提交 · 655179089f79718b85ebb3fd9f9ea196773ea2f6 · 机器未来 / Paddle

29 1月, 2019 1 次提交
- Y
  
  AnalysisConfig remove contrib namespace (#15540) · 65517908
  由 Yan Chunwei 提交于 1月 29, 2019
  
  65517908
28 1月, 2019 3 次提交
- 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
- 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
27 1月, 2019 1 次提交
- J
  - Compensation fix to LRN MKL-DNN op · 4aa7ef3c
  由 Jacek Czaja 提交于 1月 27, 2019
```
test=develop
```
  4aa7ef3c
26 1月, 2019 2 次提交
- N
  delete the usage of the const_cast · b43ea40c
  由 nhzlx 提交于 1月 26, 2019
```
test=develop
```
  b43ea40c
- Y
  
  add dynamic memory optim (#15457) · e2818c86
  由 Yan Chunwei 提交于 1月 26, 2019
  
  e2818c86
25 1月, 2019 1 次提交
- N
  fix comments · 92cf4a4c
  由 nhzlx 提交于 1月 25, 2019
```
test=develop
```
  92cf4a4c
24 1月, 2019 3 次提交
- N
  fix two bug: · 0779e355
  由 nhzlx 提交于 1月 24, 2019
```
1. graph and program_desc alignment
2. trt stream

test=develop
```
  0779e355
- N
  fix comments · 96413249
  由 nhzlx 提交于 1月 24, 2019
```
test=develop
```
  96413249
- N
  When cudnn version < 7100, there is problem with conv_fusion. · 484b3bc8
  由 nhzlx 提交于 1月 24, 2019
```
Add check for it.
test=develop
```
  484b3bc8
23 1月, 2019 4 次提交
- T
  fix debug compile of analysis pass fail · 5c68dee7
  由 tensor-tang 提交于 1月 23, 2019
```
test=develop
```
  5c68dee7
- L
  refine analyzer_bert_test to pass the ci · 353b5f06
  由 luotao1 提交于 1月 23, 2019
```
test=develop
```
  353b5f06
- N
  change the input to a smaller value · e6218c1d
  由 nhzlx 提交于 1月 23, 2019
```
test=develop
```
  e6218c1d
- F
  
  analyzer bert tester · 4a33a44f
  由 fuchang01 提交于 1月 22, 2019
  
  4a33a44f
22 1月, 2019 2 次提交
- N
  fix comments · 2f4aee36
  由 nhzlx 提交于 1月 22, 2019
```
test=develop
```
  2f4aee36
- N
  fix trt stream bug. · ec213730
  由 nhzlx 提交于 1月 22, 2019
```
BUG: After continuing to input different data, the output cannot be aligned
test=develop
```
  ec213730
21 1月, 2019 2 次提交
- F
  add python inference api (#15248) · d60751fb
  由 flame 提交于 1月 21, 2019
```
add python inference api
```
  d60751fb
- Y
  
  fea/infer memory optim2 (#14953) · 885c4e57
  由 Yan Chunwei 提交于 1月 21, 2019
  
  885c4e57
18 1月, 2019 2 次提交
- T
  test_analyzer_mm_dnn runs in serial · 001827c2
  由 Tao Luo 提交于 1月 18, 2019
```
test=develop
```
  001827c2
- Y
  
  get tensor API add more comments (#15345) · c9e5aa19
  由 Yan Chunwei 提交于 1月 18, 2019
  
  c9e5aa19
17 1月, 2019 5 次提交
- Y
  
  make clone thread safe (#15363) · e84234b5
  由 Yan Chunwei 提交于 1月 17, 2019
  
  e84234b5
- T
  add pyramid_dnn c++ inference test · 66856308
  由 Tao Luo 提交于 1月 17, 2019
```
test=develop
```
  66856308
- N
  fix unit test bug · 8817841c
  由 nhzlx 提交于 1月 17, 2019
```
test=develop
```
  8817841c
- Y
  
  cache tensor ptr in ZeroCopyTensor (#15352) · e07900d3
  由 Yan Chunwei 提交于 1月 17, 2019
  
  e07900d3
- Y
  
  hot fix the Native clone (#15344) · b7916440
  由 Yan Chunwei 提交于 1月 17, 2019
  
  b7916440
16 1月, 2019 4 次提交

N
fix win build bug · b95f2ff8
由 nhzlx 提交于 1月 16, 2019
```
test=develop
```
b95f2ff8
N
add trt int8 calibration support · 312fe0ec
由 nhzlx 提交于 1月 16, 2019
```
fix comments

test=develop
```
312fe0ec

Optimize while_op for test (#14764) · 568cc2ff

由 Yiqun Liu 提交于 1月 16, 2019

* Simplify the compare op for CPU.

* Use asynchronous tensor copy in reshape_op's kernel.

* Optimize while_op for test, avoiding creating variables every time.
test=develop

* Enable the cache of kernel type and kernel function.
test=develop

* Enable profiling with gperftools.

* Remove flags for testing, and fix the linking error.
test=develop

* Delete the codes of ChooseKernel.
test=develop

* Fix bug when preparing ExecutorPrepareContext for while_op.

* Fix missing depending on grpc libraries.

* Remove the redundant print.
test=develop

* Follow comments.

* Remove the codes related to prepare the ExecutorPrepareContext for while_op.
test=develop

568cc2ff

N

fix analysis config bug. · b2ba3471
由 nhzlx 提交于 1月 16, 2019

b2ba3471

15 1月, 2019 1 次提交
- B
  Disable conv3d mkldnn in dam (#15335) · a152a5c7
  由 bingyanghuang 提交于 1月 15, 2019
```
* disable conv3d mkldnn in dam

* Add some comments  test=develop
```
  a152a5c7
14 1月, 2019 6 次提交
- T
  disable seqpool test on mac or without mkl · 1a95cd22
  由 tensor-tang 提交于 1月 14, 2019
```
test=develop
```
  1a95cd22
- T
  fix multi-threads in ZeroCopyProfile · 2411ed42
  由 Tao Luo 提交于 1月 14, 2019
```
test=develop
```
  2411ed42
- T
  
  adjust acc on mac · 7035f051
  由 tensor-tang 提交于 1月 14, 2019
  
  7035f051
- T
  fix fuse square mat order and refine test · d618e483
  由 tensor-tang 提交于 1月 13, 2019
```
test=develop
```
  d618e483
- T
  
  add fuse pass of sequared mat sub fusion · a5d2a6d1
  由 tensor-tang 提交于 1月 13, 2019
  
  a5d2a6d1
- T
  adjust the acc since the refer result is too large · 84e023ea
  由 tensor-tang 提交于 1月 13, 2019
```
test=develop
```
  84e023ea
13 1月, 2019 3 次提交
- T
  adjust diff since abs is too large · 4461a458
  由 tensor-tang 提交于 1月 13, 2019
```
test=develop
```
  4461a458
- T
  refine and fix test · ca6fdc6e
  由 tensor-tang 提交于 1月 13, 2019
```
test=develop
```
  ca6fdc6e
- T
  
  add repeated fc relu pass · a89296ac
  由 tensor-tang 提交于 1月 12, 2019
  
  a89296ac

机器未来 / Paddle 与 Fork 源项目一致

机器未来 / Paddle
与 Fork 源项目一致