提交 · 1a7399b47ed86125acffb233c40ac3b2cadf5bc5 · PaddlePaddle / Paddle

21 3月, 2022 2 次提交

由 hong 提交于 3月 21, 2022

* switch eager mode and change it

* set default is eager

* set default is eager

* fix error; test=develop

* fix some error; test=develop

* update

* upd

* update code; test=develop

* update

* fix some bug; test=develop

* fix bug; test=develop

* fix bug; test=develop

* fix bug; test=develop

* fix error; test=develop

* format; test=develop
Co-authored-by: NJiabinYang <360788950@qq.com>

56c43ccd

[Eager grad] Refactor partial grad logic (#40693) · facda828

由 Weilong Wu 提交于 3月 21, 2022

* Refactor partial_grad/backward logic

* Add DuplicateCheck and polish code

* Refactor partial_grad/backward more clearly

* Refactor GeneralGrad by SingleInstance

facda828

19 3月, 2022 2 次提交

Z
Call sparse op from python (#40608) · 95fbbc5b
由 zhangkaihuo 提交于 3月 19, 2022
```
* call sparse api from python
```
95fbbc5b

support inplace in dygraph eager_fluid state (#40400) · 8e612903

由 pangyoki 提交于 3月 19, 2022

* [Eager] Support eager grad interface, draft version

* Support eager grad interface with allow_unused and multi startup_op

* Fix code format

* Fix allow_unused case, return PyNone if tensor not initialize

* Support output's stop_gradient related to create_graph

* Support grad exception case in eager mode, fix coverage CI

* Update ToPyObject, return PyNone if not initialize

* AccumulationNode add FLAGS_retain_grad_for_all_tensor

* Fix ci issue

* Fix CI issue

* fix, use core.eager.Tensor

* Add func SetBufferSlotRankZeros for GradTensorHolder

* Support retain_graph by using ClearTensorWrappers

* Support retain_graph by using ClearTensorWrappers

* Update retain_graph and no_grad_vars related test case

* Update code gen logic for ClearTensorWrappers

* Fix by override statement

* fix override func args

* Support retain_graph, update unit tests

* Updated ClearTensorWrappers logic

* fix grad python interface

* Use deep copy and update unit tests

* Polish code

* Polish code

* Fix CI issue, Deep copy only use when user set grad_tensors

* Fix CI, use Backward instead RunBackward

* Fix CI, Declare kernel explicitly in test file

* Polish, remove vector of TensorWrapper

* Refactor the logic of grad/backward, polish codes

* Update code after merge upstream develop

* Polish after merge upstream develop

* Update to adapt new GradNodeBase superclass

* Fix error introduced during conflict resolution

* support inplace strategy in eager_fluid state

* solve conflict

* nothing

* Update purify potential_startup_nodes logic

* Fix errors

* Polish code

* Remove useless args for ToPyObject

* Remove useless TensorWrappersSet

* fix record conflict

* Fix code-format, re-install pre-commit

* fix tensor_wrapper bug

* Fix pre-process logic for potential_startup_ops

* Update unit tests, use eager mode

* Fix conflicts

* fix unittest timeout

* little change
Co-authored-by: NWeilong Wu <veyron_wu@163.com>

8e612903

18 3月, 2022 2 次提交

Supported Complex2Real Conversion for Eager Dygraph (#39878) · e3b2a035

由 Zhanlue Yang 提交于 3月 18, 2022

* Supported Complex2Real Conversion for Eager Dygraph

* Supported Complex2Real Conversion for Eager Dygraph

* Enabled complex type promotion test for matmul_v2

* Fix CI issues

* Merged adj_edges_ with GradSlotMeta

* Fixed monir issue

* Adjusted num runs

* Recovered Eager performance tests configurations

* Recovered Eager performance tests configurations

* Adjusted performance tests configurations

* Fixed Minor Issues with performance tests

* Moved out Edge from GradSlotMeta

* Fixed issues from merge

* Fixed typo

* Addressed review comments

* Fixed minor issues

e3b2a035

Z
Refactored Final State Python-C Code Generation Scripts (#40650) · 35a5e8ee
由 Zhanlue Yang 提交于 3月 18, 2022
```
* Refactored Final State Python-C Code Generation Scripts.

* Bug fix
```
35a5e8ee

17 3月, 2022 4 次提交

[Bug fixes] Fix partial grad conflicts (#40655) · 60899549

由 Weilong Wu 提交于 3月 17, 2022

* [Eager] Support eager grad interface, draft version

* Support eager grad interface with allow_unused and multi startup_op

* Fix code format

* Fix allow_unused case, return PyNone if tensor not initialize

* Support output's stop_gradient related to create_graph

* Support grad exception case in eager mode, fix coverage CI

* Update ToPyObject, return PyNone if not initialize

* AccumulationNode add FLAGS_retain_grad_for_all_tensor

* Fix ci issue

* Fix CI issue

* fix, use core.eager.Tensor

* Add func SetBufferSlotRankZeros for GradTensorHolder

* Support retain_graph by using ClearTensorWrappers

* Support retain_graph by using ClearTensorWrappers

* Update retain_graph and no_grad_vars related test case

* Update code gen logic for ClearTensorWrappers

* Fix by override statement

* fix override func args

* Support retain_graph, update unit tests

* Updated ClearTensorWrappers logic

* fix grad python interface

* Use deep copy and update unit tests

* Polish code

* Polish code

* Fix CI issue, Deep copy only use when user set grad_tensors

* Fix CI, use Backward instead RunBackward

* Fix CI, Declare kernel explicitly in test file

* Polish, remove vector of TensorWrapper

* Refactor the logic of grad/backward, polish codes

* Update code after merge upstream develop

* Polish after merge upstream develop

* Update to adapt new GradNodeBase superclass

* Fix error introduced during conflict resolution

* Update purify potential_startup_nodes logic

* Fix errors

* Polish code

* Remove useless args for ToPyObject

* Remove useless TensorWrappersSet

* Fix code-format, re-install pre-commit

* Fix pre-process logic for potential_startup_ops

* Update unit tests, use eager mode

* Fix conflicts

60899549

Y

move activation sigmoid (#40626) · ed8a9370
由 YuanRisheng 提交于 3月 17, 2022

ed8a9370

[Eager Grad] Support eager grad interface (#40170) · 4db8cf24

由 Weilong Wu 提交于 3月 17, 2022

* [Eager] Support eager grad interface, draft version

* Support eager grad interface with allow_unused and multi startup_op

* Fix code format

* Fix allow_unused case, return PyNone if tensor not initialize

* Support output's stop_gradient related to create_graph

* Support grad exception case in eager mode, fix coverage CI

* Update ToPyObject, return PyNone if not initialize

* AccumulationNode add FLAGS_retain_grad_for_all_tensor

* Fix ci issue

* Fix CI issue

* fix, use core.eager.Tensor

* Add func SetBufferSlotRankZeros for GradTensorHolder

* Support retain_graph by using ClearTensorWrappers

* Support retain_graph by using ClearTensorWrappers

* Update retain_graph and no_grad_vars related test case

* Update code gen logic for ClearTensorWrappers

* Fix by override statement

* fix override func args

* Support retain_graph, update unit tests

* Updated ClearTensorWrappers logic

* fix grad python interface

* Use deep copy and update unit tests

* Polish code

* Polish code

* Fix CI issue, Deep copy only use when user set grad_tensors

* Fix CI, use Backward instead RunBackward

* Fix CI, Declare kernel explicitly in test file

* Polish, remove vector of TensorWrapper

* Refactor the logic of grad/backward, polish codes

* Update code after merge upstream develop

* Polish after merge upstream develop

* Update to adapt new GradNodeBase superclass

* Fix error introduced during conflict resolution

* Update purify potential_startup_nodes logic

* Fix errors

* Polish code

* Remove useless args for ToPyObject

* Remove useless TensorWrappersSet

* Fix code-format, re-install pre-commit

* Fix pre-process logic for potential_startup_ops

* Update unit tests, use eager mode

4db8cf24

J
fix copy_ problem by doing it with phi copy (#40521) · c1931beb
由 Jiabin Yang 提交于 3月 17, 2022
```
* fix copy_ problem by doing it with phi copy

* improve test coverage

* refactor copy with sr kernel
```
c1931beb

16 3月, 2022 2 次提交
- F
  Add yaml config for pool2d (#40563) · ac5cc136
  由 From00 提交于 3月 16, 2022
```
* Add yaml config for pool2d

* Fix CI error

* Fix code format error
```
  ac5cc136
- Z
  
  Fixed issue with default-valued attributes (#40368) · f748b433
  由 Zhanlue Yang 提交于 3月 16, 2022
  
  f748b433
15 3月, 2022 2 次提交

Z
Fixed issues with generated scale operator (#40482) · 30417999
由 Zhanlue Yang 提交于 3月 15, 2022
```
* Fixed issues with generated scale operator

* Fixed minor issues
```
30417999

Added more profile signposts to dygraph (#40201) · 36db75b4

由 Zhanlue Yang 提交于 3月 15, 2022

* Added more signposts to dygraph profiling

* Fixed minor issues

* Refactored signpost names

* Fixed typo

* Removed debug codes

* Fixed typo

* Adjusted signpost names

* Fixed issues from branch merge

36db75b4

14 3月, 2022 3 次提交

Support custom op and paddle.autograd.bacward in eager (#40423) · 227fa408

由 Jiabin Yang 提交于 3月 14, 2022

* eager, test=develop

* fix bug, test=develop

* eager, test=develop

* merge legacy to fluid

* eager, test=develop

* eager, test=develop

* Refactor TensorAdd func by template and remove gradient_accumulation in eager

* Remove needless target name

* eager, test=develop

* eager, test=develop

* Use overload instead of template

* Remove legacy code

* Remove legacy code

* selectedrows, test=develop

* Remove DataType test

* eager, test=develop

* eager, test=develop

* support gan, test=develop

* Using Tensor directly instead of using EagerTensor

* support gradient_accumulation

* make test_imperative_lod_tensor_to_selected_rows longer

* make test_imperative_lod_tensor_to_selected_rows longer

* refine code

* ptb, test=develop

* Rename all EagerTensor to Tensor

* Rename some EagerTensor to Tensor

* rename EagerTensor to EagerVariable

* eager, test=develop

* eager, test=develop

* eager, test=develop

* eager, test=develop

* add more test

* eager, test=develop

* Support copiable selected rows and merge develop

* save load, eager, test=develop

* save load, eager, test=develop

* refine, test=develop

* remove useless _set_value method

* refine, test=develop

* refine, test=develop

* revert static_runner, test=develop

* EagerTensor to Tensor, test=develop

* refine, test=develop

* refine, test=develop

* clear grad, test=develop

* merge, develop

* merge, develop

* merge, test=develop

* merge, test=develop

* Support quant and part of slice

* support legacy static save

* extend slim tests time

* remove imperative on inference

* remove imperative on inference

* merge develop

* fix typo

* fix typo

* split slice related code into 2 part for imperative and eager

* split slice from inference

* split slice from inference

* fix test_tensor_register_hook

* support custom op in eager mode

* fix inference deps error

* split eager utils from custom operator

* fix type match

* fix typo
Co-authored-by: NWang Huan <wanghuan29@baidu.com>
Co-authored-by: NWeilong Wu <veyron_wu@163.com>
Co-authored-by: Nwanghuancoder <wanghuancoder@163.com>

227fa408

Z

Adjust Yaml name parsing to satisfy Sparse-related APIs (#40480) · d6e99fe4
由 Zhanlue Yang 提交于 3月 14, 2022

d6e99fe4

Refine partial_program for new run_program OP (#40355) · afafb1c3

由 0x45f 提交于 3月 14, 2022

* refine partial_program

* fix code for test_mnist.py train

* support quantify UT

* make __fake_vars and _double_grads to lazy

* fix comments

afafb1c3

12 3月, 2022 1 次提交
- C
  Fix eager benchmark test failed (#40468) · 70f83f1d
  由 Chen Weihang 提交于 3月 12, 2022
```
* fix eager benchmark test failed

* fix test_tracer failed
```
  70f83f1d
11 3月, 2022 4 次提交

Z

Added Final State Matmul_v2 to C++ performance test (#40391) · 6d830f6c
由 Zhanlue Yang 提交于 3月 11, 2022

6d830f6c

[Phi] Remove needless deps in unittests (#40256) · 89ed57e2

由 Chen Weihang 提交于 3月 11, 2022

* remove needless deps in unittests

* add gpu marco

* fix other unittests

* fix kernel name error

* fix test_prepare_op

* fix failed dygraph unittests

* fix gpu failed tests

* fix cinn test failed

* fix cinn test failed

* fix dropout tests

89ed57e2

[Phi] Reduce grad (#40263) · f452ad5c

由 chentianyu03 提交于 3月 11, 2022

* add reduce_sum grad kernel

* add reduce_grad

* modify reduce grad

* update reduce grad functions

* fix build error

* add argument mapping

* move cast input after grad

* add dims.size=1 cpu reduce_sum grad compute method

* update reduce grad GPU

* remove raw reduce_sum_grad kernel

* modify header files

* add namespace funcs for reduce_grad_funcstions

f452ad5c

Z
Fixed issues with intermediate kernels (#40266) · 9ebe7276
由 Zhanlue Yang 提交于 3月 11, 2022
```
* Fix issues with intermediate kernels

* Fixed CI issues
```
9ebe7276

10 3月, 2022 1 次提交
- Z
  
  Supported auto code gen for sparse kernels (#40276) · 2b6da4de
  由 Zhanlue Yang 提交于 3月 10, 2022
  
  2b6da4de
09 3月, 2022 3 次提交
- Z
  [PHI] Fix some bug of code auto-gen in C++ API (#40262) · 63fb0347
  由 zyfncg 提交于 3月 09, 2022
```
* support code auto-gene for sparse backward api

* fix bug of intermediate api and name of return var
```
  63fb0347
- 0
  adapt run_program OP for eager (#40198) · 3e9601ba
  由 0x45f 提交于 3月 09, 2022
```
* adapt run_program OP for eager

* fix program_id

* refine code

* fix test
```
  3e9601ba
- J
  
  remove additional deps of phi (#40251) · 7f43055d
  由 Jiabin Yang 提交于 3月 09, 2022
  
  7f43055d
08 3月, 2022 1 次提交
- Z
  [PHI] Support string type attr in yaml (#40218) · 47d1d5af
  由 zyfncg 提交于 3月 08, 2022
```
* support str attr in yaml

* fix bug
```
  47d1d5af
04 3月, 2022 3 次提交
- Z
  Fixed GradNode default attributes issues (#40132) · 0f925968
  由 Zhanlue Yang 提交于 3月 04, 2022
```
* Fixed GradNode default attributes issues

* Reverted changes on yaml files
```
  0f925968
- Z
  Generate forward-only operators (#39962) · a6947991
  由 Zhanlue Yang 提交于 3月 04, 2022
```
* [Eager][Yaml]Supported Scalar and ScalarArray for AutoCodeGen

* Generate forward-only operators

* [Yaml]Support parsing fwd & bwd returns with name

* Fixed issues

* Fixed minor issues
```
  a6947991
- Z
  
  [Yaml]Support parsing fwd & bwd returns with name (#40107) · d2a911b4
  由 Zhanlue Yang 提交于 3月 04, 2022
  
  d2a911b4
03 3月, 2022 3 次提交

Z

[Eager][Yaml]Supported Scalar and ScalarArray for AutoCodeGen (#40080) · 97ccaa79
由 Zhanlue Yang 提交于 3月 03, 2022

97ccaa79

Support slim eager (#39874) · da47544c

由 Jiabin Yang 提交于 3月 03, 2022

* eager, test=develop

* fix bug, test=develop

* eager, test=develop

* merge legacy to fluid

* eager, test=develop

* eager, test=develop

* Refactor TensorAdd func by template and remove gradient_accumulation in eager

* Remove needless target name

* eager, test=develop

* eager, test=develop

* Use overload instead of template

* Remove legacy code

* Remove legacy code

* selectedrows, test=develop

* Remove DataType test

* eager, test=develop

* eager, test=develop

* support gan, test=develop

* Using Tensor directly instead of using EagerTensor

* support gradient_accumulation

* make test_imperative_lod_tensor_to_selected_rows longer

* make test_imperative_lod_tensor_to_selected_rows longer

* refine code

* ptb, test=develop

* Rename all EagerTensor to Tensor

* Rename some EagerTensor to Tensor

* rename EagerTensor to EagerVariable

* eager, test=develop

* eager, test=develop

* eager, test=develop

* eager, test=develop

* add more test

* eager, test=develop

* Support copiable selected rows and merge develop

* save load, eager, test=develop

* save load, eager, test=develop

* refine, test=develop

* remove useless _set_value method

* refine, test=develop

* refine, test=develop

* revert static_runner, test=develop

* EagerTensor to Tensor, test=develop

* refine, test=develop

* refine, test=develop

* clear grad, test=develop

* merge, develop

* merge, develop

* merge, test=develop

* merge, test=develop

* Support quant and part of slice

* support legacy static save

* extend slim tests time

* remove imperative on inference

* remove imperative on inference

* merge develop

* fix typo

* fix typo

* split slice related code into 2 part for imperative and eager

* split slice from inference

* split slice from inference

* fix test_tensor_register_hook
Co-authored-by: NWang Huan <wanghuan29@baidu.com>
Co-authored-by: NWeilong Wu <veyron_wu@163.com>
Co-authored-by: Nwanghuancoder <wanghuancoder@163.com>

da47544c

Z

[Eager][YAML] Supported array-type parsing for output tensors (#40058) · 71c69507
由 Zhanlue Yang 提交于 3月 03, 2022

71c69507

02 3月, 2022 1 次提交

[Eager] open eager when WITH_PYTHON (#39979) · 9af72957

由 wanghuancoder 提交于 3月 02, 2022

* open eager when WITH_PYTHON, test=develop

* refine, test=develop

* refine, test=develop

* add DWITH_PYTHON for gen_fluid_lib, test=develop

9af72957

01 3月, 2022 2 次提交

Z

Added attr & tensor type mapping for final state codegen (#39997) · 852a872f
由 Zhanlue Yang 提交于 3月 01, 2022

852a872f

Fixed auto codegen for intermediate tensors (#39797) · 2592805b

由 Zhanlue Yang 提交于 3月 01, 2022

* Refactored GradNodeAccumulation data structure and behaviour

* Fixed CI issues

* Fix compilation issues

* Fixed minor issues

* Reverted changes for intermediate and OverwriteOutput

* fixed minor issue

* Fixed auto codegen for intermediate tensors

* Removed restriction on AccumulationNode modification

* Fixed CI Coverage issues

* Adjusted Log contents

* Fixed CI issues

2592805b

28 2月, 2022 1 次提交

[Pten->Phi PR4] Rename pten in funcs to phi (#39961) · eb42dd52

由 Chen Weihang 提交于 2月 28, 2022

* rename pten_utils to phi_utils

* rename pten_utils target

* rename Pten to Phi

* replace pten with phi

* resolve conflict

eb42dd52

26 2月, 2022 1 次提交
- W
  [Eager Hook] Support GradientHook and ReduceHook, expose related interface to python (#39893) · a456dda6
  由 Weilong Wu 提交于 2月 26, 2022
```
* Support Eager Hook, expose interface to python

* Fix CI issue
```
  a456dda6
25 2月, 2022 1 次提交
- Z
  
  Fixed Python-C AutoCodeGen issues (#39897) · b56ac35c
  由 Zhanlue Yang 提交于 2月 25, 2022
  
  b56ac35c
24 2月, 2022 1 次提交
- C
  [PTen->Phi PR3] Rename pten make target to phi (#39832) · f77019a0
  由 Chen Weihang 提交于 2月 24, 2022
```
* rename pten to phi

* fix infrt compile failed

* resolve conflict
```
  f77019a0

PaddlePaddle / Paddle 1 年多 前同步成功

PaddlePaddle / Paddle
1 年多前同步成功