提交 · 39d6a985bc54de58cf567334e119f1c8fae572f1 · 机器未来 / Paddle

11 4月, 2019 8 次提交
- R
  fix some comments, include cosine_decay,l2_normalize,pixel_shuffle (#16763) · 39d6a985
  由 ruri 提交于 4月 11, 2019
```
* fix some comments, include cosine_decay,l2_normalize,pixel_shuffle

* Add api.spec, test=develop

* update api.spec, test=develop

* add api.spec,test=develop

* test=develop

* test=develop

* fix conflict,test=develop
```
  39d6a985
- G
  Merge pull request #16765 from guru4elephant/gpu_dataset_train · aa46caf3
  由 guru4elephant 提交于 4月 11, 2019
```
add gpu training for Executor.train_from_dataset
```
  aa46caf3
- W
  
  disable the share lib for protobuf test=develop (#16778) · b6150e1f
  由 wopeizl 提交于 4月 11, 2019
  
  b6150e1f
- W
  
  remove append_LARS not used api test=develop (#16703) · 8b587320
  由 Wu Yi 提交于 4月 11, 2019
  
  8b587320
- T
  Merge pull request #16738 from luotao1/high_level_api_test · f96446ca
  由 Tao Luo 提交于 4月 11, 2019
```
reduce CI time of high_level_api tests
```
  f96446ca
- C
  Make test_parallel_executor_seresnet.py Faster (#16701) · 610c6442
  由 chengduo 提交于 4月 10, 2019
```
* slimming test_parallel_executor_seresnet.py
```
  610c6442
- D
  remove all warnings · 3c2d2368
  由 dongdaxiang 提交于 4月 11, 2019
```
test=develop
```
  3c2d2368
- Y
  Add an option to enable the cache of expected kernel in train phase. (#16724) · 112f1614
  由 Yiqun Liu 提交于 4月 11, 2019
```
* Add an option to enable the cache of expected kernel in train phase.
test=develop

* Change the default value of cache_expected_kernel to true.
```
  112f1614
10 4月, 2019 14 次提交
- L
  
  disable memory_optimize and inpalce strategy by default, test=develop (#16760) · 2e07c19a
  由 liuwei1031 提交于 4月 10, 2019
  
  2e07c19a
- D
  remove comment in data_feed.cc · ea07eb8c
  由 dongdaxiang 提交于 4月 10, 2019
```
develop=test
```
  ea07eb8c
- T
  add WITH_HIGH_LEVEL_API option, default OFF · 544f91de
  由 Tao Luo 提交于 4月 10, 2019
```
test=develop
```
  544f91de
- G
  
  Update dataset.py · ba98872d
  由 guru4elephant 提交于 4月 10, 2019
  
  ba98872d
- G
  
  Update nccl_context.h · e349a744
  由 guru4elephant 提交于 4月 10, 2019
  
  e349a744
- D
  add gpu training for Executor.train_from_dataset · 05464e7c
  由 dongdaxiang 提交于 4月 10, 2019
```
test=develop
```
  05464e7c
- T
  remove non-existent test_image_classification_resnet · 7d0ed2a4
  由 Tao Luo 提交于 4月 10, 2019
```
test=develop
```
  7d0ed2a4
- Z
  Fix unittests which takes too long time (#16713) · 674aed6a
  由 Zeng Jinle 提交于 4月 10, 2019
```
* fix too long unittest
recommit
test=develop

* add fake_reader.py
test=develop
```
  674aed6a
- G
  
  Fix grpc log message. (#16735) · bf606bce
  由 gongweibao 提交于 4月 10, 2019
  
  bf606bce
- J
  
  test=develop, refine transformer (#16734) · 7060c8d8
  由 Jiabin Yang 提交于 4月 10, 2019
  
  7060c8d8
- Z
  fix activation grad op desc maker (#16715) · 9f7b027d
  由 Zeng Jinle 提交于 4月 09, 2019
```
test=develop
```
  9f7b027d
- L
  Merge pull request #16561 from junjun315/move-api-to-root · 9bd44b94
  由 lujun 提交于 4月 10, 2019
```
Move dygraph api to root
```
  9bd44b94
- G
  
  Fix dgc spelling mistakes. (#16733) · 95410652
  由 gongweibao 提交于 4月 10, 2019
  
  95410652
- W
  
  modify the build script for new ci test=develop (#16732) · 00279fdc
  由 wopeizl 提交于 4月 10, 2019
  
  00279fdc
09 4月, 2019 11 次提交
- L
  
  avoid optimize variable used in subblock, test=develop (#16739) · fdb719a1
  由 liuwei1031 提交于 4月 09, 2019
  
  fdb719a1
- K
  Merge pull request #16439 from heavengate/resize_scale · ed971564
  由 Kaipeng Deng 提交于 4月 09, 2019
```
add attr scale. test=develop
```
  ed971564
- T
  Merge pull request #16725 from tensor-tang/pass/disable_seqpool · 1a21d08f
  由 Tao Luo 提交于 4月 09, 2019
```
disable seqpool concat pass by default saving CI time
```
  1a21d08f
- L
  only use the latest version variable for inplace strategy (#16736) · a18ef10c
  由 liuwei1031 提交于 4月 09, 2019
```
* bug-fix, test=develop

* tweak code, test=develop
```
  a18ef10c
- T
  rename high_level_api tests · 38f01b67
  由 Tao Luo 提交于 4月 09, 2019
```
test=develop
```
  38f01b67
- H
  Fix op registry (#16677) · 2146293d
  由 Huihuang Zheng 提交于 4月 09, 2019
```
list of fixed ops:
lookup_table_op
space_to_depth_op
squared_l2_distance_op
squared_l2_norm_op
teacher_student_sigmoid_loss_op
tree_conv_op
warpctc_op

test=develop
```
  2146293d
- T
  Merge pull request #16711 from luotao1/has_attr · 5c364cda
  由 Tao Luo 提交于 4月 09, 2019
```
reduce hasAttr elapsed time in RunImpl
```
  5c364cda
- T
  disable seqpool concat pass by default saving CI time · d6c1b5a7
  由 tensor-tang 提交于 4月 09, 2019
```
test=develop
```
  d6c1b5a7
- L
  
  merge conflict, test=develop · 14db0680
  由 lujun 提交于 4月 09, 2019
  
  14db0680
- B
  
  fix training validation test=develop (#16698) · 1c8b34dd
  由 baojun 提交于 4月 08, 2019
  
  1c8b34dd
- L
  
  merge conflict, test=develop · 92c8ac8a
  由 lujun 提交于 4月 09, 2019
  
  92c8ac8a
08 4月, 2019 7 次提交
- J
  make less batch of tests to fit ci (#16706) · a06f4b2b
  由 Jiabin Yang 提交于 4月 08, 2019
```
* make less batch of tests to fit ci

* test=develop, invoke ci

add some comments back
```
  a06f4b2b
- C
  Add unit test for fuse all_reduce ops (#16699) · 55b15db5
  由 chengduo 提交于 4月 08, 2019
```
* test fuse all_reduce
```
  55b15db5
- L
  reduce hasAttr elapsed time in RunImpl · 4098ba29
  由 luotao1 提交于 4月 08, 2019
```
test=develop
```
  4098ba29
- L
  
  Merge branch 'develop' into has_attr · f89a9c5d
  由 luotao1 提交于 4月 08, 2019
  
  f89a9c5d
- T
  Merge pull request #16339 from luotao1/core_opt_choose_kernel · ad4a1bd1
  由 Tao Luo 提交于 4月 08, 2019
```
Cache the chosen kernel of operators
```
  ad4a1bd1
- L
  reduce hasAttr elapsed time in RunImpl · 6afc97ca
  由 luotao1 提交于 4月 08, 2019
```
test=develop
```
  6afc97ca
- Y
  
  disable reuse port test=develop (#16704) · 55e3c694
  由 Yan Xu 提交于 4月 08, 2019
  
  55e3c694

机器未来 / Paddle 与 Fork 源项目一致

机器未来 / Paddle
与 Fork 源项目一致