提交 · 645e81f0704515915545c23a3c6d452a84e34218 · PaddlePaddle / Paddle

18 5月, 2023 4 次提交

[inference][trt]Remove trt sparse weight api (#53905) · 1007690b

由 Zhang Jun 提交于 5月 18, 2023

* Revert "[inference][trt]add trt sparse weights switch (#53562)"

This reverts commit 4a69a536.

* remove kSPARSE_WEIGHTS

* remove kFASTER_DYNAMIC_SHAPES_0805 and add 'TrtMajorVersion' function

1007690b

Y

adjust inference lib dir (#53091) · 9b0f621c
由 Yuanle Liu 提交于 5月 18, 2023

9b0f621c

Fused elementwises kernels and ops (#51427) · fb4a6ecf

由 Hulek 提交于 5月 18, 2023

* Fused elementwises kernels and ops

* change fuse pass name

* adjust .pbtxt files

* adjust quantization attributes

* add missing arguments and fix others, review fixed

* simplify fused kernel registration

* fix elementwise unit tests

* reuse one fused elementwise op

* adjust proto

* Add supported datatypes

* Change 'Scale' to 'scale' in tests, change some tests to onednn

* Revert breaking changes

* Fix unit tests

* Delete obsolete test cases

* Delete commented out code

* Fix codestyle

* delete temporary condition

* fix conflicts and delete duplicate fusing

* Fix code after merge

* Move tests to new directory

* fix tests volatility

* Rename test_elementwise_add_onednn_op.py to test_elementwise_add_mkldnn_op.py

* Update CMakeLists.txt add mkldnn op test

---------
Co-authored-by: NSilv3S <slawomir.siwek@intel.com>

fb4a6ecf

fix -Werror=format-security (#53886) · 6d7076cc
由 engineer1109 提交于 5月 18, 2023

6d7076cc

16 5月, 2023 3 次提交
- S
  
  [XPU] Add sigmoid_elementmul_xpu_fuse_pass (#53580) · 32e36b15
  由 sprouteer 提交于 5月 16, 2023
  
  32e36b15
- Y
  [Inference] clean unused code/target for reduce inference so volume (PART I) (#53762) · 51ecd933
  由 Yuanle Liu 提交于 5月 16, 2023
```
* remove prelu land ookuip_table plugin, adjust .h include location

* clean code and adjust some .h

* update
```
  51ecd933
- Z
  [inference][trt]Remove unused code from teller.cc (#53758) · 2a94b817
  由 Zhang Jun 提交于 5月 16, 2023
```
* remove unused code
```
  2a94b817
15 5月, 2023 2 次提交

[inference Zero-Dim][trt] Add Zero-Dim tensor support for clip, cast,... · cc9aedaf

由 bukejiyu 提交于 5月 15, 2023

[inference Zero-Dim][trt] Add Zero-Dim tensor support for clip, cast, flatten_contiguous_range (#53769)

* [inference Zero-Dim][trt]clip,cast,flatten_contiguous_range trt op converter support zero dim

cc9aedaf

remove some [-Wunsed-parameter]warning (#53679) · ca2ea162

由 Galaxy1458 提交于 5月 15, 2023

* test,test=develop

* test,test=develop

* test,test=develop

* test,test=develop

* test,test=develop

* test,test=develop

ca2ea162

12 5月, 2023 4 次提交
- R
  
  [CustomDevice] add inference MP support, PART1 (#53702) · c497b43f
  由 ronnywang 提交于 5月 12, 2023
  
  c497b43f
- W
  
  [Inference] Update switch stream logical. (#53589) · eb97f4f0
  由 Wilber 提交于 5月 12, 2023
  
  eb97f4f0
- Y
  [inference zero dim] softmax, stack op trt converter support zero dim (#53729) · 05d3fc81
  由 Yuanle Liu 提交于 5月 12, 2023
```
* softmax support

* support stack
```
  05d3fc81
- W
  
  fix gpu mem alloc: use phi::memory_utils::Alloc (#53721) · 6cd7609a
  由 Wangzheee 提交于 5月 12, 2023
  
  6cd7609a
11 5月, 2023 6 次提交
- X
  [Inference Zero-Dim] Support trt 0dim of gelu, hard_swish, hard_sigmoid and leaky_relu (#53714) · b150b168
  由 xiaoxiaohehe001 提交于 5月 11, 2023
```
* support_act
* delete_silu
```
  b150b168
- Y
  [inference Zero-Dim]prelu trt converter support zero dim tensor (#53634) · 82c73884
  由 Yuanle Liu 提交于 5月 11, 2023
```
* prelu op trt converter support zero dim
```
  82c73884
- Z
  
  [inference][trt]add trt sparse weights switch (#53562) · 4a69a536
  由 Zhang Jun 提交于 5月 11, 2023
  
  4a69a536
- Z
  
  [inference Zero-Dim]add equal, elementwise_op trt 0d (#53704) · 04e5e7b7
  由 Zhang Jun 提交于 5月 11, 2023
  
  04e5e7b7
- X
  [Paddle-Inference] Support trt 0dims of expand_as_v2 and mish. (#53627) · aebff6d7
  由 xiaoxiaohehe001 提交于 5月 11, 2023
```
* support_expand_mish
```
  aebff6d7
- 张
  
  昇腾和寒武纪相关代码退场 npu相关代码退场2 (#53568) · 0d45ac73
  由张春乔提交于 5月 11, 2023
  
  0d45ac73
10 5月, 2023 1 次提交
- Y
  
  scale, square, sum, swish trt op converter support zero dim (#53660) · 6a279dfd
  由 Yuanle Liu 提交于 5月 10, 2023
  
  6a279dfd
09 5月, 2023 2 次提交
- W
  
  Support trt cuda graph. (#53406) · ea0abf93
  由 Wilber 提交于 5月 09, 2023
  
  ea0abf93
- Z
  [Paddle-TRT] Del 2 useless pass (#53414) · aec4e38f
  由 zhoutianzi666 提交于 5月 09, 2023
```
* delete delete_fill_constant_op_pass and unsqueeze2_eltwise_fuse_pass
```
  aec4e38f
08 5月, 2023 4 次提交
- W
  
  A copy of https://github.com/PaddlePaddle/Paddle/pull/51438 (#53587) · 70180df5
  由 Wilber 提交于 5月 08, 2023
  
  70180df5
- 张
  
  rm npu (#53566) · 6d396ace
  由张春乔提交于 5月 08, 2023
  
  6d396ace
- Z
  [inference][trt]Unary operation support 0d (#53506) · 10f9249b
  由 Zhang Jun 提交于 5月 08, 2023
```
* fix trt Unary operation do not support 0d when TRT < 8.6

* update unary ut

* add rsqrt to unary_list

* move rsqrt to act_list
```
  10f9249b
- Y
  [Paddle-TRT] add generic plugin for lookup_table_v2(embedding) op (#53539) · fca8595e
  由 Yuanle Liu 提交于 5月 08, 2023
```
* add embedding generic plugin， not enabled
```
  fca8595e
06 5月, 2023 5 次提交
- Y
  [inference][trt] add lookup_table op trt converter, use trt gather layer (#53554) · 08b44e67
  由 Yuanle Liu 提交于 5月 06, 2023
```
* add lookup_table op trt converter

* update
```
  08b44e67
- C
  
  XPU Support external stream (#53334) · 99399f32
  由 csy0225 提交于 5月 06, 2023
  
  99399f32
- fix brpc double link (#53512) · 03fe3ce5
  由 zhenhailiu 提交于 5月 06, 2023
```
* polish

* polish

* polish

* polish

* polish

* polish

* polish

* polish

* polish

* polish

* polish
```
  03fe3ce5
- Z
  
  [inference][trt] add reduce_all and reduce_any (#53088) · 12406cad
  由 Zhang Jun 提交于 5月 06, 2023
  
  12406cad
- W
  Add trt pow converter. (#53462) · 5a44bf7e
  由 Wilber 提交于 5月 06, 2023
```
* Add trt pow converter.

* update to use AddConstantLayer

* add dims=0 ut
```
  5a44bf7e
05 5月, 2023 1 次提交
- S
  
  [XPU] Fusion of gather and assign operators to fused_mt op for reducing memory usage (#53262) · 2039115c
  由 shentanyue 提交于 5月 05, 2023
  
  2039115c
04 5月, 2023 2 次提交
- W
  
  Fix a bug in constant folding pass (#53456) · ace61b8b
  由 weishengying 提交于 5月 04, 2023
  
  ace61b8b
- Y
  
  tensor should be defined (#53449) · 72e235d0
  由 Yuanle Liu 提交于 5月 04, 2023
  
  72e235d0
28 4月, 2023 1 次提交
- Z
  [inference][trt]trt support 0 dims (#53383) · 64adfe7a
  由 Zhang Jun 提交于 4月 28, 2023
```
* trt support 0 dim

* trt support 0 dim

* update activation ut
```
  64adfe7a
27 4月, 2023 3 次提交
- Y
  
  scale trt converter support int64 (#53388) · 182b6f83
  由 Yuanle Liu 提交于 4月 27, 2023
  
  182b6f83
- Z
  
  xpu quant weight only (#53306) · 1c97aa69
  由 zhupengyang 提交于 4月 27, 2023
  
  1c97aa69
- W
  
  revert pr https://github.com/PaddlePaddle/Paddle/pull/46779 (#53373) · 2c12abd7
  由 Wilber 提交于 4月 27, 2023
  
  2c12abd7
25 4月, 2023 2 次提交
- Y
  
  [Paddle Inference] add generic plugin for p_norm (#53278) · 00f747f2
  由 Yuanle Liu 提交于 4月 25, 2023
  
  00f747f2
- Y
  [PHI]Add flags macro for PHI (#52991) · 22e96bde
  由 YuanRisheng 提交于 4月 25, 2023
```
* add flags for phi

* fix compile bugs

* fix ci bugs

* fix inference bugs

* fix cinn' bugs

* fix cinn bugs

* perfect code according comment

* fix ci bugs

* fix ci bugs
```
  22e96bde

PaddlePaddle / Paddle 大约 1 年 前同步成功

PaddlePaddle / Paddle
大约 1 年前同步成功