提交 · 44e0393c1c4675125778da3c555b6e0093f8810d · PaddlePaddle / Paddle

12 6月, 2023 3 次提交
- N
  
  bump black to 2023 style (#54523) · 44e0393c
  由 Nyakku Shigure 提交于 6月 12, 2023
  
  44e0393c
- Z
  
  refine attribute name (#54516) · e73ddd6c
  由 zhangbo9674 提交于 6月 12, 2023
  
  e73ddd6c
- R
  Revert "【第三方库离线编译】add pslib, pslib_brpc cache (#54466)" (#54541) · b04689e9
  由 risemeup1 提交于 6月 12, 2023
```
This reverts commit df1f2b5d.
```
  b04689e9
11 6月, 2023 5 次提交
- R
  Revert "[cmake] fix cmake with the same name already exists (#54391)" (#54539) · 667c8707
  由 risemeup1 提交于 6月 11, 2023
```
This reverts commit a6b9120a.
```
  667c8707
- R
  Revert "Fix compile error with C++17 on distribution (#54411)" (#54538) · 9037c1c6
  由 risemeup1 提交于 6月 11, 2023
```
This reverts commit 8b1839be.
```
  9037c1c6
- R
  【第三方库离线编译】add pslib, pslib_brpc cache (#54466) · df1f2b5d
  由 RedContritio 提交于 6月 11, 2023
```
* add pslib cache

* fix bug
```
  df1f2b5d
- G
  [cmake] fix cmake with the same name already exists (#54391) · a6b9120a
  由 gouzil 提交于 6月 11, 2023
```
* [cmake] fix cmake error

* [cmake]add rm build file
```
  a6b9120a
- H
  Fix compile error with C++17 on distribution (#54411) · 8b1839be
  由 huangjiyi 提交于 6月 11, 2023
```
* “update”

* update

* update

* update

* update

* test

* update
```
  8b1839be
10 6月, 2023 4 次提交
- L
  
  Fix bugs in fused_linear_epilogue (#54512) · 0a704e14
  由 limingshu 提交于 6月 10, 2023
  
  0a704e14
- H
  
  [CustomOP] delete experimental namespace of sigmoid (#54479) · 8f07f503
  由 HongyuJia 提交于 6月 10, 2023
  
  8f07f503
- L
  
  fix segment fault of protobuf (#54511) · d740f67a
  由 Leo Chen 提交于 6月 10, 2023
  
  d740f67a
- Z
  [IR] Fix ir StorageManager bug (#54503) · 3020ad03
  由 zhangbo9674 提交于 6月 10, 2023
```
* fix ir storage bug

* refine code

* refine code

* fix bug

* refine code

* refine code
```
  3020ad03
09 6月, 2023 13 次提交
- W
  
  [IR] add positon member in new ir operation. (#54483) · 8dfcf03e
  由 winter-wang 提交于 6月 09, 2023
  
  8dfcf03e
- P
  support add(x_float32, bfloa16_) or add(x_float32, y_float16) (#54415) · b3232936
  由 pangengzheng 提交于 6月 09, 2023
```
* support add(x_float32, bfloa16_) or add(x_float32, y_float16)

* polish

* fix test
```
  b3232936
- C
  
  Support gcc12 [CUDA<12.0] & fix some bugs (#54437) · 06b8fbb0
  由 chalsliu 提交于 6月 09, 2023
  
  06b8fbb0
- J
  
  [CINN] reopen cumsum prim+cinn unittest (#54494) · 2e4ac019
  由 jiangcheng 提交于 6月 09, 2023
  
  2e4ac019
- L
  Auto generate code for elementwise_max (#54412) · 57564bdf
  由 lzydev 提交于 6月 09, 2023
```
* auto generate code for elementwise_max

* auto generate code for elementwise_max

* fix composite ops

* fix bug of fmax
```
  57564bdf
- X
  modify multiply_grad_node create to Reduce memory in special condition (#54383) · f3d4c78f
  由 xiaoguoguo626807 提交于 6月 09, 2023
```
* modify multiply_grad_node create

* add place choose

* segment fault modify
```
  f3d4c78f
- K
  [IR] Adapt startup program (#54452) · 1681edc8
  由 kangguangli 提交于 6月 09, 2023
```
* adapt_startup_program

* refactor program translator

* polish
```
  1681edc8
- Y
  
  [AMP] Polish the documentation of check_numerics. test=document_fix (#54457) · 8a1af374
  由 Yiqun Liu 提交于 6月 09, 2023
  
  8a1af374
- J
  
  fix reduce.h (#54476) · 2dcc622d
  由 jiangfan06 提交于 6月 09, 2023
  
  2dcc622d
- Z
  [IR] Refine IR builder and throw methods (#54396) · 3a452e4e
  由 zhangbo9674 提交于 6月 09, 2023
```
* refine code

* refine code

* refine code

* refine code

* refine code

* refine code

* refine code

* fix bug

* refine code

* refine code

* refine code

* refine code

* refine code

* delete unused code

* delete unused code

* refine code
```
  3a452e4e
- Z
  
  refine xpu inference api (#54342) · b62b384b
  由 zhupengyang 提交于 6月 09, 2023
  
  b62b384b
- N
  bump ruff to 0.0.272 and update config (#54449) · 8f65f72e
  由 Nyakku Shigure 提交于 6月 09, 2023
```
* bump ruff to 0.0.271 and update config

* exclude third_party

* bump ruff to 0.0.272

* refine config
```
  8f65f72e
- H
  
  [XPU] add registration of SplitWithNumKernel with int64. (#54478) · 4a77cf53
  由 houj04 提交于 6月 09, 2023
  
  4a77cf53
08 6月, 2023 15 次提交
- 周
  [Paddle Inference] remove redunant op (#54442) · 24efc13b
  由周周周提交于 6月 08, 2023
```
* remove redunant
```
  24efc13b
- J
  [CINN] add `paddle.framework.core.is_run_with_cinn()` api for running check (#54355) · a0d59f9d
  由 jiangcheng 提交于 6月 08, 2023
```
* [CINN] add paddle.device.is_run_with_cinn() api for running check

* for pass ci converge

* add check whl should compiled with cinn

* fix ci coverage bug

* remove exposed API

* import is_run_with_cinn explicitly
```
  a0d59f9d
- C
  
  enable sort ip in launch (#54435) · f3eded6c
  由 Chitsing KUI 提交于 6月 08, 2023
  
  f3eded6c
- L
  eager call all2all to avoid p2p hang in lazy init (#54431) · 56fd25b8
  由 Leo Chen 提交于 6月 08, 2023
```
* eager call all2all to avoid p2p hang in lazy init

* update
```
  56fd25b8
- Y
  
  add warning to pp (#54448) · 2a2af7d7
  由 Yuang Liu 提交于 6月 08, 2023
  
  2a2af7d7
- C
  [AMP Prim OP]support some prim ops for bf16 dtype part3 (#54368) · e64a18db
  由 Charles-hit 提交于 6月 08, 2023
```
* support some prim ops bf16 dtype

* fix cmake
```
  e64a18db
- W
  
  [XPU]add fp16 kernels (#54410) · fd9c555c
  由 wz1qqx 提交于 6月 08, 2023
  
  fd9c555c
- Y
  
  xpu support auto growth allocator (#54121) · 168fac13
  由 ykkk2333 提交于 6月 08, 2023
  
  168fac13
- A
  
  [IR]Polish BFloat16Type code and Add related UT (#54440) · 78fc636e
  由 Aurelius84 提交于 6月 08, 2023
  
  78fc636e
- H
  [IR]add kernel dialect (#54428) · 950a29b2
  由 hong 提交于 6月 08, 2023
```
* add kernel dialect

* change DenseTensorTypeStorage to DenseTensorType

* add test case`
```
  950a29b2
- R
  
  delete requests (#54429) · b3da971d
  由 risemeup1 提交于 6月 08, 2023
  
  b3da971d
- C
  [AMP Prim OP]support some prim ops for bf16 dtype part5 (#54422) · bb89c0c8
  由 Charles-hit 提交于 6月 08, 2023
```
* support some prim ops for bf16 dtype

* remove useless code
```
  bb89c0c8
- Y
  [AMP] Add check_numerics API. (#54301) · a5444592
  由 Yiqun Liu 提交于 6月 08, 2023
```
* Add outputs to check_numerics_kernel.

* Add check_numerics to yaml.

* Add API and unittest.

* Add check_nan_inf_level as argument of check_numerics_kernel.

* Add more unittests.

* Fix static API implementation and unittest.

* Move the implementation of check_numerics to paddle.amp.

* Fix import error.
```
  a5444592
- Y
  
  output tensor hook support while op (#54432) · 418d2796
  由 Yuanle Liu 提交于 6月 08, 2023
  
  418d2796
- R
  Revert "【第三方库离线编译】add pslib, pslib_brpc cache (#54370)" (#54441) · 2f781a34
  由 risemeup1 提交于 6月 08, 2023
```
This reverts commit dd99a115.
```
  2f781a34

PaddlePaddle / Paddle 大约 1 年 前同步成功

PaddlePaddle / Paddle
大约 1 年前同步成功