提交 · fb80048d36173c29c4e9db9b5d18585243727771 · Crayon鑫 / Paddle

26 7月, 2022 2 次提交

B

Move bmm OP from fluid to phi (#44496) · fb80048d
由 BiynXu 提交于 7月 26, 2022

fb80048d

Improve CI unittest parallel execution strategy (#44334) · ff216f18

由 zhangbo9674 提交于 7月 26, 2022

* paralle_test_daily

* test=paralle_test_daily

* test=paralle_test_daily

* test=parallel_test_daily

* test=paralle_test_daily

* test=paralle_test_daily

* test=paralle_test_daily

* test=paralle_test_daily

* test=paralle_test_daily

* test=paralle_test_daily

* test=paralle_test_daily

* test=paralle_test_daily

* test pre_test_bak

* test cfs

* test_cfs,test=paralle_test_daily

* test_cfs,test=paralle_test_daily

* fix nightly test name,test=paralle_test_daily

* fix nightly test name,test=paralle_test_daily

* test ci parallel speed

* refine parallel rule,test=paralle_test_daily

ff216f18

25 7月, 2022 11 次提交

[Auto Parallel] Add dist op cost (#44146) · d0f4465d

由 caozhou 提交于 7月 25, 2022

* update comp cost

* add dist default op cost

* add dist fill constant batch size like op cost

* add elewise op cost

* add fill_constant_batch_size_like op cost unittest

* add unittest and remove fill_constant_batch_size_like grad op cost

* add to cmakelist

* fix unittest bug

d0f4465d

Fc fp16 (#44578) · a54c6953

由 ming1753 提交于 7月 25, 2022

* (modified) fc support fp16

* __CUDA_ARCH__ version

* delete half

* delete half

* add half support

* add half support

* add half support

a54c6953

[MLU]transpose convbpf output to HWCN for better performance (#44552) · 3244a9de
由光明和真理提交于 7月 25, 2022

3244a9de

【Hackathon No.21】为 Paddle 新增 SoftMarginLoss (#42364) · f9cd526b

由 yangguohao 提交于 7月 25, 2022

* 2022-04-28

* 2022-04-28_V2

* 2022-04-30

* 2022-04-30_V2

* 2022-05-01

* 2022-05-02

* 2022-05-02_V2

* 2022-05-05_V1

* 2022-05-06_V1

* 2022-05-07_V1

* Update loss.py

* 2022-05-07_V2

* 2022-05-13_V1

* Update test_soft_margin_loss.py

* Update loss.py

* Update loss.py

* 2022-05-16_V1

* 2022-05-19_V1

* 2022-05-20_V1

* Update test_soft_margin_loss.py

* 2022-06-01_V1

* 2022-06-05

* 2022-06-07

* 2022-06-07

* 2022-06-08

* 2022-06-08_V2

* 2022-06-17-code_style

* Modify python

* 2022-06-20

* for

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

f9cd526b

A
[dy2st]Add ProgramHelper to polish build program logic in autoparallel.Engine (#44513) · 243acdb4
由 Aurelius84 提交于 7月 25, 2022
```
* [dy2st]Add ProgramHelper to polish build program logic in autoparallel.Engine

* refine code
```
243acdb4
Q

skip dist test cases if mlu card number only one, test=develop (#44549) · ead696ae
由 Qi Li 提交于 7月 25, 2022

ead696ae
W
Phi gird sampler migration (#44562) · b470bd18
由 Wang Bojun 提交于 7月 25, 2022
```
* add_ymal_utest for phi grid_sampler op
```
b470bd18
Z
add swish using TensorRT layer (#44561) · c5a1e49c
由 Zhang Jun 提交于 7月 25, 2022
```
* update

* empty commit

* update

* update

* update
```
c5a1e49c
L

[Phi] Migrate squared_l2_norm_op to phi (#44492) · 3e170163
由 lyq 提交于 7月 25, 2022

3e170163
W
[JitLayer]Fix jit.save error when save params combined (#44504) · c0a29d2f
由 WangZhen 提交于 7月 25, 2022
```
* Fix jit.save error when save params combined

* Change dict_value to list
```
c0a29d2f
Z
Fix bug of amp code-gen (#44570) · e32e4a1d
由 zyfncg 提交于 7月 25, 2022
```
* fix bug of amp code_gen

* fix bug
```
e32e4a1d

22 7月, 2022 13 次提交
- M
  Fc fp16 (#44558) · 6b6f7a21
  由 ming1753 提交于 7月 22, 2022
```
* (modified) fc support fp16

* __CUDA_ARCH__ version

* delete half

* delete half
```
  6b6f7a21
- [Sparse]add sparse unary api(expm1/deg2rad/rad2deg/relu6/leaky_relu) (#44432) · 19d9c736
  由 zhouweiwei2014 提交于 7月 22, 2022
  
  19d9c736
- H
  
  support send_partial, recv_partial and allgather_partial in ProcessGroupNCCL (#44444) · 18c77325
  由 Haohongxiang 提交于 7月 22, 2022
  
  18c77325
- F
  [phi] move inverse op from fluid to phi (#44471) · ea4b2c5e
  由 freeliuzc 提交于 7月 22, 2022
```
* move inverse from fluid to phi with unitest bug

* fix bug, add eager op yaml
```
  ea4b2c5e
- Q
  add xpu lars_momentum/pow2_decay (#44448) · 8ccbb863
  由 QingshuChen 提交于 7月 22, 2022
```
*test=kunlun
```
  8ccbb863
- Y
  
  Add code of occupancy computing on DCU and avoid threadID bug for DCU profiler (#44520) · 8037901b
  由 yuguo 提交于 7月 22, 2022
  
  8037901b
- M
  
  (modified) fc support fp16 (#44540) · fcfaa104
  由 ming1753 提交于 7月 22, 2022
  
  fcfaa104
- A
  [CustomDevice] register Copy for custom device (#44200) · 3b0aa75e
  由 Aganlengzi 提交于 7月 22, 2022
```
* [CustomDevice] register Copy for custom device

* [CustomDevice] register Copy for custom device

* [CustomDevice] register Copy for custom device

* merge and add uts

* merge and add uts

* fix for blocking and unittests coverage
```
  3b0aa75e
- Z
  
  commit (#44534) · db864f0b
  由 zhoutianzi666 提交于 7月 22, 2022
  
  db864f0b
- F
  
  [MLU] add floor kernel and grid_sampler kernel (#44498) · 1c0120e2
  由 fwenguang 提交于 7月 22, 2022
  
  1c0120e2
- C
  
  fix arg_max to select first index (#44521) · 5ee4a21a
  由 cifar10 提交于 7月 22, 2022
  
  5ee4a21a
- X
  
  shufflechannelfix (#44516) · a2b39320
  由 xiaoxiaohehe001 提交于 7月 22, 2022
  
  a2b39320
- W
  
  add batch stream (#44524) · 4f86092b
  由 Wilber 提交于 7月 22, 2022
  
  4f86092b
21 7月, 2022 14 次提交

Fc fp16 (#44505) · 3e1280ea

由 ming1753 提交于 7月 21, 2022

* fc support fp16

* add a ‘,’ on paddle_pass_builder.cc

* fc support fp16 on non-cuda.

3e1280ea

[ Dy2static ] infer_program may be incorrect in amp mode. (#44487) · 185a900f

由 xiongkun 提交于 7月 21, 2022

* fix the outputs of net is x,x

* add unittest for duplicate output

* fix

* fix _infer_program use the original program not the amp program.

* get _***program_id back and avoid duplicate cache
ing

* fix

185a900f

[jit] jit support property.proto (#44337) · 0aa344f0

由 Hui Zhang 提交于 7月 21, 2022

* add property.proto, can compiled

* property get and deserilize

* support get float

* format code

* format code

* add unittest

* add more set method

* fix grammar error

* Update paddle/fluid/jit/property.h
Co-authored-by: NAurelius84 <zhangliujie@baidu.com>

* Update paddle/fluid/jit/property.cc
Co-authored-by: NAurelius84 <zhangliujie@baidu.com>

* Update paddle/fluid/jit/property.cc
Co-authored-by: NAurelius84 <zhangliujie@baidu.com>

* Update paddle/fluid/jit/property.cc
Co-authored-by: NAurelius84 <zhangliujie@baidu.com>

* fix comment

* fix error throw

* fix property save unit test

* fix error info

* fix copyright and header import

* reorder jit property tensor datatype
Co-authored-by: NAurelius84 <zhangliujie@baidu.com>

0aa344f0

C

[MLU] transpose avg_pool2d to NHWC for better performance. (#44475) · 7ab0e336
由 Chenxiao Niu 提交于 7月 21, 2022

7ab0e336
Z
[Paddle-TRT] fix_fill_constant (#44481) · c3ba8056
由 zhoutianzi666 提交于 7月 21, 2022
```
* fix_fill_constant

* fix_fill_constant

* fix_ernie
```
c3ba8056

[ Phi ] svd transfer (#44392) · ba89a3d3

由 xiongkun 提交于 7月 21, 2022

* svd cpu forward

* svd gpu forward

* transfer the backward of svd

* remove cusolver in svd_grad

* svd kernel bug fix

* fix bugs

* fix bugs.

* fix bug

ba89a3d3

W
[Eager] bilinear_tensor_product yaml (#44459) · 55e5ab82
由 wanghuancoder 提交于 7月 21, 2022
```
* bilinear_tensor_product yaml
```
55e5ab82
[Sparse] Add sparse addmm kernel (dense+coo*dense->dense,dense+csr*dense->dense) (#44451) · 78b5c103
由 zhouweiwei2014 提交于 7月 21, 2022

78b5c103

[JitLayer]Pybind PEFunction and call phi api in layer_test (#44465) · a0bccd9e

由 WangZhen 提交于 7月 21, 2022

* Support predictor function in JitLayer

* Pybind PEFunction

* Pybind PEFunction and call phi api in layer_test

* Call sqrt phi API

* Polish flags

* Fix comments

a0bccd9e

L
Replace with dygraph op calling method. (#44331) · 5b3f91df
由 Li Min 提交于 7月 21, 2022
```
* Replace with dygraph op calling method.
```
5b3f91df
C

fix tensor stream error in custom op (#44500) · 32b3469a
由 Chen Weihang 提交于 7月 21, 2022

32b3469a
A
[Dy2Sta]Fix Segment Fault while training multi-card if params have no grad (#44485) · 32c97a9d
由 Aurelius84 提交于 7月 21, 2022
```
* [Dy2Sta]Fix Segment Fault while training multi-card if params have no grad

* fix unittest
```
32c97a9d
Z
add slot attr for push sparse op (#44422) · 85c6937b
由 zhaocaibei123 提交于 7月 21, 2022
```
* add slot attr for push sparse op

* add pybind

* remove fleet

* add unittest

* fix
```
85c6937b
Y
add adaptive pool and softmax with cross entropy supports different axis, * test = kunlun (#44428) · 1a7f2de3
由 ykkk2333 提交于 7月 21, 2022
```
* add xpu pnorm op and fix pool op, *test=kunlun

* add adaptive pool, and softmax with cross entropy supports different axis, *test=kunlun
```
1a7f2de3

Crayon鑫 / Paddle 与 Fork 源项目一致

Crayon鑫 / Paddle
与 Fork 源项目一致