提交 · 056599a738a2634c163dea044b188f7770501182 · PaddlePaddle / Paddle

21 3月, 2019 1 次提交
- L
  add expected_kernel_cache_pass · 056599a7
  由 luotao1 提交于 3月 21, 2019
```
test=develop
```
  056599a7
20 3月, 2019 11 次提交

L

Merge branch 'develop' into core_opt_choose_kernel · bfdab00e
由 luotao1 提交于 3月 20, 2019

bfdab00e
T
Merge pull request #16301 from luotao1/runtime_context_pass · a5124ee0
由 Tao Luo 提交于 3月 20, 2019
```
add runtime_context_cache_pass
```
a5124ee0
B
Add softmax_with_cross_entropy_op to ngraph engine (#16304) · 2de263a5
由 baojun 提交于 3月 20, 2019
```
* Add softmax_with_cross_entropy_op test=develop

* simplify implementation test=develop
```
2de263a5
L
Merge branch 'core_opt_choose_kernel' of https://github.com/Xreki/Paddle into... · 6c6a3922
由 luotao1 提交于 3月 20, 2019
```
Merge branch 'core_opt_choose_kernel' of https://github.com/Xreki/Paddle into core_opt_choose_kernel
```
6c6a3922
R
Merge pull request #16202 from shippingwang/add_sqrt_doc · a3b8028d
由 ruri 提交于 3月 20, 2019
```
update sqrt explaination
```
a3b8028d

由 chengduo 提交于 3月 19, 2019

* fuse all_reduce
test=develop

* add fuse_parameter_groups_size
test=develop

* Polish code
test=develop

* Fix travis-ci
test=develop

* Add SetGroupAccordingToLayers and SetGroupAccordingToGroupSize
test=develop

* Add SetGroupAccordingToMemorySize
test=develop

* fix multi_devices_graph
test=develop

* reset params_grads
test=develop

* Polish code
test=develop

f26ba5bd

Z
Merge pull request #16274 from sneaxiy/fix_grad_maker · d0ef6825
由 Zeng Jinle 提交于 3月 19, 2019
```
Remove unused variables in op grad maker
```
d0ef6825
B
Minor ngraph fix (#16270) · 804afc51
由 baojun 提交于 3月 19, 2019
```
* take care edge cases test=develop

* use pragma test=develop
```
804afc51
T
Merge pull request #16280 from luotao1/cos_sim_infershape · 9195c3bb
由 Tao Luo 提交于 3月 20, 2019
```
refine cos_sim infershape
```
9195c3bb

Collective ops (#15572) · 6382b62f

由 Wu Yi 提交于 3月 20, 2019

* wip allreduce in op

* wip

* wip

* wip

* wip adding test

* wip for conflict with mp mode

* fix tests test=develop

* fix cpu build test=develop

* fix travis clang format test=develop

* fix cpu build test=develop

* update api.spec test=develop

* delete comment test=develop

* fix cpplint test=develop

* fix test=develop

* follow comment test=develop

* add file test=develop

* fix build test=develop

* update test=develop

* to be compatible with sync_bn, and fix mp mode in develop test=develop

6382b62f

S
fix op grad maker · 023a3a3d
由 sneaxiy 提交于 3月 19, 2019
```
test=develop
```
023a3a3d

19 3月, 2019 15 次提交
- L
  add runtime_context_cache_pass · 82af8031
  由 luotao1 提交于 3月 19, 2019
```
test=develop
```
  82af8031
- T
  Merge pull request #16287 from PaddlePaddle/revert-16002-runtime_context · b9fc80a1
  由 Tao Luo 提交于 3月 19, 2019
```
Revert "cache runtime_context"
```
  b9fc80a1
- W
  [enhence] Make step_input of dynamic_rnn support custom lod level. (#15972) · 18911b6e
  由 whs 提交于 3月 19, 2019
```
* Make step_input support custom lod level.
test=develop

* Fix API.spec
test=develop

* Fix API.spec.
test=develop

* Fix API.spec
test=develop

* Add default value in document of step_input.
test=develop

* Fix document.
test=develop

* Fix API.spec
test=develop
```
  18911b6e
- L
  refine cos_sim infershape · c05af910
  由 luotao1 提交于 3月 19, 2019
```
test=develop
```
  c05af910
- H
  Merge pull request #16258 from phlrain/fix_concat_1 · d3acf680
  由 Hongyu Liu 提交于 3月 19, 2019
```
Fix concat
```
  d3acf680
- T
  
  Revert "cache runtime_context" · 7d2740db
  由 Tao Luo 提交于 3月 19, 2019
  
  7d2740db
- T
  Merge pull request #16256 from tensor-tang/refine/seqenum · ead558b7
  由 tensor-tang 提交于 3月 19, 2019
```
refine sequence enumerate op
```
  ead558b7
- Q
  Merge pull request #16214 from velconia/imperative_infer_var_type · c7f1f3ed
  由 Qiyang Min 提交于 3月 19, 2019
```
Implement imperative infer var type
```
  c7f1f3ed
- Z
  fix api doc (#16201) · f8df9eb3
  由 Zeng Jinle 提交于 3月 18, 2019
```
test=develop
```
  f8df9eb3
- J
  [MKL-DNN] Fix to crash of Transformer when mkldnn is to be used (#16233) · 13816dd4
  由 Jacek Czaja 提交于 3月 19, 2019
```
* - Fix to crash of Transformer when mkldnn is to be used

Desc: TensorCopy was not setting MKLDNN primitive descriptor when layout was to be kMKLDNN

test=develop

* - Enable transformer for mkl-dnn

test=develo

* - Compilation fix

test=develop

* - Removed manual selection of MKL-DNN ops to be used in Transformer test

test=develop
```
  13816dd4
- Y
  Fix the bug in fp16 backward kernel (#16269) · 7e20e769
  由 Yibing Liu 提交于 3月 19, 2019
```
test=develop
```
  7e20e769
- S
  
  add api.spec, test=develop · 97c60518
  由 shippingwang 提交于 3月 19, 2019
  
  97c60518
- W
  Add cpu_quantize_placement_pass for C-API quantization (#16265) · af030088
  由 Wojciech Uss 提交于 3月 19, 2019
```
* Add cpu_quantize_placement_pass for C-API quantization

test=develop

* added a comment on required pass attributes

test=develop
```
  af030088
- T
  Merge pull request #16002 from luotao1/runtime_context · dbb92ee4
  由 Tao Luo 提交于 3月 19, 2019
```
cache runtime_context
```
  dbb92ee4
- S
  
  Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_sqrt_doc · 4f42504e
  由 shippingwang 提交于 3月 19, 2019
  
  4f42504e
18 3月, 2019 13 次提交
- M
  fix set data type bug · 565b19b7
  由 minqiyang 提交于 3月 18, 2019
```
test=develop
```
  565b19b7
- M
  
  Fix py_func_op's problem · 8364688c
  由 minqiyang 提交于 3月 18, 2019
  
  8364688c
- Z
  Merge pull request #16188 from sneaxiy/fix_const_cast · 6429d2a8
  由 Zeng Jinle 提交于 3月 18, 2019
```
Remove const_cast in optimizers
```
  6429d2a8
- M
  Polish code style · b40e41fb
  由 minqiyang 提交于 3月 18, 2019
```
test=develop
```
  b40e41fb
- X
  Enable INT8 transpose kernel for MobileNet-SSD improvement. (#16159) · e818fa10
  由 xiaolil1 提交于 3月 18, 2019
```
* Enable INT8 transpose kernel for MobileNet-SSD improvement.
test=develop

* Refine the license year.
test=develop

* Delete redundant code.
test=develop

* Add axis check.
test=develop
```
  e818fa10
- X
  Merge pull request #16247 from panyx0718/imperative · 374abcf3
  由 Xin Pan 提交于 3月 18, 2019
```
add more imperative layer tests.
```
  374abcf3
- T
  Merge pull request #16219 from luotao1/fc_infershape · c072998a
  由 Tao Luo 提交于 3月 18, 2019
```
refine fc_infershape
```
  c072998a
- T
  update load persistables for increment, test=develop (#15576) · 8ea4218c
  由 tangwei12 提交于 3月 18, 2019
```
* update load persistables for increment, test=develop

* update load persistables for increment, test=develop

* update API Spec, test=develop

* update API Spec, test=develop

* add doc, test=develop

* add doc, test=develop

* Update lookup_table_utils.py

* Update API.spec

* Update lookup_table_utils.py

test=develop

* Update API.spec

test=develop

* fix api spec

* Update lookup_table_utils.py

test=develop
```
  8ea4218c
- P
  
  fix conncat; test=develop · dcba2e72
  由 phlrain 提交于 3月 18, 2019
  
  dcba2e72
- P
  
  Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into develop · 955fad7a
  由 phlrain 提交于 3月 18, 2019
  
  955fad7a
- P
  
  fix concat; test=develop · a7fe3b50
  由 phlrain 提交于 3月 18, 2019
  
  a7fe3b50
- T
  refine seq enum op · 50931dee
  由 tensor-tang 提交于 3月 18, 2019
```
test=develop
```
  50931dee
- Q
  Merge pull request #16198 from velconia/imperative_train_speed · 8e4ad008
  由 Qiyang Min 提交于 3月 18, 2019
```
Improve imperative mode training speed
```
  8e4ad008

PaddlePaddle / Paddle 大约 1 年 前同步成功

PaddlePaddle / Paddle
大约 1 年前同步成功