提交 · 310edc0d0c9050f1f01c108655493c1935c00214 · 机器未来 / Paddle

24 12月, 2019 4 次提交
- L
  Update layers used in ptb model to use auto-generated op functions in dygraph mode (#21724) · 310edc0d
  由 Leo Chen 提交于 12月 24, 2019
```
* update layers, test=develop

* fix input numpy, test=develop

* fix bugs, test=develop

* follow commments, test=develop

* update getitem, test=develop
```
  310edc0d
- L
  change qat_performance with mobilenet, change batch_size of qat2_resnet50 (#21895) · 9dff56e8
  由 lidanqing 提交于 12月 24, 2019
```
test=develop
```
  9dff56e8
- F
  Update iou_similarity op to support non-normalized bbox (#21671) · 6b9fbcf3
  由 FDInSky 提交于 12月 24, 2019
```
Update iou_similarity op to support non-normalized bbox
```
  6b9fbcf3
- G
  
  Modify the while_loop API (#21844) · 46f9184a
  由 guofei 提交于 12月 24, 2019
  
  46f9184a
23 12月, 2019 2 次提交
- G
  
  Fix default label dim of label_smooth_op. test=develop (#21862) · 7689b6aa
  由 Guo Sheng 提交于 12月 23, 2019
  
  7689b6aa
- G
  optimize fc jit (#21878) · d4dda862
  由 GaoWei8 提交于 12月 23, 2019
```
test=develop
```
  d4dda862
20 12月, 2019 3 次提交

C

fix softmax_with_cross_entropy_fix bug, test=develop (#21810) · 2b941736
由 Chen Weihang 提交于 12月 20, 2019

2b941736

add table id in cache shuffle (#21585) · c3cf42d0

由 Thunderbrook 提交于 12月 20, 2019

* general table

* add sparse table
test=develop

* no cvm
test=develop

* add no_cvm
test=develop

* add note
test=develop

* code style
test=develop

* code style
test=develop

* code style
test=develop

* code style
test=develop

* code style
test=develop

* add key of optimizer
test=develop

* solve pslib stop core
test=develop

* barrier
test=develop

* add notes
test=develop

* add table id in cache shuffle
test=develop

* table id
test=develop

* code style
test=develop

c3cf42d0

Disable memory opt pass when DNNL is on (#21826) · 253e6642

由 Michał Gallus 提交于 12月 20, 2019

* Disable memory opt pass when DNNL is on

* Refine comment above mem optimization pass enablement

test=develop

253e6642

19 12月, 2019 5 次提交
- C
  Speed GEO dense calc & communication (#21579) · a86f11b5
  由 Chengmo 提交于 12月 19, 2019
```
* test=develop, speed dense calc & communication
```
  a86f11b5
- W
  handle multi-inputs with empty inputs for mkldnn_concat_op (#21827) · 666c3bb9
  由 Wojciech Uss 提交于 12月 19, 2019
```
test=develop
```
  666c3bb9
- Z
  Add some debug flags to auto growth allocator (#21766) · aa4d6a5d
  由 Zeng Jinle 提交于 12月 18, 2019
```
* add some debug flags to auto growth allocator, test=develop

* add comments about auto growth, test=develop
```
  aa4d6a5d
- G
  Make While Op could run on GPU place and add while_loop unittest (#21672) · 8b7c50f4
  由 guofei 提交于 12月 19, 2019
```
1. Make while_op accept GPU conditional data
2. Add more complex test cases for while_loop API
```
  8b7c50f4
- W
  
  fix batch_norm_grad infer shape=0 & add allreduce enforce shape, test=develop (#21801) · 17299b8d
  由 WangXi 提交于 12月 19, 2019
  
  17299b8d
18 12月, 2019 2 次提交

Fix Backward Bugs in Conditional Block (#21809) · 557bce77

由 Huihuang Zheng 提交于 12月 18, 2019

The fixed bugs:

1. The condition sub-graph is not pruned
2. When backward graph is extremely simple, the whole backward ops are pruned.

557bce77

X
fix compiled error when with_pslib=on (#21769) · 0eb4d990
由 xujiaqi01 提交于 12月 18, 2019
```
* fix compiled error of butil when with_pslib=on and with_testing=on
* test=develop
```
0eb4d990

17 12月, 2019 1 次提交
- H
  
  Fix That conditional_block_op Doesn't Have InferShape (#21733) · 0677a1c1
  由 Huihuang Zheng 提交于 12月 17, 2019
  
  0677a1c1
16 12月, 2019 6 次提交
- Z
  Fix softmax cuda bug (#21720) · a5a8d144
  由 zhaoyuchen2018 提交于 12月 16, 2019
```
* Fix softmax cuda bug

* Refine multihead log and softmax logic
```
  a5a8d144
- K
  yolo_box OP add Attr(clip_bbox). (#21620) · 943a4449
  由 Kaipeng Deng 提交于 12月 16, 2019
```
* yolo_box OP add Attr(clip_bbox). test=develop
```
  943a4449
- M
  Re-anble vgg and resnet101 models download (#21713) · a5159d84
  由 Michał Gallus 提交于 12月 16, 2019
```
test=develop
```
  a5159d84
- L
  Fix elementwise_pow bug on CUDA place with integer (#21675) · 7181afd7
  由 Leo Chen 提交于 12月 16, 2019
```
* fix elementwise_pow bug on integer, test=develop

* use llrint to support elementwise_pow_grad, test=develop

* add some tests, test=develop

* revert grad functor, test=develop
```
  7181afd7
- 石
  
  fix analysis_predictor when func is called multiple times, test=release/1.6 (#21665) · 2bb13582
  由石晓伟提交于 12月 16, 2019
  
  2bb13582
- L
  Add fc-dequantize squash in cpu_quantize_squash_pass for ernie model (#21714) · d3a96632
  由 lidanqing 提交于 12月 16, 2019
```
* fc-dequantize squash
test=develop

* change according to reviews
test=develop

* change PADDLE_ENFORCE
test=develop

* add second test when fc-dequant do not fuse
test=develop

* change all related PADDLE_ENFORCE
test=develop
```
  d3a96632
15 12月, 2019 2 次提交
- C
  Rename paddle throw error macro (#21657) · 1fd1f06f
  由 Chen Weihang 提交于 12月 15, 2019
```
* rename paddle throw error macro, test=develop

* fix new error use case, test=develop
```
  1fd1f06f
- W
  
  fix std::min type in nan_inf, test=develop (#21725) · 8754cbd1
  由 WangXi 提交于 12月 15, 2019
  
  8754cbd1
12 12月, 2019 4 次提交

L
polish cmake, test=develop (#21681) · fbe3ac21
由 Leo Chen 提交于 12月 12, 2019
```
* polish cmake, test=develop

* add current directory to LD_LIBRARY_PATH, test=develop
```
fbe3ac21

Add reshape int8 mkldnn op (#21428) · d419b859

由 joanna.wozna.intel 提交于 12月 12, 2019

* Add reshape int8 op

test=develop

* Change test to CPUPlace

test=develop

* Correct tests

test=develop

d419b859

W

Rewrite check nan inf tools (#21076) · 8a0f611b
由 WangXi 提交于 12月 12, 2019

8a0f611b

memory leak for cpu (#21174) · 9ad940fd

由 tangwei12 提交于 12月 12, 2019

* add fake init for the trainer, fix large memory hold in the trainer
* do not merge recv vars from a remote endpoint, test=develop
* add recv and save op, merge slice var in one op, save memory
* remove hsigmoid with pull sparse, test=develop

9ad940fd

11 12月, 2019 5 次提交
- Z
  there is bug for inference using auto grwoth allocator (#21621) · fbbd94a6
  由 Zhaolong Xing 提交于 12月 11, 2019
```
test=develop
```
  fbbd94a6
- Z
  Make OperatorWithKernel::InferShape abstract (#21633) · 73461a7a
  由 Zeng Jinle 提交于 12月 11, 2019
```
* make OperatorWithKernel::InferShape virtual, test=develop

* fix test_prepare_op by relu, test=develop
```
  73461a7a
- M
  add `no_need_buffer_slots` interface to pybind (#21575) · 686f0ecb
  由 mapingshuo 提交于 12月 11, 2019
```
* add no_need_buffer_slots interface to pybind
```
  686f0ecb
- Z
  
  fix op_registry, add ignore op_function_impl.h, test=develop (#21654) · 6828f368
  由 Zeng Jinle 提交于 12月 11, 2019
  
  6828f368
- G
  Modify padding strategy: remove weight copy in fc padding (#21650) · 5af0c7ba
  由 GaoWei8 提交于 12月 11, 2019
```
test=develop
```
  5af0c7ba
10 12月, 2019 6 次提交

Refine dygraph DataLoader implementation (#21634) · d96acc33

由 Chen Weihang 提交于 12月 10, 2019

* refine dygraph dataloader & polish related code, test=develop

* refine code based review comment, test=develop

d96acc33

W

fix the mean grad OP performance improvement test=develop (#21658) · 5eec8cf5
由 wangchaochaohu 提交于 12月 10, 2019

5eec8cf5
Z

refine some grad op makers, test=develop (#21629) · 29f64c8c
由 Zeng Jinle 提交于 12月 10, 2019

29f64c8c
M
Dropout with seed (#21590) · e2d849b9
由 mapingshuo 提交于 12月 10, 2019
```
* add seed op
```
e2d849b9

MKL-DNN 1.0 Update (#20162) · e81f0228

由 Adam 提交于 12月 10, 2019

* MKLDNN v1.0 rebase to Paddle 1.6
test=develop

* Add hacky paddle::string::to_string() implementation

* vectorize<int64-t>() -> vectorize() cleanup
test=develop

* PADDLE_ENFORCE and void_cast fixes
test=develop

* Rebase changes
test=develop

* Cosmetics
test=develop

* Delete MKL from mkldnn.cmake
test=develop

* CMake debug commands
test=develop

* Delete MKLDNN_VERBOSE and rebase fixes
test=develop

* Rebase fixes
test=develop

* Temporarily disable int8 resnet101 vgg16 and vgg19 tests
test=develop

* Add libmkldnn.so.1 to python setup
test=develop

* Add libmkldnn.so.1 to inference_lib cmake after rebase
test=develop

* Post rebase fixes + FC int8 changes
test=develop

* Fix LRN NHWC
test=develop

* Fix NHWC conv3d
test=develop

* Windows build fix + next conv3d fix
test=develop

* Fix conv2d on AVX2 machines
test=develop

e81f0228

R
fix: fail to call ZeroCopyTensor::mutable_data() when device_id is no… (#21461) · 7f5d532a
由 rensilin 提交于 12月 10, 2019
```
* ZeroCopyTensor::mutable_data in the right device, test=develop

* add unittest for zerocopy, test=develop
```
7f5d532a

机器未来 / Paddle 与 Fork 源项目一致

机器未来 / Paddle
与 Fork 源项目一致