提交 · 7d91974c91b427d12e40a644518993a22dc0b1a3 · PaddlePaddle / Paddle

25 2月, 2021 3 次提交

add cache for VariableWrapper (#30880) · ca3b6bcf

由 chentianyu03 提交于 2月 25, 2021

* add cache for VariableWrapper

* modify args names and vlog level

* format code style

* add log when set cache to variable_wrapper

* add log when set cache to variable_wrapper

* add comment to variableWrapper cache

* format code style

ca3b6bcf

J

OneDNN hardswish integration (#30211) · 2f116534
由 jakpiase 提交于 2月 25, 2021

2f116534

[CustomOp] Support attributes as func input in custom op (#31128) · e8cdb49a

由 Chen Weihang 提交于 2月 25, 2021

* add simple attr support and test

* add int, float attr support

* support other attribute

* add custom attrs test in cmake

* polish details

* fix test failed

* add backward test

* update test flags

e8cdb49a

24 2月, 2021 4 次提交
- Z
  
  modify custom op dependent from paddle_framework to paddle_custom_op (#31195) · ffbf7135
  由 Zhou Wei 提交于 2月 24, 2021
  
  ffbf7135
- L
  align the default value of some configuration for fleet to that of single cards (#30740) · dc8dfba3
  由 lilong12 提交于 2月 24, 2021
```
* update, test=develop
```
  dc8dfba3
- T
  support multi node in heterps (#31102) · c4f279fe
  由 Thunderbrook 提交于 2月 24, 2021
```
* push multi node

* multi node

* MultiThread

* remove log

* solve bug in 30829
```
  c4f279fe
- C
  [CustomOp] Add new paddle custom op so (#31141) · 1ce96fa1
  由 Chen Weihang 提交于 2月 23, 2021
```
* add new custom op so

* fix use new method error

* fix test failed
```
  1ce96fa1
23 2月, 2021 3 次提交

added support for fake_quantize_dequantize_abs_max op in quantization… (#30896) · 5d6a8c7b

由 alncat 提交于 2月 23, 2021

* added support for fake_quantize_dequantize_abs_max op in quantization inference pass

* remove const_cast to pass ci

* remove compare operator to pass ci-coverage

* added detailed error message for unregistered tensorrt_subgrah_pass

5d6a8c7b

Unification of BF16 enablement process (#31034) · 781df300

由 joanna.wozna.intel 提交于 2月 23, 2021

* Unification of bfloat16 enablement process and refactor

* Remove unnecessary function

* Standardize the output name search

781df300

Q

[ROCM] update fluid framework for rocm (part2), test=develop (#31010) · a60d93fb
由 Qi Li 提交于 2月 23, 2021

a60d93fb

22 2月, 2021 4 次提交
- T
  support save multi sparse table in one path (#31108) · 565354f6
  由 Thunderbrook 提交于 2月 22, 2021
```
* save multi table one path

* format
```
  565354f6
- Q
  
  [ROCM] update fluid framework for rocm (part3), test=develop (#31011) · 50967135
  由 Qi Li 提交于 2月 22, 2021
  
  50967135
- Q
  
  [ROCM] update fluid framework for rocm (part1), test=develop (#31009) · 8fe09faf
  由 Qi Li 提交于 2月 22, 2021
  
  8fe09faf
- Z
  [2.0Custom OP]Support New Custom OP on Windows (#31063) · adaec007
  由 Zhou Wei 提交于 2月 22, 2021
```
* [2.0.1]Support New Custom OP on windows

* fix CI

* fix code style

* fix CI

* fix CI

* fix coverage

* fix CI

* fix CI
```
  adaec007
20 2月, 2021 3 次提交
- C
  Remove PE special profiler (#30886) · 6b3371e0
  由 Chengmo 提交于 2月 20, 2021
```
* remove pe special profiler

* add profiler info
```
  6b3371e0
- T
  add squeeze_op/unsqueeze_op on kunlun;fix conv op and parallel... · d5323dab
  由 TTerror 提交于 2月 20, 2021
```
add squeeze_op/unsqueeze_op on kunlun;fix conv op and parallel executor;optimize lookup_table op (#31056)

* add squeeze_op/unsqueeze_op on kunlun; fix conv op and parallel executor on kunlun; optimize lookup_table op on kunlun

* update squeeze/unsqueeze op
```
  d5323dab
- J
  
  hide useless headers and add complex support (#31074) · 628451af
  由 Jiabin Yang 提交于 2月 20, 2021
  
  628451af
18 2月, 2021 1 次提交

Add Conv Transpose BF16 (#30877) · caf9d398

由 joanna.wozna.intel 提交于 2月 18, 2021

* Add conv transpose BF16

* Share function GetWeightsTz

* Adjust to review and fix op compatibility

* Add bias to unique handler name

* Remove errors related to paddle enforce

* Add conv2d_transpose to bf16 list and kernel refator

caf9d398

10 2月, 2021 1 次提交

New custom operator extension mechanism (#30690) · f649442d

由 Chen Weihang 提交于 2月 09, 2021

* initial commit: simple demo

* polish copyright format

* add grap op simple demo

* adapt uncertain number of argument

* change trait marco name

* add place & dtype support for add kernel

* add dispath and infershape func

* poish code & add notes

* add dynamic_loader dep for paddle_framework

* add new custom op test dir

* polish impl details

* add unittest for new custom op

* fix failed unittest

* Costum op (#1)

* fix compile error

* wrap framework tensor with LoDTensor

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* add CustomTensor default constructor

* add size() for CustomTensor

* make size const for CustomTensor

* refactor place related api to circle the concept

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* make place const

* make Tensor copy

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* remove additional head of framework

* use back to shared ptr for custom tensor

* use back to shared ptr for custom tensor

* use back to shared ptr for custom tensor

* use back to shared ptr for custom tensor

* use back to shared ptr for custom tensor

* use back to shared ptr for custom tensor

* add gpu test

* merge latest cwh code in

* adjust ut code of custom op

* adjust ut code of custom op

* adjust ut code of custom op

* Remove ShareData from user && Change CustomTensor to Tensor && Support more data type (#2)

* fix compile error

* wrap framework tensor with LoDTensor

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* add CustomTensor default constructor

* add size() for CustomTensor

* make size const for CustomTensor

* refactor place related api to circle the concept

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* make place const

* make Tensor copy

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* remove additional head of framework

* use back to shared ptr for custom tensor

* use back to shared ptr for custom tensor

* use back to shared ptr for custom tensor

* use back to shared ptr for custom tensor

* use back to shared ptr for custom tensor

* use back to shared ptr for custom tensor

* add gpu test

* merge latest cwh code in

* adjust ut code of custom op

* adjust ut code of custom op

* adjust ut code of custom op

* adjust ut code of custom op

* adjust ut code of custom op

* hid share data from and to

* rename CustomTensor to Tensor

* refactor register design & add test

* change op_funtion to op_meta_info

* split op meta info into .h and .cc

* move get methods into friend class

* move OpMetaInfoHelper into framework space

* move CustomTensorUtils into framework space

* change pybind api name

* move PD C API into op meta info

* add register custom op api

* remove inference cmake change

* refactor copy to api && change Reshape to lowercase && support more dtype && add more test (#3)

* fix compile error

* wrap framework tensor with LoDTensor

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* add CustomTensor default constructor

* add size() for CustomTensor

* make size const for CustomTensor

* refactor place related api to circle the concept

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* make place const

* make Tensor copy

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* remove additional head of framework

* use back to shared ptr for custom tensor

* use back to shared ptr for custom tensor

* use back to shared ptr for custom tensor

* use back to shared ptr for custom tensor

* use back to shared ptr for custom tensor

* use back to shared ptr for custom tensor

* add gpu test

* merge latest cwh code in

* adjust ut code of custom op

* adjust ut code of custom op

* adjust ut code of custom op

* adjust ut code of custom op

* adjust ut code of custom op

* hid share data from and to

* rename CustomTensor to Tensor

* support multi dtype

* remove lod, make reshape lowercase, add copy test and refactor copy api

* remove lod, make reshape lowercase, add copy test and refactor copy api

* remove lod, make reshape lowercase, add copy test and refactor copy api

* remove lod, make reshape lowercase, add copy test and refactor copy api

* fix copy to error

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* polish detail & error message

* polish test details

* Add cast api && Change copy related api to copy_to && add more test (#4)

* fix compile error

* wrap framework tensor with LoDTensor

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* add CustomTensor default constructor

* add size() for CustomTensor

* make size const for CustomTensor

* refactor place related api to circle the concept

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* fix compile error

* make place const

* make Tensor copy

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* debug CustomTensor core

* remove additional head of framework

* use back to shared ptr for custom tensor

* use back to shared ptr for custom tensor

* use back to shared ptr for custom tensor

* use back to shared ptr for custom tensor

* use back to shared ptr for custom tensor

* use back to shared ptr for custom tensor

* add gpu test

* merge latest cwh code in

* adjust ut code of custom op

* adjust ut code of custom op

* adjust ut code of custom op

* adjust ut code of custom op

* adjust ut code of custom op

* hid share data from and to

* rename CustomTensor to Tensor

* support multi dtype

* remove lod, make reshape lowercase, add copy test and refactor copy api

* remove lod, make reshape lowercase, add copy test and refactor copy api

* remove lod, make reshape lowercase, add copy test and refactor copy api

* remove lod, make reshape lowercase, add copy test and refactor copy api

* fix copy to error

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add more test

* add type cast

* add cast and make copy to api

* add cast and make copy to api

* add cast and make copy to api

* add cast and make copy to api

* merge cwh code

* merge cwh code

* merge cwh code

* merge cwh code

* merge cwh code

* add more error log

* add more error log

* polish code

* used for test

* remove test comment

* remove test comment

* fix uint8 type error

* fix lost uint8 type error

* add test for coverage

* polish details by reviewer comments

* add prefix for DISABLE_COPY_AND_ASSIGN
Co-authored-by: NJiabin Yang <360788950@qq.com>

f649442d

09 2月, 2021 2 次提交
- W
  
  Fix the problem that the number of ops executed by xpu is wrong (#30961) · 14d039e4
  由 WangXi 提交于 2月 09, 2021
  
  14d039e4
- A
  
  Fix LayerNorm tester for gcc4.8 (#30962) · 3ba69809
  由 Adam Osewski 提交于 2月 09, 2021
  
  3ba69809
05 2月, 2021 2 次提交
- W
  
  add include for heterbox_trainer.cc, develop=test (#30910) · aab3a301
  由 wanghuancoder 提交于 2月 05, 2021
  
  aab3a301
- A
  More UT for LayerNormFuse pass (#30891) · 092a2b14
  由 Adam Osewski 提交于 2月 05, 2021
```
* Additionally change to not throw error from inside pass.
```
  092a2b14
04 2月, 2021 1 次提交
- W
  use iwyu clean include second time, test=develop (#30829) · 35c5b23f
  由 wanghuancoder 提交于 2月 04, 2021
```
* use iwyu clean include second time, test=develop
```
  35c5b23f
03 2月, 2021 1 次提交
- A
  
  Layer normalization fuse pass. (#30721) · 4f066e31
  由 Adam Osewski 提交于 2月 03, 2021
  
  4f066e31
01 2月, 2021 2 次提交
- T
  dump to cpu (#30750) · cb66c53c
  由 Thunderbrook 提交于 2月 01, 2021
```
* dump to cpu

* format

* format

* format
```
  cb66c53c
- W
  
  Fleet distributed strategy support pure fp16 (#30754) · 31ed9c9e
  由 WangXi 提交于 2月 01, 2021
  
  31ed9c9e
28 1月, 2021 1 次提交
- A
  
  fixed compilation error on gcc 4.8.x due to the usage of isfinite (#30733) · 5b59499e
  由 alncat 提交于 1月 28, 2021
  
  5b59499e
27 1月, 2021 2 次提交
- L
  
  [Kunlun] fix dead lock for exec_op_count_ (#30718) · 67abfc15
  由 liuyuhui 提交于 1月 27, 2021
  
  67abfc15
- A
  
  modified conv+bn fuse pass to fix wrong mask in mask rcnn (#30704) · 5ace20fc
  由 alncat 提交于 1月 27, 2021
  
  5ace20fc
26 1月, 2021 1 次提交
- L
  
  update, test=develop (#30692) · 7fbc68a2
  由 lilong12 提交于 1月 26, 2021
  
  7fbc68a2
25 1月, 2021 2 次提交
- A
  More precise mkldnn kernel rules in GetExpectedKernelType (#29840) · 5bf25d1e
  由 arlesniak 提交于 1月 25, 2021
```
* More precise mkldnn kernel choice in GetExpectedKernelType

* Fixes after review

* Refresh develop for CI

* CI experiment

* get back from CI exper
```
  5bf25d1e
- J
  
  [oneDNN] Cache oneDNN stream not to recreate in each oneDNN op (#30358) · 173660be
  由 Jacek Czaja 提交于 1月 25, 2021
  
  173660be
21 1月, 2021 1 次提交
- T
  solve build gpu task core (#30626) · 1bebc092
  由 Thunderbrook 提交于 1月 21, 2021
```
* build gpu task core

* format
```
  1bebc092
20 1月, 2021 1 次提交
- L
  
  [Kunlun] Add condition_variable and notify() in BindThreadedSSAGraphExecutor (#30586) · e5b0d9e1
  由 liuyuhui 提交于 1月 20, 2021
  
  e5b0d9e1
19 1月, 2021 2 次提交
- L
  
  Fix bug: GetAttrValue should deal with attr with attrType vector<double> (#30536) · ff25c5b3
  由 liym27 提交于 1月 19, 2021
  
  ff25c5b3
- L
  unify calling cudaSetDevice (#30470) · 81217a94
  由 Leo Chen 提交于 1月 19, 2021
```
* unify calling cudaSetDevice

* fix compile
```
  81217a94
18 1月, 2021 2 次提交
- H
  
  Ascend Framework Part1: OP & Wrapper (#30281) · 40ede126
  由 hutuxian 提交于 1月 18, 2021
  
  40ede126
- L
  
  [Kunlun]PR3: add xpu executor, multi xpu card train function optimization (#30317) · 843dc3cd
  由 liuyuhui 提交于 1月 18, 2021
  
  843dc3cd
16 1月, 2021 1 次提交
- A
  [oneDNN] Refactor fuse pass helper functions to one place. (#30460) · c5ffad12
  由 Adam Osewski 提交于 1月 16, 2021
```
* Move pass tester helper functions to single common place.

* Use helper functions in two more fuse pass tests.
```
  c5ffad12

PaddlePaddle / Paddle 大约 1 年 前同步成功

PaddlePaddle / Paddle
大约 1 年前同步成功