace.
```
  2388d0e7
- N
  add min_subgraph_size attr to tensorrt config · 71636e67
  由 nhzlx 提交于 12月 26, 2018
```
test=develop
```
  71636e67
- P
  fix test issues on windows · 01c00b07
  由 peizhilin 提交于 12月 26, 2018
```
test=develop
```
  01c00b07
- P
  add cuda dso support for windows · 1e7f83e6
  由 peizhilin 提交于 12月 26, 2018
```
test=develop
```
  1e7f83e6
- T
  code style fix, test=develop (#15045) · dc8eca82
  由 tangwei12 提交于 12月 26, 2018
```
* code style fix, test=develop
```
  dc8eca82
- N
  faster rcnn input is presistable. (fix it in paddle-trt) · a6aa8ea7
  由 nhzlx 提交于 12月 26, 2018
```
test=develop
```
  a6aa8ea7
- H
  Fix conv_elementwise_add2_act pass · 956cf921
  由 hjchen2 提交于 12月 26, 2018
```
test=develop
```
  956cf921
- W
  Make topk op support variable k. (#15044) · 2314f2eb
  由 whs 提交于 12月 26, 2018
```
* Make topk op support variable k.
test=develop

* Fix tensor type.
test=develop
```
  2314f2eb
- S
  fix conflict with develop · 179acc60
  由 sneaxiy 提交于 12月 26, 2018
```
test=develop
```
  179acc60
- D
  
  fix ci error. test=develop · 3ea2f415
  由 dzhwinter 提交于 12月 26, 2018
  
  3ea2f415
- W
  Fp16 training (#14992) · 856f0da0
  由 Wu Yi 提交于 12月 26, 2018
```
* wip

* wip

* wip

* wip for test

* add fp16 tests test=develop

* fix cpu build test=develop

* fix test=develop

* fix py3 tests test=develop

* fix lr_scheduler dtype test=develop

* fix test=dvelop

* test fix ci compile test=develop

* fix build and merge test=develop

* fallback momentumop change to general test=develop

* make fp16 lr schedule simple test=develop

* fix ut test=develop

* fix tests test=develop

* remove fp16 learning rate cast test=develop
```
  856f0da0
25 12月, 2018 5 次提交
- B
  Fix issue which cause abnormal CPU usage in stack op · e821b12f
  由 Brian Liu 提交于 12月 25, 2018
```
Stack OP has much higher CPU cost than expected in release mode.
Caused by DebugStringEx() in base class OperatorWithKernel. Actually
this issue occur for each OP which hasn't implement it's own
GetExpectedKernelType().

test=develop
```
  e821b12f
- C
  Move GetTensor to tensor_util (#15011) · b9fb03cf
  由 chengduo 提交于 12月 25, 2018
```
* refine tensor
test=develop

* refine tensor
test=develop

* fix device_context log
test=develop
```
  b9fb03cf
- T
  add seq pool inference test · d46a140d
  由 tensor-tang 提交于 12月 25, 2018
```
test=develop
```
  d46a140d
- T
  
  support more input fake data · d4931a2a
  由 tensor-tang 提交于 12月 25, 2018
  
  d4931a2a
- N
  add affine_channel fuse. · ce3782c1
  由 nhzlx 提交于 12月 25, 2018
```
fix conv+elemenwise fuse bug.
```
  ce3782c1

PaddlePaddle / PaddleDetection 大约 1 年 前同步成功

PaddlePaddle / PaddleDetection
大约 1 年前同步成功