提交 · 01ab8a061936aee264c8e9aeae360cf3de195e47 · BaiXuePrincess / Paddle

17 3月, 2020 4 次提交
- F
  add approximation for gelu, test=develop (#22961) · 01ab8a06
  由 Feiyu Chan 提交于 3月 17, 2020
```
add approximation for gelu, default value is False (only kernel with eigen is added, remove code for computing gelu with MKLDNN temporarily)
```
  01ab8a06
- A
  
  Revert "Change ShareDataWith() to TensorCopy() in conv_mkldnn (#22695)" (#22985) · 5842ae67
  由 Adam 提交于 3月 17, 2020
  
  5842ae67
- G
  
  Fix lod error of concat op for axis = 0 (#22538) · 1dc1f927
  由 GaoWei8 提交于 3月 17, 2020
  
  1dc1f927
- Z
  
  Override GetKernelTypeForVar to avoid device transform, test=develop (#23032) · 714b0076
  由 Zhang Ting 提交于 3月 17, 2020
  
  714b0076
16 3月, 2020 1 次提交
- W
  
  fix the conv group problem test=develop (#23025) · 112e3edb
  由 wangchaochaohu 提交于 3月 16, 2020
  
  112e3edb
13 3月, 2020 2 次提交
- W
  Add Unittest for backward of fusion group (#22932) · 3757e068
  由 wangchaochaohu 提交于 3月 13, 2020
```
* add fusion group test for backward and refine code
```
  3757e068
- C
  fix bug which input shape (#22965) · 63f3ada7
  由 chengjuntao 提交于 3月 13, 2020
```
* fix bug which input shape, test=develop

* add error type,test=develop
```
  63f3ada7
12 3月, 2020 2 次提交

W
Cast fusion for fusion group (#22876) · f0d193a2
由 wangchaochaohu 提交于 3月 12, 2020
```
* add support for expression type convert and add cast Op support in fusion group
```
f0d193a2

由 yaoxuefeng 提交于 3月 12, 2020

* update

* update test=develop

* update compile set test=develop

* update compile set test=develop

* update test=develop

* update test=develop

* update test=develop

* update compile setting test=develop

* update compile setting test=develop

* update run demo test=develop

* update test=develop

* update test=develop

* fix test=develop

* update test=develop

* update test=develop

* update test=develop

* update test=develop

* update test=develop

* update test=develop

* update test=develop

* update test=develop

* update test=develop

* update format test=develop

* update format test=develop

* update style test=develop

* update style test=develop

* change style test=develop

* change style test=develop

* change style test=develop

* add dataset unittest test=develop

* update test=develop

* update for record test=develop

* udpate style for record test=develop

* update for record test=develop

* update for record test=develop

* update for record test=develop

* fix format test=develop

* update test=develop

* update test=develop

* update test=develop

* update test=develop

* update test=develop

* update test=develop

* update test=develop

* update test=develop

* update test=develop

* update test=develop

* update test=develop

* update test=develop

* update test=develop

* update test=develop

* update test=develop

* fix compile warning test=develop

* add attr default test=develop

* add unittest test=develop

* fix style test=develop

* fix style test=develop

* change out_val_ifempty to out_val_if_empty test=develop

29a7a52d

11 3月, 2020 3 次提交
- W
  Speed up the matmul op, use the gemm replace the batch gemm (#22926) · f154d586
  由 wawltor 提交于 3月 11, 2020
```
In the op of gemm, we use the gemm to replace batch gemm, speed up the matmul op 
```
  f154d586
- A
  
  Change ShareDataWith() to TensorCopy() in conv_mkldnn (#22695) · 056edf39
  由 Adam 提交于 3月 11, 2020
  
  056edf39
- Z
  [Ernie GPU Optimize]: Embedding_eltwise_layernorm Fuse (#22494) · 8d6dc102
  由 Zhaolong Xing 提交于 3月 11, 2020
```
* 1. add embedding eltwise layernorm fuse
2. add embedding eltwise layernorm op
3. refine inplace_add_relu
4. refine fc_eltwise_layernorm
test=develop

* 1. refine fc
test=develop

* fix comments
test=develop

* fix comments

test=develop
```
  8d6dc102
10 3月, 2020 1 次提交
- G
  modify assign op and add unittest of assign op (#22769) · 3d8571e8
  由 guofei 提交于 3月 10, 2020
```
As the title.
```
  3d8571e8
09 3月, 2020 4 次提交

Imperative tracer refactoring (#22457) · d33c4343

由 Zeng Jinle 提交于 3月 09, 2020

* refine grad maker, test=develop

* refactor tracer stage 1, test=develop

* merge develop to solve conflict third times, test=develop

d33c4343

fix communicator when break under pyreder mode (#22911) · ad9c8f6d

由 tangwei12 提交于 3月 09, 2020

* fix communicator when breaking under PyReader mode, test=develop
* revert some vlog level to 0, test=develop

ad9c8f6d

M
add lookup_table_dequant_op (#22900) · 5ba9dfc1
由 mapingshuo 提交于 3月 09, 2020
```
add lookup_table_dequant_op
```
5ba9dfc1

[Paddle-TRT] : (Part1) Dynamic shape support (#22868) · dd67d44a

由 Zhaolong Xing 提交于 3月 09, 2020

* change the ci trt from version 5. to 6.0

* paddle-trt dynamic shape support init

* conv+bias or conv+bn dynamic shape support
test=develop

* modity trt engine opconvert
test=develop

* fix ci error
test=develop

dd67d44a

08 3月, 2020 1 次提交
- T
  
  remove vlog, test=develop (#22898) · 07e13b84
  由 tangwei12 提交于 3月 08, 2020
  
  07e13b84
05 3月, 2020 2 次提交
- W
  fix concat_mkldnn op. test=develop (#22692) · f686310d
  由 Wilber 提交于 3月 05, 2020
```
fix concat_mkldnn op when encounter extreame conditions.
```
  f686310d
- Z
  [BUG]: Multihead matmul op's ouput size should be BxSx(N*H) (#22848) · 1a533ed2
  由 Zhaolong Xing 提交于 3月 05, 2020
```
test=develop
```
  1a533ed2
03 3月, 2020 1 次提交
- 石
  change the function in op_teller, test=develop (#22794) · ddb9b46f
  由石晓伟提交于 3月 03, 2020
```
* change the function in op_teller, test=develop

* correct the commit-id, test=develop
```
  ddb9b46f
28 2月, 2020 2 次提交
- T
  
  fix typo word (#22784) · 433cef03
  由 tianshuo78520a 提交于 2月 28, 2020
  
  433cef03
- K
  
  fix detection_map. test=develop (#22705) · ebc7ffc3
  由 Kaipeng Deng 提交于 2月 28, 2020
  
  ebc7ffc3
27 2月, 2020 3 次提交

Refine adam op to improve performance, test=develop (#22346) · 72dde4ab

由 zhaoyuchen2018 提交于 2月 27, 2020

* Refine adam op, test=develop

* Fuse kernels together to reduce cpu time.

* Refine paddle enforce, test=develop

* Remove some comments, test=develop

* Refine code,test=develop

* Refine cuda kernel, test=develop

* Refine code according to comments, test=develop

72dde4ab

W

fix lod level, test=develop (#22755) · f2d1cd11
由 wangguanzhong 提交于 2月 27, 2020

f2d1cd11

Correct CPU gradients of the argsort op (#22739) · 79d71234

由 FlyingQianMM 提交于 2月 27, 2020

* Correct CPU gradients of the argsort op, form a network to test its forward and backward process, test=develop

* fix dynamic threshold error in test_argsort_op, test=develop

79d71234

26 2月, 2020 1 次提交
- G
  Change ShareDataWith() to TensorCopy() in ref_by_trainer_id (#22717) · ae8b5f11
  由 guofei 提交于 2月 26, 2020
```
As the title
```
  ae8b5f11
25 2月, 2020 3 次提交
- C
  register fp16 for assign op (#22744) · 15c26671
  由 chengjuntao 提交于 2月 25, 2020
```
* register fp16 for assign op, test=develop

* add op test for fp16, test=develop
```
  15c26671
- D
  
  fix generate_mask_labels lod level (#22743) · 1c065346
  由 dyning 提交于 2月 25, 2020
  
  1c065346
- G
  
  fix compile&runtime lod_equality of lod_reset (#22737) · ba140222
  由 GaoWei8 提交于 2月 25, 2020
  
  ba140222
24 2月, 2020 2 次提交

add partial_sum op in contrib (#22292) · 3132681e

由 ShenLiang 提交于 2月 24, 2020

* add partial_sum_op, test=develop

* modify the Paddle Error Message, test=develop

* modify the Paddle Error Message, test=develop

* modify the bug for python3, test=develop

* modify the ut for ci, test=develop

* mv to contrib, test=develop

* use check_variable_and_dtype, test=develop

* fix ci, test=develop

* fix conflict, test=dvelop

* add partial concat, test=develop

* fix the conflict, test=develop

* fix the error, test=develop

* rm SSE4, test=develop

3132681e

add partial_concat op in contrib (#22528) · e1366613

由 ShenLiang 提交于 2月 24, 2020

* add partial_concat, test=develop

* fix the grids and blocks, test=develop

* fix the Paddle_Enforce, test=develop

* fix the doc of op, test=develop

* fix the doc, test=develop

* fix the doc of the op, test=develop

* replace -1 with None, test=develop

e1366613

23 2月, 2020 1 次提交
- T
  
  fix typo words (#22653) · d2ba91aa
  由 tianshuo78520a 提交于 2月 23, 2020
  
  d2ba91aa
22 2月, 2020 2 次提交
- Y
  register fp16 kernel for some ops (#22650) (#22696) · 6e7bfe30
  由 Yibing Liu 提交于 2月 22, 2020
```
test=develop
```
  6e7bfe30
- T
  SYNC with communicaotor (#22344) · 66a31501
  由 tangwei12 提交于 2月 22, 2020
```
* add sync communicator and implement
```
  66a31501
21 2月, 2020 2 次提交
- Y
  
  Add the support of fp16 in fusion_group (#22239) · 22bbd547
  由 Yiqun Liu 提交于 2月 21, 2020
  
  22bbd547
- H
  Add PADDLE_ENFORCE to Check Sequence Length of RecurrentOp (#22673) · adfa5b83
  由 Huihuang Zheng 提交于 2月 21, 2020
```
1. Add PADDLE_ENFORCE to Check Sequence Length of RecurrentOp.
2. Also enrich PADDLE_ENFORCE error messages.
```
  adfa5b83
18 2月, 2020 1 次提交

[UT coverage] improve the mul_mkldnn_op line coverage (#22408) · d9262145

由 lidanqing 提交于 2月 18, 2020

* improve the mul_mkldnn_op line coverage
test=develop

* remove fp32 mul mkldnn kernel
test=develop

* locally refactoring
test=develop

* change according to reviews
test=develop

d9262145

17 2月, 2020 2 次提交
- Z
  [Paddle-TRT] Refine the error log about runtime batch and max_batch_size. (#22535) · a06d75a2
  由 Zhaolong Xing 提交于 2月 17, 2020
```
* fix trt log
test=develop

* fix comments
test=develop
```
  a06d75a2
- A
  
  Update MKLDNN to v1.2 (#22521) · 608447bf
  由 Adam 提交于 2月 17, 2020
  
  608447bf

BaiXuePrincess / Paddle 与 Fork 源项目一致

BaiXuePrincess / Paddle
与 Fork 源项目一致