提交 · c48a9ad56e69a5d27d1b36df8c731c9c32f84d78 · BaiXuePrincess / Paddle

17 1月, 2022 3 次提交

[Pten] Replace platform::Place to pten::Place. (#38899) · c48a9ad5

由 Wilber 提交于 1月 17, 2022

* add pten::Place data structure.

* update ci problem

* fix ci problem

* update

* using platform::Place=pten::Place

* remove BOOST_GET_CONST for CPUPlace and GPUPlace

* compile pass 25%.

* compile pass 45%

* compile pass 60%

* remove boost_get for xpu npu mlu and ipu

* compile pass on cpu and gpu.

* fix compile problem

* fix compile error.

* update

* fix ci problem

* update

* ci approve

* fix ci problem

* fix ci eager test problem

* remove BOOST_GET_CONST

* fix npu compile

c48a9ad5

S
Add NoReduce mode for ParallelExecutor (#38969) · e50d883e
由 sneaxiy 提交于 1月 17, 2022
```
* add no reduce mode for pe

* add NoReduce ut
```
e50d883e
J
Support auto prune logic in eager mode (#38960) · f81569e3
由 Jiabin Yang 提交于 1月 17, 2022
```
* support test_auto_prune_partial

* support rest of autoprune strategy in eager mode
```
f81569e3

15 1月, 2022 1 次提交

[Unify Tensors PR #7] Merged LoDTensor with Tensor, test=allcases (#38880) · 88966b28

由 Zhanlue Yang 提交于 1月 15, 2022

* Merged LoDTensor with Tensor,test=allcases

* Patched python level LoDTensor

* Fixed example code failure

* Polished function names, removed duplicated forward declarations

88966b28

14 1月, 2022 2 次提交
- Q
  [MLU]Add mean and reduce_mean op (#38872) · 7f8d5bc8
  由 qipengh 提交于 1月 14, 2022
```
* [MLU]: add mean and reduce mean op

* [MLU]add mlu pytest dir in CMakeLists.txt

* [MLU]fix tensor data

* [MLU]fix TensorToPyArray and license
```
  7f8d5bc8
- 石
  
  remove interface: DenseTensor::release, test=develop (#38937) · 9ff989ae
  由石晓伟提交于 1月 14, 2022
  
  9ff989ae
13 1月, 2022 2 次提交

[pten]Remove pten/include dir files (#38878) · 7e0292ea

由 chentianyu03 提交于 1月 13, 2022

* move dot_dev api into dot_kernel.h

* add infermate header

* modify to dotkerel in dot_op.h

* mvoe conj dev api into complex_kernel.h

* move sign dev api into  sign_kernel.h

* move scale dev api into kernel.h and remove infermete.h

* rm paddle/pten/include/math.h

* rm paddle/pten/include/math.h

* rm include dir

* rm paddle/pten/include/math.h

* fix conflict with develop branch

* rm devContext in conj_op.h

* add the missing complex_kernel header

7e0292ea

石

splits allocation for pten, test=develop (#38853) · 277cf900
由石晓伟提交于 1月 13, 2022

277cf900

12 1月, 2022 1 次提交
- J
  
  support test_auto_prune_partial (#38871) · 4640955c
  由 Jiabin Yang 提交于 1月 12, 2022
  
  4640955c
11 1月, 2022 1 次提交

[Eager] fix some eager logic (#38576) · d3686471

由 wanghuancoder 提交于 1月 11, 2022

* Rearranged Eager AutoCodeGen directory structure

* Removed USE_OP in Eager AutoCodeGen

* Enabled generation for Operators without Grad/Inputs/Outputs

* Resolved operators without input

* Fixed merge conflicts

* Enabled Eager AutoCodeGen for 10+ more operators

* Refactored Eager AutoCodeGen with more organized helper objects

* Enabled Eager AutoCodeGen for operators with multiple OpBases

* Adjusted Eager AutoCodeGen to Enable Passing Output Tensor as Input Argument

* Handled Dispensable Inputs/Outputs in Eager AutoCodeGen

* Adjusted function generation/call between Python-C API & Dygraph API

* Synchronized auto-generated Python-C API with Dygraph Forward Functions

* support more eager tensor api

* fix merge compile error

* fix compile error and fit develop code

* support pure CPU

* fix some logic error in eager_mode

* support _varbase_creator in eager mode

* Added safe_initialized interface to EagerTensor for use in processing dispensable inputs

* for eager mode

* refine

* support multiple constructor for eager tensor

* add place related code

* polish code

* specific randint with dtype of int64

* Support pure cpu test

* eager logic

* refine test in pure cpu

* eager logic

* eager logic

* eager logic, test=develop

* skip core.eager when in inference, test=develop

* refine, test=develop

* refine, test=develop

* call RetainGrad after run forward kernel, test=develop

* refine, test=develop

* support dygraph util, meta, guard test

* eager test case

* support inference test

* refine test and fix initializer failed

* modify eagertensor patch method

* add eagertensor.clear_grandint, test=develop

* refine, test=develop

* refine, test=develop

* refine, test=develop

* support create varbase and fix retain grad error

* call monkey_patch_varbase in _test_eager_guard, test=develop

* fix windows error

* split clear_gradient to clear_gradient and zero_grads, test=develop

* refine, test=develop

* refine, test=develop

* support test_imperative_basic test in eager mode

* remove additional log in variable.h

* remove additional log in variable.h

* remove additional code create in merge

* eager

* fix some eager logic, test=develop

* refine, test=develop

* refine, test=develop

* refine, test=develop

* refine, test=develop

* refine, test=develop
Co-authored-by: Njim19930609 <jim19930609@gmail.com>
Co-authored-by: NJiabinYang <360788950@qq.com>

d3686471

10 1月, 2022 2 次提交

Y

[fleet_executor] framework for big model inference (#38795) · ededcda2
由 Yuang Liu 提交于 1月 10, 2022

ededcda2

[Unify Tensors PR #5] framework::Tensor inherits from DenseTensor,test=allcases (#38632) · 5c73a6ea

由 Zhanlue Yang 提交于 1月 10, 2022

* Added shared_ptr<Allocation> member & corresponding interfaces to Storage

* Removed original pten::Allocation from Storage and adjusted the interfaces accordingly

* Fixed issues with storage offset

* Used place to malloc allocation for TensorStorage

* [Unify Tensors PR #3]Ported framework::Tensor interfaces to pten::DenseTensor

* Fixed issues with place

* Added comments

* Moved mutable_data with stream argument to DenseTensor

* Added set_offset interface

* Fixed CI issues,test=allcases

* [Unify Tensors PR #4] Port LoDTensor interfaces to DenseTensor

* Removed friend class EigenTensor/EigenMatrix/EigenVector from Tensor

* Modified framework::Tensor to inherit from DenseTensor

* Reverted changes too pten_layout() interface

* Removed friend classes

* Rearranged cfunction calls from tensor.data<void>() to tensor.data()

* Fixed CI issues

* Fixed lite issues

* Fixed data() interface issues,test=allcases

* Resolved IsInitialized() issues

* Fixed ResetHolder() issues

* Fixed MKLDNN & Storage issues

* Resolved ShareBufferWith() issues

* Fixed LoD issues

5c73a6ea

07 1月, 2022 1 次提交

Add multi tensor for adam (#38010) · fb3313e9

由 zhangbo9674 提交于 1月 07, 2022

* add multi tensor for adam

* add merged_adam op

* refine code

* refine adam compute logic

fb3313e9

06 1月, 2022 1 次提交
- W
  
  fix slot, test=develop (#38738) · 4514f16d
  由 wanghuancoder 提交于 1月 06, 2022
  
  4514f16d
05 1月, 2022 2 次提交

[Eager] Support test imperative basic in eager test_empty_grad (#38376) · 9108e777

由 wanghuancoder 提交于 1月 05, 2022

* Rearranged Eager AutoCodeGen directory structure

* Removed USE_OP in Eager AutoCodeGen

* Enabled generation for Operators without Grad/Inputs/Outputs

* Resolved operators without input

* Fixed merge conflicts

* Enabled Eager AutoCodeGen for 10+ more operators

* Refactored Eager AutoCodeGen with more organized helper objects

* Enabled Eager AutoCodeGen for operators with multiple OpBases

* Adjusted Eager AutoCodeGen to Enable Passing Output Tensor as Input Argument

* Handled Dispensable Inputs/Outputs in Eager AutoCodeGen

* Adjusted function generation/call between Python-C API & Dygraph API

* Synchronized auto-generated Python-C API with Dygraph Forward Functions

* support more eager tensor api

* fix merge compile error

* fix compile error and fit develop code

* support pure CPU

* fix some logic error in eager_mode

* support _varbase_creator in eager mode

* Added safe_initialized interface to EagerTensor for use in processing dispensable inputs

* for eager mode

* refine

* support multiple constructor for eager tensor

* add place related code

* polish code

* specific randint with dtype of int64

* Support pure cpu test

* eager logic

* refine test in pure cpu

* eager logic

* eager logic

* eager logic, test=develop

* skip core.eager when in inference, test=develop

* refine, test=develop

* refine, test=develop

* call RetainGrad after run forward kernel, test=develop

* refine, test=develop

* support dygraph util, meta, guard test

* eager test case

* support inference test

* refine test and fix initializer failed

* modify eagertensor patch method

* add eagertensor.clear_grandint, test=develop

* refine, test=develop

* refine, test=develop

* refine, test=develop

* call monkey_patch_varbase in _test_eager_guard, test=develop

* split clear_gradient to clear_gradient and zero_grads, test=develop

* refine, test=develop

* refine, test=develop

* refine, test=develop
Co-authored-by: Njim19930609 <jim19930609@gmail.com>
Co-authored-by: NJiabinYang <360788950@qq.com>

9108e777

Support EagerTensor initialization with kwargs (#38488) · 4ba6d4e4

由 Weilong Wu 提交于 1月 05, 2022

* Support EagerTensor init with kwargs

* Updated comments

* Updated unit tests case

* Refactor InitTensor related code to reduce duplicate code

* Updated the error reporting msg

* Updated VLOG msg

* Merge develop and Update EagerTensor init func

* Polish switch case, reduce some code

* Add SyntaxError unit test case

* Refactor the related initialization func of EagerTensor

* Remove ParseStopGradient and ParseZeroCopy and ParsePersistable, construct ParseBooleanArgs instead.

* Updated error msg to pass CI

* Updated PADDLE_ENFORCE error type

4ba6d4e4

31 12月, 2021 2 次提交
- F
  [MLU]support calling mlu op from python interface (#38292) · b6bf650a
  由 fwenguang 提交于 12月 31, 2021
```
* [MLU]support calling mlu op from python interface

* [MLU]fix

* fix

* [mlu]fix mlu_places

* [mlu]fix required mlu

* fix

* [MLU]fix tensor copy

* [mlu] fix MLUPlace call path
```
  b6bf650a
- W
  
  fix python ascend run error. (#38605) · 1df354e7
  由 Wilber 提交于 12月 31, 2021
  
  1df354e7
30 12月, 2021 2 次提交

Y
[Auto parallel] Make sure the id semantics of every var and op unique (#38132) · 5620214e
由 Yulong Ao 提交于 12月 30, 2021
```
* [Auto parallel] Make the id of var and op unique

* [Auto Parallel] Rename back dist_context to distop_context
```
5620214e

Support test imperative basic with fixed retain grad interface (#38548) · 2421a25a

由 Jiabin Yang 提交于 12月 30, 2021

* Rearranged Eager AutoCodeGen directory structure

* Removed USE_OP in Eager AutoCodeGen

* Enabled generation for Operators without Grad/Inputs/Outputs

* Resolved operators without input

* Fixed merge conflicts

* Enabled Eager AutoCodeGen for 10+ more operators

* Refactored Eager AutoCodeGen with more organized helper objects

* Enabled Eager AutoCodeGen for operators with multiple OpBases

* Adjusted Eager AutoCodeGen to Enable Passing Output Tensor as Input Argument

* Handled Dispensable Inputs/Outputs in Eager AutoCodeGen

* Adjusted function generation/call between Python-C API & Dygraph API

* Synchronized auto-generated Python-C API with Dygraph Forward Functions

* support more eager tensor api

* fix merge compile error

* fix compile error and fit develop code

* support pure CPU

* fix some logic error in eager_mode

* support _varbase_creator in eager mode

* Added safe_initialized interface to EagerTensor for use in processing dispensable inputs

* for eager mode

* refine

* support multiple constructor for eager tensor

* add place related code

* polish code

* specific randint with dtype of int64

* Support pure cpu test

* eager logic

* refine test in pure cpu

* eager logic

* eager logic

* eager logic, test=develop

* skip core.eager when in inference, test=develop

* refine, test=develop

* refine, test=develop

* call RetainGrad after run forward kernel, test=develop

* refine, test=develop

* support dygraph util, meta, guard test

* support inference test

* refine test and fix initializer failed

* support create varbase and fix retain grad error

* fix windows error

* support test_imperative_basic test in eager mode

* remove additional log in variable.h

* remove additional log in variable.h

* remove additional code create in merge
Co-authored-by: Njim19930609 <jim19930609@gmail.com>
Co-authored-by: NWang Huan <wanghuan29@baidu.com>

2421a25a

29 12月, 2021 1 次提交
- Y
  add dynamic mf size api (#38436) · 7411dab5
  由 yaoxuefeng 提交于 12月 29, 2021
```
add dynamic mf size api
```
  7411dab5
28 12月, 2021 1 次提交

Support test basic of Var and Layer (#38426) · 1fb80a6a

由 Jiabin Yang 提交于 12月 28, 2021

* Rearranged Eager AutoCodeGen directory structure

* Removed USE_OP in Eager AutoCodeGen

* Enabled generation for Operators without Grad/Inputs/Outputs

* Resolved operators without input

* Fixed merge conflicts

* Enabled Eager AutoCodeGen for 10+ more operators

* Refactored Eager AutoCodeGen with more organized helper objects

* Enabled Eager AutoCodeGen for operators with multiple OpBases

* Adjusted Eager AutoCodeGen to Enable Passing Output Tensor as Input Argument

* Handled Dispensable Inputs/Outputs in Eager AutoCodeGen

* Adjusted function generation/call between Python-C API & Dygraph API

* Synchronized auto-generated Python-C API with Dygraph Forward Functions

* support more eager tensor api

* fix merge compile error

* fix compile error and fit develop code

* support pure CPU

* fix some logic error in eager_mode

* support _varbase_creator in eager mode

* Added safe_initialized interface to EagerTensor for use in processing dispensable inputs

* for eager mode

* refine

* support multiple constructor for eager tensor

* add place related code

* polish code

* specific randint with dtype of int64

* Support pure cpu test

* eager logic

* refine test in pure cpu

* eager logic

* eager logic

* eager logic, test=develop

* skip core.eager when in inference, test=develop

* refine, test=develop

* refine, test=develop

* call RetainGrad after run forward kernel, test=develop

* refine, test=develop

* support dygraph util, meta, guard test

* support inference test

* refine test and fix initializer failed

* support create varbase and fix retain grad error

* fix windows error

* support test code coverage

* support test code coverage

* support test code coverage
Co-authored-by: Njim19930609 <jim19930609@gmail.com>
Co-authored-by: NWang Huan <wanghuan29@baidu.com>

1fb80a6a

26 12月, 2021 1 次提交

[PTen] Move copy kernel impl (#38421) · 73819658

由 Chen Weihang 提交于 12月 26, 2021

* add register general kernel marco

* move copy kernel impl

* revert needless change

* polish details

* fix xpu compil faild

* fix xpu compile failed

* polish format

73819658

24 12月, 2021 3 次提交

Z

[AMP] Add multi_precision for sgd (#38231) · a4d07bb9
由 zhangbo9674 提交于 12月 24, 2021

a4d07bb9
B

fix share buffer to (#38407) · 9409ff6b
由 Baibaifan 提交于 12月 24, 2021

9409ff6b

Support test imperative basic in eager (#38313) · d48f7c89

由 Jiabin Yang 提交于 12月 24, 2021

* Rearranged Eager AutoCodeGen directory structure

* Removed USE_OP in Eager AutoCodeGen

* Enabled generation for Operators without Grad/Inputs/Outputs

* Resolved operators without input

* Fixed merge conflicts

* Enabled Eager AutoCodeGen for 10+ more operators

* Refactored Eager AutoCodeGen with more organized helper objects

* Enabled Eager AutoCodeGen for operators with multiple OpBases

* Adjusted Eager AutoCodeGen to Enable Passing Output Tensor as Input Argument

* Handled Dispensable Inputs/Outputs in Eager AutoCodeGen

* Adjusted function generation/call between Python-C API & Dygraph API

* Synchronized auto-generated Python-C API with Dygraph Forward Functions

* support more eager tensor api

* fix merge compile error

* fix compile error and fit develop code

* support pure CPU

* fix some logic error in eager_mode

* support _varbase_creator in eager mode

* Added safe_initialized interface to EagerTensor for use in processing dispensable inputs

* for eager mode

* refine

* support multiple constructor for eager tensor

* add place related code

* polish code

* specific randint with dtype of int64

* Support pure cpu test

* eager logic

* refine test in pure cpu

* eager logic

* eager logic

* eager logic, test=develop

* skip core.eager when in inference, test=develop

* refine, test=develop

* refine, test=develop

* call RetainGrad after run forward kernel, test=develop

* refine, test=develop

* support dygraph util, meta, guard test

* support inference test

* refine test and fix initializer failed
Co-authored-by: Njim19930609 <jim19930609@gmail.com>
Co-authored-by: NWang Huan <wanghuan29@baidu.com>

d48f7c89

23 12月, 2021 1 次提交
- add new API: paddle.clone;Tensor.element_size;nn.utils.parameters_to_vector (#38020) · 0eb03ed7
  由 zhouweiwei2014 提交于 12月 23, 2021
```
* add new API: paddle.clone;Tensor.element_size;nn.utils.parameters_to_vector

* fix comment
```
  0eb03ed7
21 12月, 2021 2 次提交
- Z
  Fix inplace problem of setitem (#38298) · da61df5c
  由 zyfncg 提交于 12月 21, 2021
```
* add inplace_map for trace_op in pybind

* fix inplace problem of setitem

* refactor the param format  of trace_op
Co-authored-by: Npangyoki <pangyoki@126.com>
```
  da61df5c
- Y
  
  [fleet_executor] Python side fleet executor and task node (#38290) · a4afb97a
  由 Yuang Liu 提交于 12月 21, 2021
  
  a4afb97a
20 12月, 2021 3 次提交

F

[MLU]add mlu backend (#38207) · 76514a1f
由 fwenguang 提交于 12月 20, 2021

76514a1f

Add multi_tensor for momentum optimizer and clear_grads (#37564) · 0cc5e22c

由 zhangbo9674 提交于 12月 20, 2021

* add multi_tensor for momentum and clear_grads for optimizer

* fix bug for dygraph

* add unittest

* refine comment

* add param_group

* refine regularizaiton logic

* del clear_grads

* add clear_grads

* add dispensable check of None

* refine clear_grad

* fix build bug

* refine code by comment

* refine code

* add multi tensor check

* refine param_group update

* add multi tensor for static mode

* refine comments

* delete useless comma for momentum

* refine comment for momentum

* refine code by commment

0cc5e22c

Z

move the directory of fill kernels in pten (#38219) · 06128b9f
由 zyfncg 提交于 12月 20, 2021

06128b9f

17 12月, 2021 5 次提交

Support multi place constructor (#38171) · 6f439e5a

由 Jiabin Yang 提交于 12月 17, 2021

* support more eager tensor api

* support multiple constructor for eager tensor

* add place related code

* polish code

* specific randint with dtype of int64

* Support pure cpu test

* refine test in pure cpu

* refine test in pure cpu

6f439e5a

Refine some AMP operators for BERT (#37923) · d80fe268

由 sneaxiy 提交于 12月 17, 2021

* support multi precision update for LAMB

* hide some api

* fix ci uts

* fix lamb output of dygraph

* remove some changes to some PR

* try to fix Py3 CI compile error

* fix test_imperative_optimizer, add lars ut, add layer_norm ut

* fix ut, fix format

* fix ut

* fix windows ci

d80fe268

Generated CoreOpsInfos for potential use in append_op API (#38085) · e3b033f9

由 Zhanlue Yang 提交于 12月 17, 2021

* Rearranged Eager AutoCodeGen directory structure

* Removed USE_OP in Eager AutoCodeGen

* Enabled generation for Operators without Grad/Inputs/Outputs

* Resolved operators without input

* Fixed merge conflicts

* Enabled Eager AutoCodeGen for 10+ more operators

* Refactored Eager AutoCodeGen with more organized helper objects

* Enabled Eager AutoCodeGen for operators with multiple OpBases

* Adjusted Eager AutoCodeGen to Enable Passing Output Tensor as Input Argument

* Handled Dispensable Inputs/Outputs in Eager AutoCodeGen

* Adjusted function generation/call between Python-C API & Dygraph API

* Synchronized auto-generated Python-C API with Dygraph Forward Functions

* Generated CoreOpsInfos for potential use in append_op API

* Fixed CI problem

e3b033f9

K

add op/api repeat/interleave (#37981) · a7de0e66
由 kuizhiqing 提交于 12月 17, 2021

a7de0e66
Y

[fleet_executor] run time graph on python side (#38164) · fc701369
由 Yuang Liu 提交于 12月 17, 2021

fc701369

16 12月, 2021 3 次提交

J
support eager switch system (#38170) · 8305c2be
由 Jiabin Yang 提交于 12月 16, 2021
```
* support eager switch system

* polish code
```
8305c2be

Add sparse_attention mask ,test=develop (#37973) · fa463b90

由 Liu-xiandong 提交于 12月 16, 2021

Add key_padding_mask and attn_mask in sparse_attention Api

1.Key padding mask is a tensor with dimensions [batch_size, seq_len], and attention mask is a tensor with dimensions [seq_len, seq_len]. The data types of the two masks are consistent with Q, K, and V, which are float32 or float64. If the value in Mask is 0, it means that the position needs to be masked.

2.The changed files are mainly paddle/fluid/operators/sparse_attention_op.cu and python/paddle/fluid/tests/unittests/test_sparse_attention_op.py. sparse_attention has three parts: sddmm, softmax, and dsd. Adding the mask operation only needs to modify the softmax. It has no effect on the other two parts. In addition, in order to test the mask function, related tests has been added.

fa463b90

Enabled Eager AutoCodeGen for All Existing Operators & Possible Future Operators (#37969) · 08482a86

由 Zhanlue Yang 提交于 12月 16, 2021

* Rearranged Eager AutoCodeGen directory structure

* Removed USE_OP in Eager AutoCodeGen

* Enabled generation for Operators without Grad/Inputs/Outputs

* Resolved operators without input

* Fixed merge conflicts

* Enabled Eager AutoCodeGen for 10+ more operators

* Refactored Eager AutoCodeGen with more organized helper objects

* Enabled Eager AutoCodeGen for operators with multiple OpBases

* Adjusted Eager AutoCodeGen to Enable Passing Output Tensor as Input Argument

* Handled Dispensable Inputs/Outputs in Eager AutoCodeGen

* Enabled Eager AutoCodeGen for All Existing Operators & Possible Future Operators

* Fixed CI issues

08482a86

BaiXuePrincess / Paddle 与 Fork 源项目一致

BaiXuePrincess / Paddle
与 Fork 源项目一致