提交 · 80a3fd2eae795a30f8835cf7bb05c523010181a6 · s920243400 / PaddleDetection

13 6月, 2019 1 次提交

Cherry-pick of #17814 and #18030 (#18067) · 80a3fd2e

由 Wojciech Uss 提交于 6月 13, 2019

Added unit test for QAT FP32 & INT8 comparison (#17814)
Disable MKLDNN FC in Resnet50 test (#18030)

test=release/1.5

80a3fd2e

06 6月, 2019 2 次提交
- M
  
  [NGraph] Bert model for a capi, ngraph's support test=develop (#17844) · c1379bf2
  由 mozga-intel 提交于 6月 06, 2019
  
  c1379bf2
- Z
  fix: when use the load model from memory mode, the RAM occupy is high (#17788) · ae576f3c
  由 Zhaolong Xing 提交于 6月 06, 2019
```
test=develop
```
  ae576f3c
29 5月, 2019 3 次提交
- T
  add fc_mkldnn_pass in compare_mkldnn (#17712) · b4b16946
  由 Tao Luo 提交于 5月 29, 2019
```
test=develop
```
  b4b16946
- Z
  fix trt ci timeout error (#17701) · 4337009b
  由 Zhaolong Xing 提交于 5月 29, 2019
```
test=develop
```
  4337009b
- M
  
  Capi for a ngraph engine (#17037) · 5eb81fe5
  由 mozga-intel 提交于 5月 28, 2019
  
  5eb81fe5
28 5月, 2019 1 次提交

Improve mobilenetv2 INT8 performance by using INT8 relu as post-op (#17570) · 04b6c29e

由 lidanqing 提交于 5月 28, 2019

* add INT8 conv+relu6 fuse and enbale mobilentv2 INT8 test
test=develop

* change fasle and 0.0 to fuse_brelu and brelu_threshold
test=develop

change the "fuse_relu||fuse_brelu" to "unsigned_output"
test=develop

* Use relu instead of brelu as INT8 post-op because INT8 brelu is not enabled in mkldnn v0.18
test=develop

* continuous-integration fix
test=develop

04b6c29e

24 5月, 2019 2 次提交

[MKL-DNN] Add Fully Connected Op for inference only(#15226) · 0c39b97b

由 Michał Gallus 提交于 5月 24, 2019

* fuse mul and elementwise add to fc

* Reimplement the FC forward operator

* Fix FC MKLDNN integration by transposing weights

* Add FC MKLDNN Pass

test=develop

* FC MKLDNN Pass: change memcpy to std::copy

* Fix MKLDNN FC handling of mismatch input and weights dims

* Lower tolerance for MKL-DNN in resnet50 test

test=develop

* Adjust FC to support MKLDNN Op placement

test=develop

* Adjust Placement Op to set use_mkldnn attribute for graph

test=develop

* MKLDNN FC: fix weights format so that gemm version is called

test=develop

* FC MKLDNN: Remove tolerance decrease from tester_helper

* FC MKL-DNN: Refactor the code, change input reorder to weight reorder

* MKL-DNN FC: Introduce operator caching

test=develop

* FC MKL-DNN: Fix the tensor type in ExpectedKernelType

test=develop

* FC MKL-DNN: fix style changes

test=develop

* FC MKL-DNN: fallback to native on non-supported dim sizes

test=develop

* FC MKLDNN: fix CMake paths

test=develop

* FC MKLDNN: Refine placement pass graph mkldnn attribute

test=develop

* Fix Transpiler error for fuse_conv_eltwise

test=develop

* Fix missing STL includes in files

test=develop

* FC MKL-DNN: Enable new output size computation

Also, refine pass to comply with newest interface.
test=develop

* FC MKL-DNN: enable only when fc_mkldnn_pass is enabled

* FC MKL-DNN: Allow Weights to use oi or io format

* FC MKL-DNN: Adjust UT to work with correct dims

test=develop

* Enable MKL DEBUG for resnet50 analyzer

test=develop

* FC MKL-DNN: Improve Hashing function

test=develop

* FC MKL-DNN: Fix shape for fc weights in transpiler

* FC MKL-DNN: Update input pointer in re-used fc primitive

* Add log for not handling fc fuse for unsupported dims

test=develop

* FC MKL-DNN: Move transpose from pass to Op Kernel

test=develop

* FC MKL-DNN: Disable transpose in unit test

test=develop

* FC MKL-DNN: Remove fc_mkldnn_pass from default list

* Correct Flag for fake data analyzer tests

test=develop

* FC MKL-DNN: Add comment about fc mkldnn pass disablement

test=develop

* FC MKL-DNN: Disable fc in int8 tests

test=develop

0c39b97b

fix quantize_squash_pass segfault when no tensor linked to Bias (#17292) · bccb0ba4

由 Sylwester Fraczek 提交于 5月 24, 2019

* fix quantize_squash_pass segfault when there is no tensor linked do Bias input

test=develop

* add googlenet test

test=develop

* fix concat CreateKey not using input format

test=develop

bccb0ba4

22 5月, 2019 1 次提交
- L
  fix bug that saved optimal model path in test_analyzer_save_model con… (#17555) · daf88968
  由 lijianshe02 提交于 5月 22, 2019
```
* modify saved model path in analyzer_save_model.cc test=develop
```
  daf88968
21 5月, 2019 2 次提交
- T
  remove unused SERIAL compiler option (#17500) · 3d19f44a
  由 Tao Luo 提交于 5月 21, 2019
```
test=develop
```
  3d19f44a
- L
  Enabling resnet101, vgg16, vgg19 INT8v2 model tests (#17468) · 36757ed2
  由 lidanqing 提交于 5月 21, 2019
```
* Add 6 models tests support in CMake

* enabling resnet101, vgg16, vgg19 INT8v2 model tests
test=develop

* remove SERIAL
test=develop
```
  36757ed2
15 5月, 2019 1 次提交
- F
  bug fix (#17392) · e48dd92f
  由 flame 提交于 5月 15, 2019
```
fix secure bug
```
  e48dd92f
08 5月, 2019 1 次提交
- W
  improved unit test output (#17266) · 984aa905
  由 Wojciech Uss 提交于 5月 08, 2019
```
added printing data type to differentiate int8 and fp32 latency results

test=develop
```
  984aa905
07 5月, 2019 1 次提交

call SetNumThreads everytime to avoid missing omp thread setting (#17224) · 54636a19

由 Leo Zhao 提交于 5月 07, 2019

* call SetNumThreads everytime to avoid missing omp thread setting

resolve #17153
test=develop

* add paddle_num_threads into config for test_analyzer_pyramid_dnn

resolve #17153
test=develop

54636a19

05 5月, 2019 1 次提交
- W
  
  use two GPUs to run the exclusive test test=develop (#17187) · 83c4f772
  由 wopeizl 提交于 5月 05, 2019
  
  83c4f772
30 4月, 2019 1 次提交

fix bn fuse vardesc and add model saver (#17143) · 79ed1c76

由 tensor-tang 提交于 4月 30, 2019

* fix bn fuse vardesc and add model saver

test=develop

* unify save model in test helper

test=develop

* fix mkdir on windows

test=develop

* remove magic number use bn bias var desc

test=develop

79ed1c76

23 4月, 2019 1 次提交
- L
  fix runtime_context_cache bug when gpu model has an op runs only on cpu · 490e7462
  由 luotao1 提交于 4月 23, 2019
```
test=develop
```
  490e7462
22 4月, 2019 1 次提交

add parallel build script to ci … (#16901) · d9991dcc

由 wopeizl 提交于 4月 22, 2019

* add parallel build script to ci test=develop
* 1. classify the test case as single card/two cards/multiple cards type
   2. run test case according to the run type

d9991dcc

19 4月, 2019 1 次提交
- T
  disable runtime_context_cache pass by default · aa7b975b
  由 Tao Luo 提交于 4月 19, 2019
```
test=develop
```
  aa7b975b
17 4月, 2019 1 次提交
- T
  use multi-thread to speedup CI tests · bc037c13
  由 Tao Luo 提交于 4月 17, 2019
```
test=develop
```
  bc037c13
15 4月, 2019 3 次提交
- R
  minus trt ci times. · 1965a224
  由 root 提交于 4月 15, 2019
```
test=develop
```
  1965a224
- L
  add SaveOptimModel interface in analysis_predictor.h and test it in a… (#16441) · de26df44
  由 lijianshe02 提交于 4月 15, 2019
```
* add SaveOptimModel interface in analysis_predictor.h and test it in analyzer_dam_tester and analyzer_resnet50_tester test=develop
```
  de26df44
- L
  improve preprocess script and read from tar · de02d40e
  由 lidanqing 提交于 4月 15, 2019
```
test=develop
```
  de02d40e
12 4月, 2019 1 次提交

Fix the order while sorting the operators (#16756) · 93cedfdb

由 Yihua Xu 提交于 4月 12, 2019

* Fix the order when sorting operators.

test=develop

* Enable transfomer compare test item.

test=develop

* Use set to replace vector.

test=develop

93cedfdb

11 4月, 2019 1 次提交

Security issue (#16774) · 85363848

由 liuwei1031 提交于 4月 11, 2019

* disable memory_optimize and inpalce strategy by default, test=develop

* fix security issue
http://newicafe.baidu.com:80/issue/PaddleSec-3/show?from=page
http://newicafe.baidu.com:80/issue/PaddleSec-8/show?from=page
http://newicafe.baidu.com:80/issue/PaddleSec-12/show?from=page
http://newicafe.baidu.com:80/issue/PaddleSec-32/show?from=page
http://newicafe.baidu.com:80/issue/PaddleSec-35/show?from=page
http://newicafe.baidu.com:80/issue/PaddleSec-37/show?from=page
http://newicafe.baidu.com:80/issue/PaddleSec-40/show?from=page
http://newicafe.baidu.com:80/issue/PaddleSec-43/show?from=page
http://newicafe.baidu.com:80/issue/PaddleSec-44/show?from=page
http://newicafe.baidu.com:80/issue/PaddleSec-45/show?from=page

test=develop

* revert piece.cc, test=develop

* adjust api.cc,test=develop

85363848

09 4月, 2019 1 次提交
- T
  disable seqpool concat pass by default saving CI time · d6c1b5a7
  由 tensor-tang 提交于 4月 09, 2019
```
test=develop
```
  d6c1b5a7
04 4月, 2019 2 次提交
- T
  reduce all analyzer_test ci elasped time · d5c8d4ac
  由 Tao Luo 提交于 4月 04, 2019
```
test=develop
```
  d5c8d4ac
- B
  
  MKLDNN INT8 v2 readme.md (#16515) · 88ceda51
  由 bingyanghuang 提交于 4月 04, 2019
  
  88ceda51
03 4月, 2019 1 次提交
- L
  test_analyzer_int8 tests use default pass order · bd636a9e
  由 luotao1 提交于 4月 03, 2019
```
test=develop
```
  bd636a9e
02 4月, 2019 2 次提交
- W
  
  fix dataset reading and add support for full dataset (#16559) · 9b6a0296
  由 Wojciech Uss 提交于 4月 02, 2019
  
  9b6a0296
- L
  fix preprocess script with processbar, integrity check and logs (#16608) · 2ca0de3c
  由 lidanqing 提交于 4月 02, 2019
```
* fix preprocess script with processbar, integrity check and logs
test=develop

* delete unnecessary empty lines, change function name
test=develop
```
  2ca0de3c
01 4月, 2019 1 次提交
- T
  enhance analyzer_tests download · ce187104
  由 Tao Luo 提交于 4月 01, 2019
```
test=develop
```
  ce187104
29 3月, 2019 2 次提交
- W
  remove profiling from int8 test · 24983951
  由 Wojciech Uss 提交于 3月 29, 2019
```
test=develop
```
  24983951
- L
  fix some bugs of unzip and reading val list · 0d656996
  由 lidanqing 提交于 3月 28, 2019
```
test=develop
```
  0d656996
28 3月, 2019 5 次提交
- L
  add wget and unzip part and change data_dir · b46e467a
  由 lidanqing 提交于 3月 28, 2019
```
test=develop
```
  b46e467a
- L
  change script file name and data_dir location · 894aa9b2
  由 lidanqing 提交于 3月 28, 2019
```
test=develop
```
  894aa9b2
- L
  preprocess with PIL the full val dataset and save binary · 57f51e5b
  由 lidanqing 提交于 3月 28, 2019
```
test=develop
```
  57f51e5b
- S
  fixed url to dataset · 8ece7a97
  由 Sylwester Fraczek 提交于 3月 28, 2019
```
test=develop
```
  8ece7a97
- S
  create test for quantized resnet50 · fe21578a
  由 Sylwester Fraczek 提交于 3月 20, 2019
```
test=develop
```
  fe21578a

s920243400 / PaddleDetection 与 Fork 源项目一致

s920243400 / PaddleDetection
与 Fork 源项目一致