提交 · 11f94cdce9dc00d6ea47b7e61871858443d15e89 · BaiXuePrincess / Paddle

23 3月, 2020 1 次提交
- A
  
  Improve SGD jit code to work with large data (#23120) · 4f5e4540
  由 Adam 提交于 3月 23, 2020
  
  4f5e4540
22 3月, 2020 1 次提交

add dequantize_log_op and make pyramid hash support int8 weight (#22548) · 4db03190

由 Liufang Sang 提交于 3月 22, 2020

* add dequantize_log_op and make pyramid hash support int8 weight test=develop

* add unittest and update pyramid hash op test=develop

* remove paddle_enforce test=develop

* fix error message test=develop

* remove incorrent commit test=develop

* fix error message in log_dequantize test=develop

* change 2019 to 2020 test=develop

* remove useless check_grad test=develop

4db03190

21 3月, 2020 1 次提交
- Z
  
  fix read op comments, test=develop, test=document_fix (#23122) · 9258e960
  由 Zeng Jinle 提交于 3月 20, 2020
  
  9258e960
20 3月, 2020 3 次提交

Reader sequential and inference partial feed (#22699) · acfc9b8a

由 Zeng Jinle 提交于 3月 20, 2020

* sequential reader stage 1, test=develop

* fix ut, test=develop

* fix iterable=False reset bug, add some logs and polish code, test=develop

* inference feed partial data, test=develop

* Turn on keep_order=True for test, test=develop

* enhance ut to test more cases, test=develop

* test commit for reverting

* Revert "test commit for reverting", test=develop

This reverts commit 80aef42e.

* add ut of merged and unmerged results, test=develop

* add more uts for coverages and add en doc of api, test=develop

* follow comments, test=develop

* change note style, test=develop

acfc9b8a

W
update embedding_eltwise_layernorm fuse and kernel. test=develop (#23114) · 95b356a0
由 Wilber 提交于 3月 20, 2020
```
update embedding_eltwise_layernorm fuse pass and fused kernel, to support multi input
```
95b356a0

Add dygraph double grad implementation (#22939) · a31d7328

由 Zeng Jinle 提交于 3月 20, 2020

* add double grad implementation for dygraph, test=develop

* polish code, add uts, test=develop

* fix place bug, test=develop

* polish codes, add more uts for coverages, test=develop

* add no_grad_set, test=develop

* add star gan ut, test=develop

* follow comments, test=develop

a31d7328

19 3月, 2020 4 次提交

high-performance dygraph slice (#22879) · 2e2da712

由 songyouwei 提交于 3月 19, 2020

* move __getitem__ to cpp

* bug fix

* add type check and gil release

* support negative step with omitted ends
test=develop

* code refine
test=develop

* bug fix
test=develop

* slice always return different pyobj
test=develop

2e2da712

S

added mkldnn swish activation (#23041) · abee05a8
由 Sylwester Fraczek 提交于 3月 19, 2020

abee05a8
Z
fix align error (#23090) · 8c6fde9e
由 Zhaolong Xing 提交于 3月 19, 2020
```
test=develop
```
8c6fde9e

Fix div zero in fake quantize op (#22966) · 915b892a

由 Liufang Sang 提交于 3月 18, 2020

* fix div zero test=develop

* fix div zero test=develop

* add hostdevice function test=develop

* add eps when is zero test=develop

915b892a

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

Fix instag (#22632) · 29a7a52d

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

BaiXuePrincess / Paddle 与 Fork 源项目一致

BaiXuePrincess / Paddle
与 Fork 源项目一致