提交 · 2e1bca99cade4129cca043289d93a3cf77384200 · PaddlePaddle / Paddle

12 10月, 2020 22 次提交
- G
  Refine the gradient calculation errors caused by renaming in while_grad (#27814) · 2e1bca99
  由 guofei 提交于 10月 12, 2020
```
test=develop
```
  2e1bca99
- W
  add load_op_xpu for Baidu Kunlun (#27817) · 8fa4c098
  由 wanghuancoder 提交于 10月 12, 2020
```
* add load_op_xpu for Baidu Kunlun, test=kunlun

* add is_compiled_with_xpu for unit test, test=kunlun

* add is_compiled_with_xpu for unit test, test=kunlun
```
  8fa4c098
- M
  refine adam/strided_slice && fix doc for rmsprop/unstack (#27740) · 84d8e49d
  由 MRXLT 提交于 10月 12, 2020
```
* refine parameters order && doc

* update rmsprop doc

* refine adam/transpose/unstack/stride_slice

* fix bug && doc

* fix doc

* bug fix

* bug fix

* fix doc

* fix doc

* fix doc

* fix doc

* depercate old strided_slice

* update doc

* set default value for name

* update doc
```
  84d8e49d
- T
  Fix/embedding doc (#27816) · e96fc6ab
  由 tangwei12 提交于 10月 12, 2020
```
* fix fluid doc

* fix fluid doc

* fix fluid doc, test=document_fix

* fix fluid doc, test=document_fix
```
  e96fc6ab
- W
  
  Lite subgraph support arm cpu. (#27827) · 9005c5a2
  由 Wilber 提交于 10月 12, 2020
  
  9005c5a2
- J
  Mutiply allows non-tensor data input (#27690) · 2bcb7c0a
  由 joejiong 提交于 10月 12, 2020
```
Mutiply allows non-tensor data input
```
  2bcb7c0a
- J
  
  [oneDNN] adaptive pool support (#27747) · 55e63763
  由 Jacek Czaja 提交于 10月 12, 2020
  
  55e63763
- add musl option (#27798) · 6335e6a0
  由 chen.zhiyu 提交于 10月 12, 2020
  
  6335e6a0
- update CUDAPlace doc. test=document_fix (#27711) · e8a5aefb
  由 myq406450149 提交于 10月 12, 2020
  
  e8a5aefb
- Z
  use IndexList to improve performance of instance_norm op (#25132) · 16999ae4
  由 Zhang Ting 提交于 10月 12, 2020
```
* use IndexList to improve performance, test=develop

* remove EIGEN_HAS_INDEX_LIST, test=develop

* use IndexList only when EIGEN_HAS_INDEX_LIST is true
```
  16999ae4
- G
  Add flattern weight of lstm (#27192) · 36bb056e
  由 GaoWei8 提交于 10月 12, 2020
```
* add flattern weight of lstm
```
  36bb056e
- G
  error message optimization in softmax_with_cross_entropy_op (#27772) · 7779790c
  由 Guanghua Yu 提交于 10月 12, 2020
```
* error message optimization in softmax_with_cross_entropy_op

* fix some unsuited comment
```
  7779790c
- Z
  
  hsigmoid -> hsigmoid_loss/HSigmoidLoss; refine docs (#27745) · 659d04df
  由 zhupengyang 提交于 10月 12, 2020
  
  659d04df
- L
  Fix bug of Model.save (#27815) · b84d4ae3
  由 LiuChiachi 提交于 10月 12, 2020
```
* fix model bugs, inputs can be InputSpec instance

* correct error message
```
  b84d4ae3
- H
  
  Fix the param of swish (#27824) · f3e2580c
  由 hong19860320 提交于 10月 12, 2020
  
  f3e2580c
- T
  Add double grad in Squeeze and Unsqueeze (#27810) · 070ac959
  由 TeslaZhao 提交于 10月 12, 2020
```
* Add double grad in Squeeze and Unsqueeze

* Add double grad in Squeeze and Unsqueeze
```
  070ac959
- J
  add the kunlun kernel for the paddle 2.0 · d4359b0f
  由 Jack Zhou 提交于 10月 12, 2020
```
Add xpu kernel for KUNLUN core:

* accuracy op
* sign op
* scale op
* sum op

Add default atol in xpu unittest.
```
  d4359b0f
- L
  
  fix summary (#27820) · 4676f03c
  由 LielinJiang 提交于 10月 12, 2020
  
  4676f03c
- M
  
  add XPU support for shape op and reshape op (#27804) · 840d54de
  由 mapingshuo 提交于 10月 12, 2020
  
  840d54de
- W
  
  fleet combine amp dgc recompute meta optimizer (#27643) · 0a1862d1
  由 WangXi 提交于 10月 12, 2020
  
  0a1862d1
- C
  Add test attribute in channelwise_quant op, test=develop (#27742) · 8fabb1c3
  由 cc 提交于 10月 12, 2020
```
* Add test attribute in channelwise_quant op, test=develop
```
  8fabb1c3
- T
  
  Fix docker build error (#27735) · 81d3992c
  由 tianshuo78520a 提交于 10月 12, 2020
  
  81d3992c
11 10月, 2020 3 次提交
- C
  Polish jit.save/load design & remove paddle.SaveLoadConfig (#27623) · 9b49f024
  由 Chen Weihang 提交于 10月 11, 2020
```
* replace config by kwargs

* change save path form dir to prefix

* fix failed unittests

* revert unittest name change

* polish en docs

* add more tests for coverage
```
  9b49f024
- H
  
  Add Swish and ThresholdedReLU for API 2.0 (#27758) · 74d3a550
  由 hong19860320 提交于 10月 11, 2020
  
  74d3a550
- W
  
  update for windows compile. (#27813) · a2d08aa9
  由 Wilber 提交于 10月 11, 2020
  
  a2d08aa9
10 10月, 2020 15 次提交
- W
  add double grad op for matmul (#27776) · ad99e638
  由 wangxinxin08 提交于 10月 10, 2020
```
* add matmul doublegrad op

* fix compile errors

* modify code according to review

* delete float16
```
  ad99e638
- W
  
  [API 2.0]Update 2.0 api from fluid to paddle (#27802) · c425cf18
  由 Wilber 提交于 10月 10, 2020
  
  c425cf18
- I
  
  Modify the output information of the shell script,test=document_fix (#27805) · 1bada985
  由 iducn 提交于 10月 10, 2020
  
  1bada985
- Y
  Polish the documentation and examples of paddle.static.nn.fc. (#27768) · bf187c75
  由 Yiqun Liu 提交于 10月 10, 2020
```
* Reimplement paddle.static.nn.fc, mainly change the parameters and English documentation.

* Polish the documentation format.
```
  bf187c75
- Z
  
  refine APIs: brelu, hardsigmoid, hardswish, maxout (#27658) · 0025e0d8
  由 zhupengyang 提交于 10月 10, 2020
  
  0025e0d8
- Z
  
  add softmax xpu kernel (#27700) · 5098891f
  由 zhupengyang 提交于 10月 10, 2020
  
  5098891f
- L
  disable_fuse_all_reduce (#27746) · 65c06141
  由 Leo Chen 提交于 10月 10, 2020
```
* disable_fuse_all_reduce

* fix test

* fix ut
```
  65c06141
- D
  fix pool3d bug, test=develop (#27718) · f6ad2375
  由 Double_V 提交于 10月 10, 2020
```
* fix pool3d bug, test=develop

* fix unitest, test=develop

* fix test and fix pool2d bug, test=develop
```
  f6ad2375
- 石
  
  save operator version infomation to program desc, test=develop (#27668) · 0d275916
  由石晓伟提交于 10月 10, 2020
  
  0d275916
- Z
  Add api of constant in paddle.nn.initializer (#27786) · 00d401ec
  由 zhulei 提交于 10月 10, 2020
```
* Add api of constant in paddle.nn.initializer

* Add api of constant in paddle.nn.initializer

* Add api of constant in paddle.nn.initializer

* Add api of constant in paddle.nn.initializer

* Add api of constant in paddle.nn.initializer

* Add api of constant in paddle.nn.initializer

* Add api of constant in paddle.nn.initializer
```
  00d401ec
- Q
  
  fix ut error of test_recognize_digits, test=develop (#27791) · b8d2a021
  由 Qi Li 提交于 10月 10, 2020
  
  b8d2a021
- L
  
  modified sample code of add_position_encoding to use paddle.randn, test=document_fix (#27679) · c4b1faa4
  由 Li Fuchen 提交于 10月 10, 2020
  
  c4b1faa4
- L
  Fix dynamic parallel train mode for hapi (#27787) · d3b98f0d
  由 LielinJiang 提交于 10月 10, 2020
```
* fix dynamic parallel for hapi

* fix code style
```
  d3b98f0d
- L
  Fix cross, cumsum op docs test=document_fix · b53970ee
  由 LutaoChu 提交于 10月 10, 2020
```
Fix cross, cumsum op docs
```
  b53970ee
- F
  
  add alias for RNN APIs from paddle.nn.layer.rnn explicitly (#27784) · c52b6bbf
  由 Feiyu Chan 提交于 10月 10, 2020
  
  c52b6bbf

PaddlePaddle / Paddle 大约 1 年 前同步成功

PaddlePaddle / Paddle
大约 1 年前同步成功