提交 · b2bb7ec90c95f152573a4460d7f9f005dc5bd83d · BaiXuePrincess / Paddle

09 2月, 2023 2 次提交
- W
  [TRT] Transpose layernorm fusion with different input format (#50082) · b2bb7ec9
  由 Wang Bojun 提交于 2月 09, 2023
```
* trans_layernorm
```
  b2bb7ec9
- 傅
  
  fix set_value_65965 (#50340) · b3f60f39
  由傅剑寒提交于 2月 09, 2023
  
  b3f60f39
08 2月, 2023 22 次提交

fuse quantize+transpose and transpose+dequantize (#49509) · 197a4ffe

由 Paulina Gacek 提交于 2月 08, 2023

* QuantTranpose pattern is being found by pass

* quant + transpose fuse

* code style changes

* UT written, reorder fixed

* Dequantize + transpose2 fuse  added

* pass name changed

* UT added & shift corrected

* got rid of redundancy

* review changes

* AsIntermediate corrected

* compat added

197a4ffe

S
Add bf16 support for fused matmul (#50254) · b47923b4
由 Sławomir Siwek 提交于 2月 08, 2023
```
* add support for bf16 fused_ops

* fused_matmul only
```
b47923b4
W
[code style]fix cpplint codestyle (#50314) · 209d534d
由 wangxiaoning 提交于 2月 08, 2023
```
* fix codestyle

* fix std
```
209d534d
Z
[inference][trt] Disable ShapeTensor for nearest_interp_v2 when trt version < 8.2 (#50258) · fa284076
由 Zhang Jun 提交于 2月 08, 2023
```
* update

* update

* format code

* update

* Update test_trt_convert_nearest_interp_v2.py
```
fa284076
Y

Fused attention pass mp support (#50320) · e44ff495
由 Yuang Liu 提交于 2月 08, 2023

e44ff495
Z
[pglbox]hidden unzip (#50292) · a7539508
由 zmxdream 提交于 2月 08, 2023
```
* hidden unzip

* fix

* fix
```
a7539508
W

Export custom operator-related function symbols (#50238) · f9c801ff
由 weishengying 提交于 2月 08, 2023

f9c801ff
H

Use inference, save construct time (#50163) · 7a82b6de
由 HongyuJia 提交于 2月 08, 2023

7a82b6de
T

test=document_fix (#50323) · cd6ebca6
由 tianshuo78520a 提交于 2月 08, 2023

cd6ebca6
T

test=document_fix (#50322) · be2a1d94
由 tianshuo78520a 提交于 2月 08, 2023

be2a1d94
Z
Fix bn performance degradation (#50287) · 6f1ec935
由 zhangkaihuo 提交于 2月 08, 2023
```
* fix bn performance degradation
```
6f1ec935

[feature] use prim flag in shell (#50309) · 8c14b02b

由 cyber-pioneer 提交于 2月 08, 2023

* add flag

* change flag

* use prim flag

* fix code

* fix softmax prim flag

* set case timeout

8c14b02b

L

Optimize gc in executor (#50301) · 9268f392
由 LiYuRio 提交于 2月 08, 2023

9268f392

[Tensor Support unsigned] Tensor::data() supports unsigned int and bfloat16 (#50257) · 80dc81c5

由 HongyuJia 提交于 2月 08, 2023

* support unsigned int and bfloat16

* update unit test

* update DenseTensor datatype

* unsupport more datatype of mutable_data(Place)

* fix unittest

80dc81c5

R

fix gcc12 error (#48176) · ffa2ec19
由 risemeup1 提交于 2月 08, 2023

ffa2ec19

Add eager jit dir (#49369) · 32804fe8

由 risemeup1 提交于 2月 08, 2023

* add_eager_and_jit,test=ljd_test

* test

* test,test=ljd_test

* test,test=ljd_test

* test,test=ljd_test

* test,test=ljd_test

* test,test=ljd_test

* test,test=ljd_test

* add_eager_jit_dir,test=ljd_test

* fix conflict,test=ljd_test

* test,test=ljd_test

* get new precise_map,test=ljd_test

* test

32804fe8

Z

[Zero-Dim] Fix 0d axis support for argmin/argmax (#50293) · aec1e4ce
由 Zhong Hui 提交于 2月 08, 2023

aec1e4ce

[Paddle-TRT] remove engine info from RumImpl process (#50181) · b3888614

由 gaoziyuan 提交于 2月 08, 2023

* remove_engine_info

* remove_engine_info

* remove_engine_info

* change trtlayerinformation line to json

---------
Co-authored-by: Ngaoziyuan <gaoziyuan@baidu.com>

b3888614

W

Enlarge timeout for test_resnet_prim_cinn (#50285) · bb148c54
由 WangZhen 提交于 2月 08, 2023

bb148c54

Support compile pslib v3 (#50166) · 8464c04d

由 pangengzheng 提交于 2月 08, 2023

* fix feature_value.h and feature_value.cu to support pslib

* code style

* align DistPsArch pre-stable branch
---------
Co-authored-by: NSławomir Siwek <slawomir.siwek@intel.com>
Co-authored-by: NTomasz Socha <tomasz.socha@intel.com>
Co-authored-by: Nheliqi <1101791222@qq.com>
Co-authored-by: Nzqw_1997 <118182234+zhengqiwen1997@users.noreply.github.com>
Co-authored-by: Njameszhang <zhangxiaoci@baidu.com>
Co-authored-by: Nxiaoguoguo626807 <100397923+xiaoguoguo626807@users.noreply.github.com>
Co-authored-by: NFeiyu Chan <chenfeiyu@baidu.com>
Co-authored-by: NGGBond8488 <33050871+GGBond8488@users.noreply.github.com>
Co-authored-by: Nsprouteer <89541335+sprouteer@users.noreply.github.com>
Co-authored-by: Njakpiase <jakpia21@gmail.com>
Co-authored-by: NJiabin Yang <360788950@qq.com>
Co-authored-by: Nlimingshu <61349199+JamesLim-sy@users.noreply.github.com>
Co-authored-by: Nzhangbopd <1299246947@qq.com>
Co-authored-by: N张春乔 <83450930+Liyulingyue@users.noreply.github.com>
Co-authored-by: NLiYuRio <63526175+LiYuRio@users.noreply.github.com>
Co-authored-by: N姜永久 <34344716+yjjiang11@users.noreply.github.com>
Co-authored-by: NYuang Liu <liuyuang@baidu.com>
Co-authored-by: Njiangcheng <thisjiang@qq.com>
Co-authored-by: Nronnywang <ronny1996@163.com>
Co-authored-by: Nsneaxiy <32832641+sneaxiy@users.noreply.github.com>
Co-authored-by: Nhouj04 <35131887+houj04@users.noreply.github.com>
Co-authored-by: Nzhangbo9674 <82555433+zhangbo9674@users.noreply.github.com>
Co-authored-by: Ngem5 <117625383+linsheng011@users.noreply.github.com>
Co-authored-by: Nwanghuancoder <wanghuan29@baidu.com>
Co-authored-by: NRyan <44900829+DrRyanHuang@users.noreply.github.com>
Co-authored-by: NRuibiao Chen <chenruibiao@baidu.com>
Co-authored-by: Nengineer1109 <jialiang.wang@xdxct.com>
Co-authored-by: NRedContritio <RedContritio@qq.com>
Co-authored-by: Nmjxs <52824616+kk-2000@users.noreply.github.com>
Co-authored-by: NYiqun Liu <Xreki@users.noreply.github.com>
Co-authored-by: N张正海 <65210872+ccsuzzh@users.noreply.github.com>
Co-authored-by: NHongyuJia <jiahongyu@baidu.com>
Co-authored-by: Npangyoki <pangyoki@126.com>
Co-authored-by: NLoneRanger <836253168@qq.com>
Co-authored-by: NTeFeng Chen <ctfeng66@163.com>
Co-authored-by: NLeo Guo <58431564+ZibinGuo@users.noreply.github.com>
Co-authored-by: Nxiaoting <31891223+tink2123@users.noreply.github.com>
Co-authored-by: N201716010711 <87008376+201716010711@users.noreply.github.com>
Co-authored-by: Nwangxiaoning <71813629+wangxn12138@users.noreply.github.com>
Co-authored-by: NYuanle Liu <yuanlehome@163.com>
Co-authored-by: ZZK <359521840@qq.com>
Co-authored-by: Nzhangkaihuo <zhangkaihuo@baidu.com>
Co-authored-by: NRoc <30228238+sljlp@users.noreply.github.com>
Co-authored-by: NPuQing <me@puqing.work>
Co-authored-by: NZhang Jun <ewalker@live.cn>
Co-authored-by: NCharles-hit <56987902+Charles-hit@users.noreply.github.com>
Co-authored-by: Nniuliling123 <51102941+niuliling123@users.noreply.github.com>
Co-authored-by: Nwenbin <wang3323032@qq.com>
Co-authored-by: Nwangshengxiang <121413869+shengxiangwang@users.noreply.github.com>
Co-authored-by: NBo Zhang <105368690+zhangbopd@users.noreply.github.com>
Co-authored-by: NAurelius84 <zhangliujie@baidu.com>
Co-authored-by: Nzxcd <228587199@qq.com>
Co-authored-by: Nzhoutianzi666 <39978853+zhoutianzi666@users.noreply.github.com>
Co-authored-by: Ngouzil <66515297+gouzil@users.noreply.github.com>
Co-authored-by: Nzhangyikun02 <48021248+zhangyk0314@users.noreply.github.com>
Co-authored-by: NHui Zhang <zhtclz@foxmail.com>
Co-authored-by: NWang Bojun <105858416+wwbitejotunn@users.noreply.github.com>
Co-authored-by: NGuanghua Yu <742925032@qq.com>
Co-authored-by: NYUNSHEN XIE <1084314248@qq.com>
Co-authored-by: NZhong Hui <zhonghui.net@gmail.com>
Co-authored-by: Nrisemeup1 <62429225+risemeup1@users.noreply.github.com>
Co-authored-by: Nliuruyan <44316842+liuruyan@users.noreply.github.com>
Co-authored-by: NLeo Chen <chenqiuliang@baidu.com>
Co-authored-by: NYuanRisheng <yuanrisheng@baidu.com>
Co-authored-by: Nwuhuachaocoding <77733235+wuhuachaocoding@users.noreply.github.com>
Co-authored-by: NCcc <52520497+juncaipeng@users.noreply.github.com>

8464c04d

H

move mixed_vector (#50282) · 35d7d1f0
由 Huang Jiyi 提交于 2月 08, 2023

35d7d1f0

[PHI]Unify Fluid and PHI kernel (#49328) · e92e3aab

由 YuanRisheng 提交于 2月 08, 2023

* unify_kernel

* fix compile bugs

* modify macro name

* perfect code according comment

* fix compile bugs

* fix compile bugs

* fix ci bugs

* fix ci bug

* fix ci bugs

* fix ci bugs

* modify code according comment

* rm conv_fusion_op

e92e3aab

07 2月, 2023 16 次提交
- X
  
  [Paddle Inference] Fix range fp32 input below trt 8.4. (#50221) · 766a4ca9
  由 xiaoxiaohehe001 提交于 2月 07, 2023
  
  766a4ca9
- Z
  Remove axis in some elementwise api (#50190) · 1dedaada
  由 zyfncg 提交于 2月 07, 2023
```
* remove axis in some elementwise api

* fix inplace bug eager-gen

* fix bug

* revert change for CheckInplace

* polish code
```
  1dedaada
- C
  Fix bad alloc exception throw error (#50266) · 05ba95e1
  由 Chen Weihang 提交于 2月 07, 2023
```
* fix bad alloc exp error

* remove needless files
```
  05ba95e1
- 张
  
  fix div 0 error in conv1/2/3 (#49999) · 7a0fdeb9
  由张春乔提交于 2月 07, 2023
  
  7a0fdeb9
- 张
  
  fix div 0 error in chunk (#49998) · 36e5de81
  由张春乔提交于 2月 07, 2023
  
  36e5de81
- Y
  update xpu zero dim tensor ut (#50289) · 7e4b432b
  由 Yuang Liu 提交于 2月 07, 2023
```
* xpu scatter ut no backward

* update gather xpu ut
```
  7e4b432b
- 张
  fix the div 0 error of sequence_concat (#49963) · 84fe2de6
  由张春乔提交于 2月 07, 2023
```
* fix the div 0 error of sequence_concat
* Update test_sequence_concat.py
```
  84fe2de6
- R
  
  Recover spectral norm self._eps (#50278) · 20075bf3
  由 Roc 提交于 2月 07, 2023
  
  20075bf3
- W
  
  [BUG FIX] fix gpups ci (#50192) · 9a47b11d
  由 wangzhen38 提交于 2月 07, 2023
  
  9a47b11d
- Y
  
  fix op_desc set attr bug (#50281) · 2755507c
  由 Yuanle Liu 提交于 2月 07, 2023
  
  2755507c
- C
  add batch_norm composite rule (#49894) · 9b3a41b1
  由 cyber-pioneer 提交于 2月 07, 2023
```
move composite test case

remove unuseful var

add composite op blacklist
```
  9b3a41b1
- C
  
  Support build with gcc12 for CUDA less than 12.0 (#50106) · 755049f2
  由 chalsliu 提交于 2月 07, 2023
  
  755049f2
- X
  [Zero-Dim] add upsample 0D test and fix api doc (#50237) · 1a0f3490
  由 xiaoting 提交于 2月 07, 2023
```
* add upsample 0d test

* update api references

* update api references

* update api reference
```
  1a0f3490
- W
  
  support bfp16 for stage3 and offload. (#49931) · 1a4a1520
  由 wuhuachaocoding 提交于 2月 07, 2023
  
  1a4a1520
- Y
  
  Fix gather, scatter op 0d tenor GPU error. (#50271) · 05c9c0a5
  由 Yuang Liu 提交于 2月 07, 2023
  
  05c9c0a5
- L
  
  Add start interceptor and fix bug in switch scope (#50225) · 0dd41a2a
  由 LiYuRio 提交于 2月 07, 2023
  
  0dd41a2a

BaiXuePrincess / Paddle 与 Fork 源项目一致

BaiXuePrincess / Paddle
与 Fork 源项目一致