提交 · 8d8527fb616445f4a29317f25f9fcbd9eb369b37 · 机器未来 / Paddle

18 2月, 2020 3 次提交

Y
register fp16 kernel for some ops (#22650) · 8d8527fb
由 Yibing Liu 提交于 2月 18, 2020
```
test=release/1.7
```
8d8527fb
C

Add Queue.get delay for multiprocess data loader (#22604) (#22640) · 5d96b6e0
由 Chen Weihang 提交于 2月 18, 2020

5d96b6e0

multi-loss optimization by adding a DownpourOpt worker (#22025) (#22638) · 750c6f42

由 yaoxuefeng 提交于 2月 18, 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

750c6f42

17 2月, 2020 11 次提交
- Z
  [cherry-pick] :Refine the error log about runtime batch and max_batch… (#22537) · c35413bf
  由 Zhaolong Xing 提交于 2月 17, 2020
```
* [cherry-pick] :Refine the error log about runtime batch and max_batch_size. #22535
test=release/1.7

* fix comments
test=release/1.7
```
  c35413bf
- 1
  
  test=develop, add distributed tools (#22623) (#22637) · 189ac02b
  由 123malin 提交于 2月 17, 2020
  
  189ac02b
- Z
  
  fix py_func bug when out is list and add unittest case(#22596) · 77428e8f
  由 zhouwei25 提交于 2月 17, 2020
  
  77428e8f
- Z
  
  fix bug fot test_dygraph_mnist_fp16.py, test=develop (#22222) (#22536) · 370fdaa1
  由 zhongpu 提交于 2月 17, 2020
  
  370fdaa1
- G
  Modify english document and unittest of while_loop (#22615) (#22629) · da338d76
  由 guofei 提交于 2月 17, 2020
```
cherry-pick #22615
```
  da338d76
- K
  [cherry-pick] optimizer var exlcude fluid.data var. test=develop (#22618) · fd37536e
  由 Kaipeng Deng 提交于 2月 17, 2020
```
* optimizer var exlcude fluid.data var. test=develop

* fix compute priority. test=develop
```
  fd37536e
- W
  cherry-pick 22551. test=develop test=release/1.7 (#22609) · 6945a80b
  由 Wilber 提交于 2月 17, 2020
```
[cherry-pick] #22551

当一个模型中有多个fc_lstm子图的时候，且其中fc共用了同一个persistable的bias，此时不应该将bias节点删除，只将非persistable的节点去除即可。
```
  6945a80b
- F
  Golang inference API (#22503) (#22601) · a06883cd
  由 flame 提交于 2月 17, 2020
```
* support  golang inference
```
  a06883cd
- S
  [cherry-pick] fix pickle between python 2 & 3 (#22620) · edcf04ca
  由 songyouwei 提交于 2月 17, 2020
```
* cherry-pick #22555
test=release/1.7, test=develop

* cherry-pick #22621
test=release/1.7, test=develop
```
  edcf04ca
- T
  add texttable for pretty flag output (#22584) (#22626) · c000f8a2
  由 tangwei12 提交于 2月 17, 2020
```
pretty print for communicator flag
```
  c000f8a2
- T
  deprecated for distribute transpiler api (#22513) (#22617) · f517fb60
  由 tangwei12 提交于 2月 17, 2020
```
* add deprecated for distribute transpiler, will delete it after 2.0.0, test=develop
```
  f517fb60
16 2月, 2020 2 次提交

H
Enhance load program state (#22546) (#22589) · 2057f76a
由 hong 提交于 2月 16, 2020
```
* enhance load program state; test=develop

* optimize commet; test=develop
```
2057f76a

Try to fix data loader test failed problem (#22622) · 3f4687b8

由 Chen Weihang 提交于 2月 16, 2020

* split unittests in data loader test, test=release/1.7

* split unittests to different files, test=release/1.7

* remove repeat unittest, test=release/1.7

3f4687b8

13 2月, 2020 8 次提交

C
Post_training_quantization support set quant 8/16 bits (#22492) (#22577) · 926227c8
由 cc 提交于 2月 13, 2020
```
Post_training_quantization support set quant 8/16 bits 
```
926227c8

石

load inference model from memory buffer, test=release/1.7 (#22562) · baec7a35

由石晓伟提交于 2月 13, 2020

* 1. load model from memory

2. scale is no longer added when saving inference model

test=develop

* raise ci coverage, test=develop

* supports saving weights to memory. test=develop

* raise ci coverage, test=develop

* fix PADDLE_ENFORCE messages, test=develop

baec7a35

Add support for dynamic_decode(while) training. (#22231) (#22574) · 4e3c535a

由 Guo Sheng 提交于 2月 13, 2020

* Add support for dynamic_decode(while) training. test=develop

* Fix assign_op and tensor_array_read_write_op after solving conflict. test=develop

* Fix test_rnn_decode_api.py. test=develop

* Refine docs for apis in rnn.py. test=develop

* Adjust outputs of dynamic_decode. test=develop

* Remove the force_cpu update in assign_op. test=develop

* Remove the force_cpu update in assign_op. test=develop

* Make RNNCell.get_initial_states support batch_dim_idx argument. test=develop

* Rename _create_array_outof_while as _create_array_out_of_while in rnn.py.

test=release/1.7

4e3c535a

T
fix bug with compiledProgram (#22495) (#22566) · a8f85f2c
由 tangwei12 提交于 2月 13, 2020
```
* add thread barrier for the compiled program
```
a8f85f2c

[Cherry-Pick][1.7] Enable MKL-DNN Quantization of Ernie in Slim (#22575) · e78858f1

由 Michał Gallus 提交于 2月 13, 2020

* Introduce Ernie NLP

* Fix error regarding incorrect attr type

test=release/1.7
Co-authored-by: NWojciech Uss <wojciech.uss@intel.com>

e78858f1

石

update internal header files, test=develop (#22422) · f02ac218
由石晓伟提交于 2月 13, 2020

f02ac218
石
support dynamic cuda libs, test=release/1.7 (#22419) · 7a06fc7a
由石晓伟提交于 2月 13, 2020
```
* update external lite, test=develop

* switch WITH_TESTING to OFF, test=develop
```
7a06fc7a
Z

fix traced layer with non persistable vars, test=release/1.7 (#22554) · 4e53be82
由 Zeng Jinle 提交于 2月 12, 2020

4e53be82

12 2月, 2020 2 次提交
- P
  
  remove copying trt to inference lib, test=develop (#22470) (#22558) · 167de0fa
  由 Pei Yang 提交于 2月 12, 2020
  
  167de0fa
- G
  Make assign op support LoDTensorArray and modify while_loop API (#22309) (#22525) · d8a2aa5d
  由 guofei 提交于 2月 12, 2020
```
This PR makes assign op support LoDTensorArray and enable the loop_vars in
while_loop to support tuple or list.
```
  d8a2aa5d
11 2月, 2020 4 次提交

[cherry-pick] Add weight quantization in post_training_quanzitaion (#22445) (#22493) · ad2813b1

由 cc 提交于 2月 11, 2020

* Add weight quantization in post_training_quanzitaion (#22445)

* [cherry-pick]Support int16 for Tensor (#22423)

* add int16 support, test=develop, test=release/1.7
Co-authored-by: NLeo Chen <chenqiuliang@baidu.com>

ad2813b1

Z
add cp27-cp27m-gcc82 and cp27-cp27mu-gcc82 branch to support gcc8.2 compile... · ff4f0757
由 zhongpu 提交于 2月 11, 2020
```
add cp27-cp27m-gcc82 and cp27-cp27mu-gcc82 branch to support gcc8.2 compile for paddle, test=develop (#22504) (#22524)
```
ff4f0757

cherry-pick 22509. test=develop test=release/1.7 (#22527) · 49a80b45

由 Wilber 提交于 2月 11, 2020

[cherry-pick] #22509

支持不依赖nccl进行编译。

多卡下，如果没有打开WITH_NCCL开关编译，多卡不能通信，则只能选择一张卡使用

49a80b45

update. test=develop test=release/1.7 (#22518) · 59bb29db

由 Wilber 提交于 2月 11, 2020

[cherry-pick] #22484

支持不依赖nccl进行编译。

多卡下，如果没有打开WITH_NCCL开关编译，则只能使用单卡

59bb29db

10 2月, 2020 2 次提交
- Z
  [cherry-pick] [Refine Paddle-TRT INT8]: Support PaddleSlim's Resnet50 (#22485) · 14737e19
  由 Zhaolong Xing 提交于 2月 10, 2020
```
test=develop
```
  14737e19
- A
  
  [cherry-pick]polish no_grad_set of gradient and append_backward (#22440) (#22498) · d2d4a02c
  由 Aurelius84 提交于 2月 10, 2020
  
  d2d4a02c
07 2月, 2020 4 次提交
- L
  optimize performance of interpolate op (#22436) (#22489) · 0d0ea9b7
  由 LielinJiang 提交于 2月 07, 2020
```
* optimize interpolate op.
test=develop
```
  0d0ea9b7
- Z
  [Cherry-pick] [Fix BUG]: Core when multi thread + clone + paddle-tr #22442 (#22471) · 2eb3a7a9
  由 Zhaolong Xing 提交于 2月 07, 2020
```
test=release/1.7
```
  2eb3a7a9
- J
  [Cherry-pick] Add dequant-scale squash (#22409) (#22473) · 6892deb1
  由 joanna.wozna.intel 提交于 2月 07, 2020
```
* Add dequant scale squash

* Correct dequant-scale squash test

test=release/1.7
```
  6892deb1
- F
  R language support (#22417) (#22468) · 256dad09
  由 flame 提交于 2月 07, 2020
```
* R-language inference support
```
  256dad09
06 2月, 2020 2 次提交

X
change pslib url (#22454) · 20fa4f0d
由 xujiaqi01 提交于 2月 06, 2020
```
* change pslib url of release1.7 to a stable commit
* test=develop
```
20fa4f0d

[cherry-pick]revised the readme/readme_cn (#22462) · 780d656a

由 Daniel Yang 提交于 2月 06, 2020

* revised readme/readme_cn for paddle (#22219)

* these are the revised readme/readme_cn pull-request, test=develop

* this is a readme pull request, test = develop

* these are the revised readme/readme_cn pull request,test = develop

* revised version 1.6.2->1.6.3,test=release/1.7

780d656a

05 2月, 2020 2 次提交
- T
  fix sigmoid cudnn bug (#22439) (#22449) · 0be820cb
  由 Tao Luo 提交于 2月 05, 2020
```
Co-authored-by: NManjunath Bhat <manjunathbhat9920@gmail.com>
```
  0be820cb
- W
  cherry-pick 22384 and 22371. test=develop test=release/1.7 (#22453) · fb98116c
  由 Wilber 提交于 2月 05, 2020
```
[cherry-pick] #22384 and #22371

22384增加了WITH_NCCL开关

22371修改了fluid依赖lite的commit id
```
  fb98116c

机器未来 / Paddle 与 Fork 源项目一致

机器未来 / Paddle
与 Fork 源项目一致