提交 · 4a7aa7c3508c4125503015a9baac8bbc6d1d9404 · Crayon鑫 / Paddle

08 8月, 2022 9 次提交

T

move lamb_op to phi (#44899) · 4a7aa7c3
由 Thomas Young 提交于 8月 08, 2022

4a7aa7c3

【Fix en docs】 Fix document formatting*test=document_fix (#43996) · 8537edaa

由 ReganYue 提交于 8月 08, 2022

* Update creation.py

修复Returns格式问题

* 【Fix en docs】 Fix document formatting

修复Returns和Return Type格式问题。

* 修复格式，解决中英文文档不一致问题

为了解决中英文不一致的问题，统一将Returns内容第一个标点符号改成逗号。删掉Return Type

* fix大量 与 Return Type 和 Return相关的文档

* Convert Variable and LoDTensor to Tensor

* Update python/paddle/nn/functional/common.py
Co-authored-by: NNyakku Shigure <sigure.qaq@gmail.com>

* Update detection.py

* Update io.py

* Update io.py

* Update loss.py

* Update loss.py

* Update io.py

* Update io.py

* Update nn.py

* Update nn.py

* Update nn.py

* Update nn.py

* Update nn.py

* Update nn.py

* Update nn.py

* Update nn.py

* Update nn.py

* Update nn.py

* Update sequence_lod.py

* Update sequence_lod.py

* Update sequence_lod.py

* Update sequence_lod.py

* Update sequence_lod.py

* Update sequence_lod.py

* Update sequence_lod.py

* Update sequence_lod.py

* Update sequence_lod.py

* Update sequence_lod.py

* Update sequence_lod.py

* Update metrics.py

* Update metrics.py

* Update metrics.py

* Update metrics.py

* Update metrics.py

* Update nets.py

* Update reader.py

* Update nn.py

* Update io.py

* Update nn.py

* Update io.py

* remove Return Type to Returns

* Update sequence_lod.py

* Update sequence_lod.py

* Update common.py

* Update common.py

* Update extension.py

* Update creation.py

* update*test=document_fix

* Update io.py

* Update manipulation.py

* Update sequence_lod.py

* Update common.py

* Update common.py

* Update common.py

* Update common.py

* update*test=document_fix

* update*test=document_fix

* Update loss.py

* Update loss.py

* update*test=document_fix

* paddle.crop

* for ci;test=document_fix

* update io;test=document_fix

* Update io.py

* update sequence_expand docs;test=document_fix

* update note typos;test=document_fix
Co-authored-by: NNyakku Shigure <sigure.qaq@gmail.com>
Co-authored-by: NLigoml <39876205+Ligoml@users.noreply.github.com>

8537edaa

F

[MLU] fix bn_grad and hard_sigmoid_grad error (#44919) · 8573ca54
由 fwenguang 提交于 8月 08, 2022

8573ca54
C

fix custom op attr names size error (#44938) · 713c4d0d
由 Chen Weihang 提交于 8月 07, 2022

713c4d0d
W
[Eager] fix to_uva_tensor without specific gpu number (#44937) · 75d306ff
由 Weilong Wu 提交于 8月 08, 2022
```
* [Eager] fix to_uva_tensor without specific gpu number

* Update test_tensor_uva.py

update test case
```
75d306ff
L
clean includes of tensor.h (#44928) · ee9ea48d
由 Leo Chen 提交于 8月 08, 2022
```
* clean tensor.h

* fix gather_nd
```
ee9ea48d
W

[Eager] fix split op in final state (#44952) · 3f234db0
由 Weilong Wu 提交于 8月 08, 2022

3f234db0
H

[XPU] fleet collective support bkcl. (#44929) · f9165878
由 houj04 提交于 8月 08, 2022

f9165878

[Auto Parallel] Add C++ DeviceMesh (#44951) · d99cb2e1

由 Yulong Ao 提交于 8月 08, 2022

* [Auto Parallel] Add ProcessMesh, DeviceMesh and DistributedMapper

* [Auto Parallel] Remove unecessary codes

* [Auto Parallel] Comment out unnecessary cmake statements

d99cb2e1

07 8月, 2022 2 次提交
- C
  Fix kernels header pos error (#44902) · 839d8bb3
  由 Chen Weihang 提交于 8月 07, 2022
```
* fix header pos error

* fix namespace error
```
  839d8bb3
- W
  [Eager] just need to run needed nodes (#44910) · 849eccb7
  由 Weilong Wu 提交于 8月 07, 2022
```
* [Eager] just need to run needed nodes

* [Eager] just need to run needed nodes
```
  849eccb7
05 8月, 2022 16 次提交
- F
  fix 5 operator makers with typos which pass string literal to argument... · ce9d2a9e
  由 Feiyu Chan 提交于 8月 05, 2022
```
fix 5 operator makers with typos which pass string literal to argument 'generated', remove generated as parameter of AddAttr (#44935)
```
  ce9d2a9e
- Y
  [MKLDNN]Move mkldnn activation kernel to phi (#44365) · 2dfa88d2
  由 YuanRisheng 提交于 8月 05, 2022
```
* move mkldnn activation kernel

* fix compile bugs

* fix compile bugs

* deal with conflict

* fix compile bugs

* fix windows compile bugs

* mkldnn unittest fix

* change mutable to alloc

* fix unittest bugs

* modify code according comment
```
  2dfa88d2
- Z
  
  add sparse visit (#44847) · f419e341
  由 zhangkaihuo 提交于 8月 05, 2022
  
  f419e341
- Z
  
  Add feed&fetch as default deny ops. (#44708) · d4ca7ffb
  由 Zhen Wang 提交于 8月 05, 2022
  
  d4ca7ffb
- S
  Merge matmul_v1 and matmul_v2 fuse passes (#44870) · d0cf9d9d
  由 Sławomir Siwek 提交于 8月 05, 2022
```
* remove v2_transpose_reshape

* matmul_transpose_reshape

* reshape_transpose_matmul

* restore ut

* adjust old ut

* restore parallel UT ruels

* feedback from review
```
  d0cf9d9d
- O
  
  modify pybind.cc (maybe a typo) (#44883) · 1f7e9546
  由 OccupyMars2025 提交于 8月 05, 2022
  
  1f7e9546
- J
  
  Add int8 support for matmulV2 (#44908) · f3c14762
  由 joanna.wozna.intel 提交于 8月 05, 2022
  
  f3c14762
- Q
  
  [DCU] fix hipDeviceAttributeManagedMemory not support on DTK, test=develop (#44816) · 075d7219
  由 Qi Li 提交于 8月 05, 2022
  
  075d7219
- Y
  
  fix bugs when clip xshape (#44898) · df790b9b
  由 YuanRisheng 提交于 8月 05, 2022
  
  df790b9b
- D
  migrate kernel (#44841) · 62a98130
  由 duanboqiang 提交于 8月 05, 2022
```
* migrate kernel

* fix sig order

* remove header files

* remove header

* remove header

* modify logits grad
```
  62a98130
- C
  enhance fused_multi_transformer_op(post_layer_norm) (#44789) · 643c94e4
  由 carryyu 提交于 8月 05, 2022
```
* add fused_multi_transformer post_layer_norm

* add test post_layer_norm
```
  643c94e4
- Z
  update trt workspace size param (#44469) · bdce552b
  由 Zhang Jun 提交于 8月 05, 2022
```
* update trt workspace size param

* update

* update

* update

* use int64_t

* use int64_t

* upate

* update
```
  bdce552b
- Z
  
  refactor xpu tests for squeeze/unsqueeze, *test=kunlun (#44812) · 54d98963
  由 zhangxiaoci 提交于 8月 05, 2022
  
  54d98963
- F
  move fft kernels to phi (#44714) · 153f1138
  由 Feiyu Chan 提交于 8月 05, 2022
```
* move fft kernels to phi, done with cufft, pocketfft, mkl_cdft, hipfft
* make stft_op use fft from phi/kernels/funcs, clean code
```
  153f1138
- R
  
  Skip distributed Program for standalone executor (#44897) · 0cbc870e
  由 Ruibiao Chen 提交于 8月 05, 2022
  
  0cbc870e
- A
  [IPU] restore dockerfile to gcc8.2 (#44909) · 206102af
  由 Allen Guo 提交于 8月 05, 2022
```
* restore to gcc8.2

* test=document_fix
```
  206102af
04 8月, 2022 13 次提交

Matmuls with activation and elementwise_add fuses (#44655) · 0420d514

由 Sławomir Siwek 提交于 8月 04, 2022

* Add unit tests

* matmul_v2 + activation

* matmuls + elementwise_add

* matmul_v2 postops

* transform matmul to v2

* opcompat

* fix fusing matmul with multipe outs

* add shape constraints

* remove unused vars

* change pass order

* - Unit tests to be debugged

- fix

- refactor

- diagnostic

- more diagnostic

- fix

- Fix number two

- fix

- fix

- fix

- alpha added

- more fixes

- compilation fix

- removed diagnostic code

- cosmetic fixes

* lint

* add alpha constraint

* merge matmul refactor

* trigger CI

* - fix

* - another fix

* code style

* add support for matmul+elementwise_add+activation

* code style

* fix bfloat16 bugs

* change append_binary to append_sum
Co-authored-by: NJacek Czaja <jacek.czaja@intel.com>

0420d514

W

Set ALL_BACKEND for shape cpu kernel (#44884) · 5e74ba33
由 WangZhen 提交于 8月 04, 2022

5e74ba33

[Docs][en] adjust code example format (#44679) · d5de7886

由 Nyakku Shigure 提交于 8月 04, 2022

* add name attribute to code-block, test=document_fix

* remove redundant labels, test=document_fix

* remove redundant labels (from upstream), test=document_fix

* more COPY-FROM (try multiple code example), test=document_fix

* empty commit, try to trigger PR-CI-build

* fix some `Examples:` format issues

* fix some ci errors

d5de7886

Z
[Paddle-TRT] add Rnn (#44678) · ffc8defa
由 zhoutianzi666 提交于 8月 04, 2022
```
* add rnn
```
ffc8defa
J

added conv and conv_tranpose support for md (#44677) · b2727020
由 jakpiase 提交于 8月 04, 2022

b2727020
L
Addition of fp16 type support for Compare OP (#44405) · 6506668e
由 limingshu 提交于 8月 04, 2022
```
* first commit

* add fp16 ctest files for compare op

* add cpu register of float16 for compare ops
```
6506668e
C

fix bug (#44875) · c693a027
由 ccrrong 提交于 8月 04, 2022

c693a027
D
[XPU] add merged_momentum including fp32 and fp16 (#44824) · 4922376c
由 dongfangshenzhu 提交于 8月 04, 2022
```
* add merged_momentum *test=kunlun

* add merged_momentum *test=kunlun

* add fp16 to merged_momentum,*test=kunlun
```
4922376c
W
[Eager] fix slice's input mistake (#44855) · cfc9bf76
由 Weilong Wu 提交于 8月 04, 2022
```
* [Eager] fix slice's input mistake

* add tests for slice
```
cfc9bf76

phi_fill_diagonal_tensor (#44649) · 2140e825

由 zhiboniu 提交于 8月 04, 2022

* phi_fill_diagonal_tensor

* delete extra lines

* update

* add legacy api test

* rename sig

2140e825

Phi generate_proposals_v2 (#44436) · 566c80ff

由 zhiboniu 提交于 8月 04, 2022

* phi_generate_proposals_v2

* remove old kernels

* optest add eager_check

* del lod

* update

* update

* update test_detection with_lod

* update nms_util

* remove old nms_util.h

566c80ff

mv fold & unpool to phi (#44836) · e9994f2e

由 xiaoting 提交于 8月 04, 2022

* fix conflicts

* mv unused file

* revert backward.h

* revert lu_unpack kernel

* rm .cu file

* Update lu_unpack_kernel.cc

* format phi yaml

e9994f2e

W
convert support multi block. (#44866) · b4a4eef2
由 Wilber 提交于 8月 04, 2022
```
* convert support multi block.

* update
```
b4a4eef2

Crayon鑫 / Paddle 与 Fork 源项目一致

Crayon鑫 / Paddle
与 Fork 源项目一致