提交 · a18c0d4242d88e9a67406230904375e4aa6dc153 · PaddlePaddle / PaddleDetection

29 1月, 2019 11 次提交
- T
  cache fc kernel · a18c0d42
  由 tensor-tang 提交于 1月 29, 2019
```
test=develop
```
  a18c0d42
- T
  cache softmax kernel func · 6e1ee7fb
  由 tensor-tang 提交于 1月 29, 2019
```
test=develop
```
  6e1ee7fb
- T
  Merge pull request #15563 from tensor-tang/jit/softmax · c7449227
  由 tensor-tang 提交于 1月 29, 2019
```
refine softmax kernel
```
  c7449227
- T
  Merge pull request #15570 from luotao1/bert · 245b1f05
  由 Tao Luo 提交于 1月 29, 2019
```
fix compiler error, use len20 dataset for bert
```
  245b1f05
- J
  Merge pull request #15470 from JiabinYang/feature/imperative · 075df09f
  由 Jiabin Yang 提交于 1月 29, 2019
```
Add simple RNN in imperative
```
  075df09f
- Q
  Merge pull request #15558 from velconia/imperative_resnet · b69996c2
  由 Qiyang Min 提交于 1月 29, 2019
```
Refine Batch Norm
```
  b69996c2
- L
  fix compiler error, use len20 dataset for bert · 5504425e
  由 luotao1 提交于 1月 29, 2019
```
test=develop
```
  5504425e
- Y
  
  AnalysisConfig remove contrib namespace (#15540) · 65517908
  由 Yan Chunwei 提交于 1月 29, 2019
  
  65517908
- J
  Merge pull request #15418 from jerrywgz/refine_nms · 7bc8481c
  由 jerrywgz 提交于 1月 29, 2019
```
Refine nms
```
  7bc8481c
- W
  fix default create_parameter dtype maching initializers (#15521) · ab471584
  由 Wu Yi 提交于 1月 29, 2019
```
* fix default create_parameter dtype maching initializers test=develop

* update type check test=develop

* update test=develop
```
  ab471584
- T
  refine softmax and use with cache · d59f7335
  由 tensor-tang 提交于 1月 28, 2019
```
test=develop
```
  d59f7335
28 1月, 2019 29 次提交
- T
  add softmax mix and mkl code · 7383eefd
  由 tensor-tang 提交于 1月 28, 2019
```
test=develop
```
  7383eefd
- T
  add hmax, hsum jitcode · 50945685
  由 tensor-tang 提交于 1月 28, 2019
```
test=develop
```
  50945685
- T
  add jit kernel hsum, hmax and softmax refer code · 81177258
  由 tensor-tang 提交于 1月 25, 2019
```
test=develop
```
  81177258
- T
  Merge pull request #15485 from luotao1/fc500110-bert_test · 67e4450c
  由 Tao Luo 提交于 1月 28, 2019
```
add bert analyzer test
```
  67e4450c
- Q
  Merge pull request #15312 from velconia/add_pyramid_dnn_support · 6000a6e7
  由 Qiyang Min 提交于 1月 28, 2019
```
Use malloc and free in JeMalloc
```
  6000a6e7
- J
  Merge pull request #15534 from JiabinYang/fix/multi_output_support_imperative · fd286f35
  由 Jiabin Yang 提交于 1月 28, 2019
```
test=develop, fix/multi_output_support_imperative
```
  fd286f35
- M
  Clear all parameters' gradient · 07822fef
  由 minqiyang 提交于 1月 28, 2019
```
test=develop
```
  07822fef
- Z
  Merge pull request #15545 from sneaxiy/fix_debug_nccl_error · bf7dedcb
  由 Zeng Jinle 提交于 1月 28, 2019
```
Fix nccl unittest error in debug mode
```
  bf7dedcb
- M
  Polish code · 49a7fba8
  由 minqiyang 提交于 1月 28, 2019
```
test=develop
```
  49a7fba8
- M
  Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into imperative_resnet · 159c4073
  由 minqiyang 提交于 1月 28, 2019
```
test=develop
```
  159c4073
- M
  Polish code · edf742cf
  由 minqiyang 提交于 1月 28, 2019
```
test=develop
```
  edf742cf
- M
  Fix batch_norm's stop_gradient bug · 5c776877
  由 minqiyang 提交于 1月 28, 2019
```
test=develop
```
  5c776877
- L
  Merge branch 'develop' into fc500110-bert_test · e31aef9f
  由 luotao1 提交于 1月 28, 2019
```
test=develop
```
  e31aef9f
- Q
  Always create variables in analysis_predictor before OptimizeInferenceProgram. (#15533) · a6910f90
  由 qingqing01 提交于 1月 28, 2019
```
Otherwise, some other persistable variable (like RAW type) will not be created
```
  a6910f90
- T
  Merge pull request #15530 from luotao1/remove_with_doc · 748c2d3e
  由 Tao Luo 提交于 1月 28, 2019
```
remove legacy WITH_DOC option
```
  748c2d3e
- J
  
  test=develop, use parameters() to get parameters · fff67a94
  由 JiabinYang 提交于 1月 28, 2019
  
  fff67a94
- Y
  
  add version support (#15469) · b62b756b
  由 Yan Chunwei 提交于 1月 28, 2019
  
  b62b756b
- Y
  
  infer get program (#15511) · 526790e6
  由 Yan Chunwei 提交于 1月 28, 2019
  
  526790e6
- J
  
  test=develop, merge develop · 2e309b11
  由 JiabinYang 提交于 1月 28, 2019
  
  2e309b11
- J
  
  remove test split op in imperative · 0ea7c9c1
  由 JiabinYang 提交于 1月 28, 2019
  
  0ea7c9c1
- M
  
  Fix mnist · 79d62c54
  由 minqiyang 提交于 1月 28, 2019
  
  79d62c54
- J
  
  polish code and add comments for Embedding · 3dfbef29
  由 JiabinYang 提交于 1月 28, 2019
  
  3dfbef29
- T
  Merge pull request #15537 from baojun-nervana/rm_ngraph_operator · 3c224e7e
  由 tensor-tang 提交于 1月 28, 2019
```
rm ngraph_operator.cc test=develop
```
  3c224e7e
- X
  Merge pull request #15516 from panyx0718/imperative3 · c11afdb5
  由 Xin Pan 提交于 1月 28, 2019
```
imperative supports multi grad ops
```
  c11afdb5
- S
  fix compile error in distributed mode · ba4f43fd
  由 sneaxiy 提交于 1月 28, 2019
```
test=develop
```
  ba4f43fd
- T
  Merge pull request #15531 from jczaja/prv-googlenet-fix · b9191902
  由 Tao Luo 提交于 1月 28, 2019
```
Performance and functional fixes to LRN
```
  b9191902
- H
  Enable INT8 Calibration Unit Test for MobileNet-V1 (#15539) · f8251580
  由 Haihao Shen 提交于 1月 28, 2019
```
* Enable mobilenet UT in separate test class; use download cache by paddle download utility and cache unzip; and fix typo; test=develop

* Extract cache_unzipping function for reuse; format code style; test=develop

* Simplify the test code by define a combined function for both downloading and unzipping; test=develop
```
  f8251580
- J
  
  test=develop, polish code and merge develop · 53d558cd
  由 JiabinYang 提交于 1月 28, 2019
  
  53d558cd
- Z
  Merge pull request #15242 from NHZlX/trt_int8_ultimate_version · 97b76c94
  由 Zhaolong Xing 提交于 1月 28, 2019
```
add trt int8 support
```
  97b76c94

PaddlePaddle / PaddleDetection 大约 1 年 前同步成功

PaddlePaddle / PaddleDetection
大约 1 年前同步成功