提交 · 68ab16444abc54db77caae99a40f841d53016a81 · PaddlePaddle / Paddle

22 12月, 2018 1 次提交
- T
  add eng doc of jit kernel and follow comments · 68ab1644
  由 tensor-tang 提交于 12月 21, 2018
```
test=develop
```
  68ab1644
21 12月, 2018 2 次提交

[Feature] Add Temporary Allocator (#14875) · 79bd6dfa

由 chengduo 提交于 12月 21, 2018

* Add Temporal Allocator

* add Temporay Allocator to DeviceContext
test=develop

* code refine
test=develop

* fix mean_iou
test=develop

* Add DeviceTemporaryAllocator
test=develop

* fix conv_op bug
test=develop

* small fix
test=develop

* code refine
test=develop

* log refine
test=develop

* fix unit test
test=develop

* move double check

* refine concat_and_split
test=develop

* add limit_of_temporary_allocation
test=develop

* fix name
test=develop

79bd6dfa

T
fix deps of jit benchmark · e99e4e99
由 tensor-tang 提交于 12月 21, 2018
```
test=develop
```
e99e4e99

20 12月, 2018 12 次提交

J
- Added transpose/transpose2 MKLDNN grad ops · d05094ee
由 Jacek Czaja 提交于 12月 20, 2018
```
test=develop
```
d05094ee
C
Refine merge_selected_rows Doc (#14748) · a015a8a3
由 chengduo 提交于 12月 20, 2018
```
* add doc for MergeSelectedRows
test=develop

* checkout selected_rows
test=develop
```
a015a8a3
T
Revert "[Feature] Fp16 training for resnet50 (#14850)" · da87f7a6
由 typhoonzero 提交于 12月 20, 2018
```
This reverts commit 3d750f9c.
```
da87f7a6
T
fix enum style · 1aaec571
由 tensor-tang 提交于 12月 20, 2018
```
test=develop
```
1aaec571
S
polish infer shape of py_func op · 490eb906
由 sneaxiy 提交于 12月 20, 2018
```
test=develop
```
490eb906
T
follow comment: reuse time function and change to upper case · facfecbd
由 tensor-tang 提交于 12月 20, 2018
```
test=develop
```
facfecbd
Y
Fix the regression issue and add the group unitest for conv2d (#14932) · 3babc801
由 Yihua Xu 提交于 12月 20, 2018
```
* Add test items for mkldnn conv2d

* Fix the regression issue and pass the unit test for conv2d and conv3d

test=develop
```
3babc801
C
Code Clean parallel_executor.py (#14849) · 550e7e41
由 chengduo 提交于 12月 20, 2018
```
* refine parallel_executor

* remove uncessary code
test=develop
```
550e7e41

[Feature] Fp16 training for resnet50 (#14850) · 3d750f9c

由 Wu Yi 提交于 12月 20, 2018

* wip

* wip

* wip

* wip for test

* add fp16 tests test=develop

* fix cpu build test=develop

* fix test=develop

* fix py3 tests test=develop

* fix lr_scheduler dtype test=develop

* fix test=dvelop

* test fix ci compile test=develop

* fix build and merge test=develop

* fallback momentumop change to general test=develop

3d750f9c

L
add deterministic result unit-test · 13367866
由 luotao1 提交于 12月 20, 2018
```
test=develop
```
13367866

lookup table utils fix (#14730) · 45dd3491

由 tangwei12 提交于 12月 20, 2018

* fix utils config, test=develop

* reweite tools in lookup table utils, test=develop

* merge develop, test=develop

* add copy in setup, test=develop

* update api spec, test=develop

* add doc, test=develop

* prettifying doc, test=develop

* Update API.spec

update api.spec, test=develop

* Update lookup_table_utils.py

test=develop

* Update lookup_table_utils.py

test=develop

* merge develop, test=develop

* merge develop, test=develop

* init fix, test=develop

* fix in downloads, test=develop

* fix in setup.in, test=develop

45dd3491

Avoid calling VLOG() when program exiting (#14937) · e7675dd5

由 Wang Guibao 提交于 12月 20, 2018

* Avoid calling VLOG() when program exiting
test=develop

* Avoid calling VLOG() when program exiting
test=develop

e7675dd5

19 12月, 2018 20 次提交
- T
  fix typo of mix impl of lstm · a369c802
  由 tensor-tang 提交于 12月 19, 2018
```
test=develop
```
  a369c802
- T
  fix build on win, fix use condition of crf decoding and layer norm and · d651a911
  由 tensor-tang 提交于 12月 19, 2018
```
enhance test precision

test=develop
```
  d651a911
- T
  MKLDNN residual connection fuse: fixing accuracy problem (#14874) · e9eee0de
  由 Tomasz Patejko 提交于 12月 19, 2018
```
* MKLDNN residual connection fuse: conv op reused

test=develop

* MKLDNN residual connection fuse: added prints for checking fuse

test=develop

* MKLDNN residual connection fuse: add more prints

test=develop

* MKLDNN residual connection fuse: add hash function. test=develop

* MKLDNN residual connection fuse: add hash to elementwise_add

test=develop

* MKLDNN residual connection fuse: add more hashes. test=develop

* MKLDNN residual connection fuse: added hashes to relu

test=develop

* MKLDNN residual connection fuse: do not fuse when fuse_relu is on

* MKLDNN residual connection fuse: check if fuse_relu attribute is set

test=develop

* MKLDNN residual connection fuse: comment out some printouts

* MKLDNN residual connection fuse: remove unused functions in the pass code

* MKLDNN residual connection fuse: delete commented hashes and printouts

* MKLDNN residual connection fuse: remove unnecessary includes. test=develop
```
  e9eee0de
- T
  clean code and remove unused files · d53c4756
  由 tensor-tang 提交于 12月 19, 2018
```
test=develop
```
  d53c4756
- T
  add more impls of lstm and gru and fix build on win · c187a7c6
  由 tensor-tang 提交于 12月 19, 2018
```
test=develop
```
  c187a7c6
- X
  move more and fix while · 1fe3ac35
  由 Xin Pan 提交于 12月 19, 2018
```
test=develop
```
  1fe3ac35
- J
  - Added reusing MKL-DNN primitives for Transpose MKL-DNN op · 709d9e3c
  由 Jacek Czaja 提交于 12月 19, 2018
```
test=develop
```
  709d9e3c
- T
  fix lstm and gru jitcode · 83d075aa
  由 tensor-tang 提交于 12月 19, 2018
```
test=develop
```
  83d075aa
- X
  convert more · 9ef8a768
  由 Xin Pan 提交于 12月 19, 2018
```
test=develop
```
  9ef8a768
- X
  convert more interface to avoid scope · 87699388
  由 Xin Pan 提交于 12月 19, 2018
```
test=develop
```
  87699388
- T
  
  add more impls of sigmoid and vtanh · f332f589
  由 tensor-tang 提交于 12月 19, 2018
  
  f332f589
- X
  fix · 8c19f0bf
  由 Xin Pan 提交于 12月 19, 2018
```
test=develop
```
  8c19f0bf
- M
  Enable mul operator for a ngraph engine (#14801) · 9035bb81
  由 mozga-intel 提交于 12月 19, 2018
```
* Enable mul operator for a ngraph
test=develop

* Enable activation ops test
test=develop

* Remove unused line
test=develop
```
  9035bb81
- T
  
  enable layer norm intrinsic code · ea259c63
  由 tensor-tang 提交于 12月 18, 2018
  
  ea259c63
- G
  
  Add size enforce (#14919) · b849157e
  由 gongweibao 提交于 12月 19, 2018
  
  b849157e
- X
  convert GetInputVarPtrs and GetOutputVarPtrs · 4dd61e72
  由 Xin Pan 提交于 12月 19, 2018
```
test=develop
```
  4dd61e72
- J
  [MKL-DNN ]Added transpose/transpose2 Op (#14872) · aa6e9c30
  由 Jacek Czaja 提交于 12月 19, 2018
```
* - Added transpose MKLDNN Op
- Few basic UT works
- Added 1D transpose
- implementing generic mem desc for MKLDNN transpose
- Modified trnaspose op to support  more dimensional data eg. 5,6..10
- Added is_test attribute to transpose op

test=develop

* - Added support for MKLDNN::memory::format::any for Transpose MKLDNN op

test=develop

* - Additional transpose mkldnn op correction to mkldnn layout

test=develop

* Cosmetic fixes

test=develop

* - Removed const_cast to obey coding standard

test=develop
```
  aa6e9c30
- X
  fix · 52d3903a
  由 Xin Pan 提交于 12月 19, 2018
```
test=develop
```
  52d3903a
- X
  
  convert more infer shape · 0e0983cc
  由 Xin Pan 提交于 12月 18, 2018
  
  0e0983cc
- X
  convert more · 62eb43ba
  由 Xin Pan 提交于 12月 18, 2018
```
test=develop
```
  62eb43ba
18 12月, 2018 5 次提交
- T
  
  enable crf decoding intrinsic code · b1516783
  由 tensor-tang 提交于 12月 18, 2018
  
  b1516783
- T
  
  add crf_decoding and layer norm intrisic code · 4cc7707d
  由 tensor-tang 提交于 12月 18, 2018
  
  4cc7707d
- T
  
  fix confilcts · 10c340c9
  由 tensor-tang 提交于 12月 18, 2018
  
  10c340c9
- T
  
  fix build · 6648995f
  由 tensor-tang 提交于 12月 17, 2018
  
  6648995f
- X
  MLP forward backward · 63240326
  由 Xin Pan 提交于 12月 13, 2018
```
test=develop
```
  63240326

PaddlePaddle / Paddle 大约 1 年 前同步成功

PaddlePaddle / Paddle
大约 1 年前同步成功