提交 · 9b031026d69988eaa792115a960872ba9e0ccf9b · Crayon鑫 / Paddle

21 6月, 2022 12 次提交
- C
  
  add mlu expand_as_v2 kernel (#43393) · 9b031026
  由 cifar10 提交于 6月 21, 2022
  
  9b031026
- Z
  
  [MLU]: add argsort/argsort_grad kernel (#43574) · d876952d
  由 zhaoying9105 提交于 6月 21, 2022
  
  d876952d
- S
  resort .cu headers, set clang-format not sort include block and consider .cu... · 829723f2
  由 Sing_chan 提交于 6月 21, 2022
```
resort .cu headers, set clang-format not sort include block and consider .cu as main source file (#43633)
```
  829723f2
- N
  
  delete the log printing in layout autotune (#43676) · 195bb3ed
  由 niuliling123 提交于 6月 21, 2022
  
  195bb3ed
- S
  Generalize conv+activation fuse pass (#43382) · 347e4b2e
  由 Sławomir Siwek 提交于 6月 21, 2022
```
* consolidate conv act passes

* generalize conv_activation

* integrate conv+act tests

* code style format

* whitespaces

* remove timeout from old tests

* implement comments from review

* restore ut

* whitespace

* code style

* transpose

* fixes after review

* method for gettin act

* Change Paddle_enforce error type

* code format

* add missing opcompats
```
  347e4b2e
- J
  
  Correct elementwise quantization (#43693) · 9aa89b99
  由 joanna.wozna.intel 提交于 6月 21, 2022
  
  9aa89b99
- Z
  
  slice large tensor for cudnn_softmax (#43681) · bd5e97d3
  由 Zhang Ting 提交于 6月 21, 2022
  
  bd5e97d3
- F
  
  [MLU] add deformable_conv kernel (#43630) · 827d9992
  由 fwenguang 提交于 6月 21, 2022
  
  827d9992
- S
  
  remove .clang-format in paddle/fluid to use the same config (#43678) · b2df4c76
  由 Sing_chan 提交于 6月 21, 2022
  
  b2df4c76
- C
  [MLU] add mlu kernel for elementwise_max_grad (#43608) · f586110d
  由 cambriconhsq 提交于 6月 21, 2022
```
* [MLU] add mlu kernel for elementwise_max_grad

* [MLU] modify mlu kernel elementwise_min_grad impl
```
  f586110d
- A
  [IPU] add activation ops (#43662) · 2353db3a
  由 Allen Guo 提交于 6月 21, 2022
```
* add argmin and argsort ops (#800)

* add argmin and arsort ops

* Add dot bmm ops (#803)

* add bmm

* add dot op

* clean CreateConst

* clean CreateCast

* add activation ops (#808)

* add activation ops

* fix 1function-redefined error
```
  2353db3a
- Z
  
  fix sign-compare warning4 (#43625) · 2a795dfa
  由 zhangchunle 提交于 6月 21, 2022
  
  2a795dfa
20 6月, 2022 12 次提交
- W
  
  Add passes and plugins for distributed inference of NLU (#43049) · 007f3614
  由 whs 提交于 6月 20, 2022
  
  007f3614
- Z
  
  add cross_op cuda kernel (#43558) · ec3e0a13
  由 zhangbopd 提交于 6月 20, 2022
  
  ec3e0a13
- J
  
  fix current expected place error in multicard (#43664) · abbd4abb
  由 Jiabin Yang 提交于 6月 20, 2022
  
  abbd4abb
- J
  Fix for oneDNN layernorm for begin_norm_axis != last_dim (#43476) · b6bc6f7a
  由 jakpiase 提交于 6月 20, 2022
```
* fix for layer_norm

* minor fix
```
  b6bc6f7a
- L
  Revert "Revert md-in-tensor refactoring (#43564)" (#43598) · 8727bb7c
  由 lidanqing 提交于 6月 20, 2022
```
This reverts commit 1ec626b1.
```
  8727bb7c
- C
  
  Update CUDA and TensorRT version for CI · 393e6fea
  由 chalsliu 提交于 6月 20, 2022
  
  393e6fea
- Z
  Support more dimensions in MMHA (#43612) · 03f9e598
  由 Zhang Zheng 提交于 6月 20, 2022
```
* support more dimensions

* fix
```
  03f9e598
- 【Sparse】add new API/OP(csr->csr) of SparseTensor softmax (#43475) · 2ddbc647
  由 zhouweiwei2014 提交于 6月 20, 2022
```
* add new API/OP(csr->csr) of SparseTensor softmax

* fix comment
```
  2ddbc647
- W
  add dymf to gpups in python (#43497) · a4cfa5ae
  由 wangguanqun 提交于 6月 20, 2022
```
* gpups default config and dataset

* codestyle

* add unittest

* code style

* add dymf to gpups

* codestyle

* add static.nn.cvm import

* PSERVER_DEBUG

* add fs config to worker desc

* update unittest

* unittest

* remove gpups unittest

* remove gpups unittest

* static check
```
  a4cfa5ae
- K
  
  fix gpu log typo (#43634) · 87beee54
  由 kuizhiqing 提交于 6月 20, 2022
  
  87beee54
- S
  
  make paddle build and unit test in python virtual environment. (#43620) · 63e2a7dd
  由 Sing_chan 提交于 6月 20, 2022
  
  63e2a7dd
- Z
  
  test=windows_ci (#43374) · f12cb756
  由 zhangchunle 提交于 6月 20, 2022
  
  f12cb756
18 6月, 2022 2 次提交
- W
  
  add cuda11.6 cudnn8.4 trt8.4 image (#43396) · cf0d6027
  由 Wilber 提交于 6月 18, 2022
  
  cf0d6027
- remove unuse cuSparse function (#43626) · 4a08c781
  由 zhouweiwei2014 提交于 6月 18, 2022
  
  4a08c781
17 6月, 2022 14 次提交

Z
fix batch csr (#43553) · 03517d8a
由 zhangkaihuo 提交于 6月 17, 2022
```
* fix to_sparse_csr
```
03517d8a
C

[MLU] add truncated_gaussian_random kernel. (#43575) · 5a5649c2
由 Chenxiao Niu 提交于 6月 17, 2022

5a5649c2

Support optional residual add in fused_attention and fused_feedforward. (#43474) · 19e866f9

由 Yiqun Liu 提交于 6月 17, 2022

* Support optional residual add in fused_attention and fused_feedforward.

* Add checkpoint and add the check of add_residual when pre_layer_norm is false.

* Add TODO and change the python api to add add_residual argument.

19e866f9

F

[MLU]add mlu kernel for where op (#43441) · 2540b023
由 fuyou765 提交于 6月 17, 2022

2540b023
F

[MLU]add mlu kernel for tile op (#43389) · 539a9e60
由 fuyou765 提交于 6月 17, 2022

539a9e60
F

[MLU]add mlu kernel for expand_v2 op (#43353) · 6a179e48
由 fuyou765 提交于 6月 17, 2022

6a179e48
C

[MLU] add mlu kernel for iou_similarity (#43503) · f3a09de4
由 cambriconhsq 提交于 6月 17, 2022

f3a09de4
Q

[MLU]add elementwise op (#43491) · 74cc73bb
由 qipengh 提交于 6月 17, 2022

74cc73bb
Z
[MLU]: add shape kernel (#43347) · feebbe15
由 zhaoying9105 提交于 6月 17, 2022
```
* [MLU]: add shape kernel

* [MLU]: set output from cpu to mlu in shape kernel
```
feebbe15

fix en docs of paddle.vision.transforms.Normalize,... · 4ba2023e

由 Liyulingyue 提交于 6月 17, 2022

fix en docs of paddle.vision.transforms.Normalize, paddle.vision.models.alexnet, paddle.vision.models.mobilenet, ops apis  (#42380)

* Normalize; test=document_fix

* alexnet; test=document_fix

* alexnet; test=document_fix

* vgg; test=document_fix

* mobilenetv2; test=document_fix

* mobilenetv1; test=document_fix

* alexnet; test=document_fix

* PSRoIPool; test=document_fix

* psroi_pool; test=document_fix

* roi_align; test=document_fix

* RoIAlign; test=document_fix

* alexnet; test=document_fix

* MobileNetV1; test=document_fix

* mobilenetv2;test=document_fix

* vgg; test=document_fix

* psroi_pool;test=document_fix

* ops; test=document_fix

* Normalize;test=document_fix

* normlize;test=document_fix

* Update alexnet.py

* for ci;test=document_fix

* for ci;test=document_fix
Co-authored-by: NLigoml <39876205+Ligoml@users.noreply.github.com>

4ba2023e

[PaddleHackathon No.14] (#41183) · 74c74b0f

由 yangguohao 提交于 6月 17, 2022

* 2022-04-28

* 2022-05-04

* 2022-05-05_V1

* 2022-05-05_V1

* Update loss.py

* Update loss.py

* 2022-06-01_hook

* 2022-06-05

* 2022-06-07

* 2022-06-07_V2

* 2022-06-07_V2

* 2022-06-17_codestyle

74c74b0f

L

Modify the large case in test_generate_proposals_v2_op to small for kunlun. *test=kunlun (#43587) · f10d26e9
由 Leo Guo 提交于 6月 17, 2022

f10d26e9

update api in creation.py and search.py; test=document_fix (#43150) · bba7c5b9

由 David Nicolas 提交于 6月 17, 2022

* update api in creation and search.py

* Update creation.py

* Update search.py

* Update search.py

* for ci;test=document_fix

* for ci;test=document_fix
Co-authored-by: NLigoml <39876205+Ligoml@users.noreply.github.com>

bba7c5b9

refactor sum unit test,*test=kunlun (#43561) · 0993c3b2
由 z8hanghuan 提交于 6月 17, 2022

0993c3b2

Crayon鑫 / Paddle 与 Fork 源项目一致

Crayon鑫 / Paddle
与 Fork 源项目一致