提交 · c3cbf0b8ef85d0b8e2f46fc5281aaa69c77ec9f1 · Crayon鑫 / Paddle

05 11月, 2018 3 次提交
- T
  remove the warning log since do not have avx2, avx512 flags · e09a7c79
  由 tensor-tang 提交于 11月 05, 2018
```
test=develop
```
  e09a7c79
- T
  throw error when mismatch cpu version · f524c1b6
  由 tensor-tang 提交于 10月 23, 2018
```
test=develop
```
  f524c1b6
- X
  fix to only check block 0 · c2d70fca
  由 Xin Pan 提交于 11月 05, 2018
```
test=develop
```
  c2d70fca
02 11月, 2018 10 次提交
- X
  make it warn · aaeedd0f
  由 Xin Pan 提交于 11月 02, 2018
```
test=develop
```
  aaeedd0f
- X
  add more debug info. · ddd2225b
  由 Xin Pan 提交于 11月 02, 2018
```
test=develop
```
  ddd2225b
- S
  
  test=develop · 203027ca
  由 sneaxiy 提交于 11月 02, 2018
  
  203027ca
- W
  enable pyreader use pin memory (#14066) · 91b2851c
  由 Wu Yi 提交于 11月 02, 2018
```
* enable pyreader use pin memory

* add py reader pin memory test test=develop
```
  91b2851c
- W
  Add affine grid generator op (#12238) · 0c319e0b
  由 whs 提交于 11月 02, 2018
```
* Add affine grid generator.

* fix ffine grid.

* Add unitest.

* Add CPU kernel and fix unitest.

* Fix CPU kernel.

* Refine code.
test=develop

* Fix python api.
test=develop

* Update python api.
test=develop

* Fix comment.
test=develop

* Rename affine_grid_generator to affine_grid and enhence unitest.
test=develop

* Fix unitest.
test=develop
```
  0c319e0b
- T
  [1.1] Load vars on PSERVER (#14037) · d325e668
  由 tangwei12 提交于 11月 02, 2018
```
* fix dim0 in _load_slice_up_vars

* fix dim0 in _load_slice_up_vars, fix innershape in delete_var_op

* Revert "fix lookuptable in reduce strategy"

This reverts commit 0e722c5e

* add unit test for dist

* add unit test for dist, test=develop

* cancel revert, test=develop
```
  d325e668
- D
  
  api change: create_variable_for_type_inference. test=develop · e99da0b5
  由 dengkaipeng 提交于 11月 01, 2018
  
  e99da0b5
- Y
  
  fix graph pattern detector (#14186) · f76fee64
  由 Yan Chunwei 提交于 11月 01, 2018
  
  f76fee64
- T
  
  fix word2vec related inference unit-tests (#14203) · fe8f1785
  由 Tao Luo 提交于 11月 02, 2018
  
  fe8f1785
- C
  fix merge lod_tensor bug (#14199) · e1742050
  由 chengduo 提交于 11月 02, 2018
```
test=develop
```
  e1742050
01 11月, 2018 10 次提交
- T
  refine vmul jitcode · 85bcb286
  由 tensor-tang 提交于 11月 01, 2018
```
test=develop
```
  85bcb286
- T
  
  refine jitcode and add vmul jitcode implementation · a3377f7b
  由 tensor-tang 提交于 11月 01, 2018
  
  a3377f7b
- D
  
  fix API.spec not add defaults. test=develop · da8ee1fb
  由 dengkaipeng 提交于 11月 01, 2018
  
  da8ee1fb
- C
  Refine GetTensorFromVar (#14160) · 2ccf77d1
  由 chengduo 提交于 11月 01, 2018
```
* fix GetTensorFromVar
test=release/1.1

* refine GetTensorFromVar
test=develop
```
  2ccf77d1
- T
  remove unused WITH_FAST_BUNDLE_TEST option · 5ac575cf
  由 Tao Luo 提交于 11月 01, 2018
```
test=develop
```
  5ac575cf
- Y
  fix(PE): use shared_ptr<BlockingQueue> for cross thread communication (#14136) · c21597cf
  由 Yu Yang 提交于 11月 01, 2018
```
It seems that the blocking queue might be destroyed early than Run
method complete. It might because the Run method throw some unhandled
exception. However, it should be shared_ptr when multthread access an
resource. So change BlockingQueue as a shared_ptr.

test=develop
```
  c21597cf
- T
  
  refine and init jitkernel vmul · a53b1b0b
  由 tensor-tang 提交于 11月 01, 2018
  
  a53b1b0b
- T
  
  add jit gencode · 2139b9f6
  由 tensor-tang 提交于 11月 01, 2018
  
  2139b9f6
- Y
  
  fix simple_on_word2vec random fail (#14171) · 06e508ab
  由 Yan Chunwei 提交于 10月 31, 2018
  
  06e508ab
- T
  MKLDNN conv residual data: primitive reuse interface used. Reorder done when formats are different · 8899d422
  由 Tomasz Patejko 提交于 10月 31, 2018
```
test=develop
```
  8899d422
31 10月, 2018 7 次提交
- C
  add int and int64 dtype for gather_op (#14175) · b73708d2
  由 chengduo 提交于 10月 31, 2018
```
test=develop
```
  b73708d2
- T
  
  MKLDNN conv residual data: residual data is reorder when formats are incorrect · f11934cb
  由 Tomasz Patejko 提交于 10月 30, 2018
  
  f11934cb
- Y
  fix tensor array bug (#14166) · 62a0fe08
  由 Yan Chunwei 提交于 10月 31, 2018
```
remove the optimized but buggy implementation
```
  62a0fe08
- C
  refine op_handle (#14178) · ed087f82
  由 chengduo 提交于 10月 31, 2018
```
test=develop
```
  ed087f82
- D
  
  add math formula for exclusive/inclusive mode in avg pool. test=develop · 7333fe8e
  由 dengkaipeng 提交于 10月 31, 2018
  
  7333fe8e
- Y
  feat(platform): lazy initialization of devicecontext in pool (#14067) · 90d9e5ae
  由 Yu Yang 提交于 10月 31, 2018
```
* feat(platform): lazy initialization of devicecontext in pool

Use std::async(deferer, []{...}) to lazy initialize DeviceContext in Pool

test=develop

* Add future includes

test=develop
```
  90d9e5ae
- S
  add depthwise conv mkldnn pass · 4e2aaf01
  由 Sylwester Fraczek 提交于 10月 30, 2018
```
added depthwise conv mkldnn pass which for MKLDNN changes depthwise_conv operator to conv operator because for mkldnn this is the same api
test=develop
```
  4e2aaf01
30 10月, 2018 10 次提交
- Y
  
  add fused broadcast op unit test, test=develop · 6bfa6a0a
  由 Yancey1989 提交于 10月 30, 2018
  
  6bfa6a0a
- X
  add a small test to verify tensor type · e2db0b9b
  由 Xin Pan 提交于 10月 30, 2018
```
test=develop
```
  e2db0b9b
- Y
  
  fix api_impl ci error (#14140) · 70ce6dcd
  由 Yan Chunwei 提交于 10月 30, 2018
  
  70ce6dcd
- X
  fix a few more tests · a943134a
  由 Xin Pan 提交于 10月 29, 2018
```
test=develop
```
  a943134a
- C
  Fix sum_op's GetExpectedKernelType (#14112) · 2f639113
  由 chengduo 提交于 10月 30, 2018
```
* fix sum_op's GetExpectedKernelType
test=develop

* fix ci fail
test=develop
```
  2f639113
- X
  add program check · 5839e323
  由 Xin Pan 提交于 10月 26, 2018
```
test=develop
```
  5839e323
- G
  position encoding && log loss · ba22624d
  由 gmcather 提交于 10月 29, 2018
```
test=develop
```
  ba22624d
- T
  remove with_inference option · 3a96d41d
  由 Tao Luo 提交于 10月 30, 2018
```
test=develop
```
  3a96d41d
- J
  
  test=develop · 7c45e77c
  由 JiabinYang 提交于 10月 30, 2018
  
  7c45e77c
- S
  
  test=develop · 5e5d2223
  由 sneaxiy 提交于 10月 26, 2018
  
  5e5d2223

Crayon鑫 / Paddle 与 Fork 源项目一致

Crayon鑫 / Paddle
与 Fork 源项目一致