提交 · 699d5f26388d01666692565ec8e8f4599c993208 · BaiXuePrincess / Paddle

14 9月, 2016 3 次提交

modify RecurrentGradientMachine to support unequal length inputs · 699d5f26

由 zhangruiqing01 提交于 9月 12, 2016

* modify RecurrentGradientMachine to support hasSubSeq sequence inlinks with the same number of sentence but different number of tokens for each sentence

Change-Id: Ic71f00a4bb346b4fa93e650dfb4b1a0d8d2338b0

699d5f26

H
use HPPL_STREAM_DEFAULT for layer computation · 0f91ea7e
由 Haonan 提交于 9月 07, 2016
```
Change-Id: Id66da7b7f5bf9ec80cc19b347e4fb822a5a6f197
```
0f91ea7e

fixed issues with synchronizing streams when copy from gpu to cpu · 688eeefa

由 Haonan 提交于 9月 06, 2016

* by default, synchronize default_stream after resizeAndCopyFrom
* add sync in some places after resizeAndCopyFrom using other streams

688eeefa

12 9月, 2016 1 次提交
- Q
  
  fix CUDNN_VERSION for backward of CudnnBatchNormLayer (#61) · dcd87fd6
  由 qingqing01 提交于 9月 12, 2016
  
  dcd87fd6
09 9月, 2016 5 次提交
- Y
  Fix a PyDataProvider2 bug when use calc_batch_size · 0c8aeffb
  由 Yu Yang 提交于 9月 09, 2016
```
* Need PyGuard when set args for calc_batch_size
```
  0c8aeffb
- Y
  
  Exit when pip install failed · b76682c5
  由 Yu Yang 提交于 9月 09, 2016
  
  b76682c5
- G
  
  fix redundant macro in hl_device_functions.cuh · b159a36e
  由 gangliao 提交于 9月 09, 2016
  
  b159a36e
- Y
  Fix bug in NO_AVX when using lstm · 4a880f03
  由 Yu Yang 提交于 9月 09, 2016
```
* libpaddle_cuda add wrong -mavx to some cpp files.
```
  4a880f03
- Y
  Lazy install Paddle wheels · 7cf8e0c9
  由 Yu Yang 提交于 9月 06, 2016
```
* install wheels when invoke paddle script if current python don't
  have paddle packages, or installed a previous version.
* Also add `make install` to travis
```
  7cf8e0c9
08 9月, 2016 8 次提交
- L
  fix unitest of test_RecurrentGradientMachine, and some tiny doc update · dbaabc94
  由 Luo Tao 提交于 9月 07, 2016
```
Change-Id: I028e402c964ca4f4431cbf8153bea4379dd4df70
```
  dbaabc94
- H
  
  bug fix for hl_matrix_classification_error · 903d5c7e
  由 He 提交于 9月 07, 2016
  
  903d5c7e
- Y
  Fix 32-bit gcc compile warnings. · fdd40e55
  由 Yu Yang 提交于 9月 07, 2016
```
Change-Id: Ibc39ca1d1a27d0d28569e29f41a5647659f8c764
```
  fdd40e55
- X
  Fix ThreadParameterUpdater · 7ad55a4e
  由 xuwei06 提交于 9月 06, 2016
```
The reference return type causes ThreadParameterUpdater.cpp:123 seg fault under gcc5.4.

Change-Id: I7a1c155892722076a7cb48793b83d5ee525747d1
```
  7ad55a4e
- L
  Update Jumbo package to 0.8.0b0. · 721b09ee
  由 liuyuan04 提交于 9月 07, 2016
```
Change-Id: I0b8608feab8f6be5094e8981fc5f65cb401ed415
```
  721b09ee
- Y
  revert CRFLayer, remove wrong gpu support · fbfd24e6
  由 Yu Yang 提交于 9月 07, 2016
```
Change-Id: I636cf13af5becb1168bc9749266b55580c46f6c9
```
  fbfd24e6
- H
  Argument concat for subsequence start positions · 1e1a33b5
  由 Haonan 提交于 9月 06, 2016
```
Change-Id: Ia60c008a8c922f66e6b5e2ca3e488fc4625d6506
```
  1e1a33b5
- Y
  
  Add travis support to PaddlePaddle · daaf5a42
  由 Yu Yang 提交于 9月 08, 2016
  
  daaf5a42
06 9月, 2016 2 次提交

fix bug in findLastSet · d6d85add

由 liaogang 提交于 9月 05, 2016

* findLastSet function: size_t could be  uint, ulong, ulonglong
* add default cuda 7.0 in flags.cmake

Change-Id: Ica9d09520e6a1468e5fc97e2c29c1ea7c73dafbb

d6d85add

L
delete Bits.h and move its func into utils/Util.h · 291318e5
由 liaogang 提交于 9月 02, 2016
```
Change-Id: Ib7f7d3c65c9850b8420f9a4799835a7ea03572bd
```
291318e5

03 9月, 2016 1 次提交
- Y
  Remove the extra endif by merge error · c974a993
  由 Yu Yang 提交于 9月 02, 2016
```
Change-Id: Ic35a54a94cbd981defff5b44a063dc680cdaaa7c
```
  c974a993
02 9月, 2016 2 次提交
- Y
  
  Remove the extra endif by merge error · ecdeaad9
  由 Yu Yang 提交于 9月 02, 2016
  
  ecdeaad9
- F
  
  fixed build issue of double definition of atomicAdd on modern GPUs · c9dc794e
  由 Fan Yang 提交于 9月 01, 2016
  
  c9dc794e
01 9月, 2016 5 次提交
- D
  Refine code and comments for CRMNormLayer. · 54d4968d
  由 dangqingqing 提交于 9月 01, 2016
```
ISSUE=4612671

git-svn-id: https://svn.baidu.com/idl/trunk/paddle@1501 1ad973e4-5ce8-4261-8a94-b56d1f490c56
```
  54d4968d
- L
  âfix bug in cuda_aggregate · 8ae1b325
  由 liaogang 提交于 9月 01, 2016
```
ISSUE=4608831

git-svn-id: https://svn.baidu.com/idl/trunk/paddle@1498 1ad973e4-5ce8-4261-8a94-b56d1f490c56
```
  8ae1b325
- H
  fix cuda-8.0 compile · 24acf8d5
  由 hedaoyuan 提交于 9月 01, 2016
```
ISSUE=4613907

git-svn-id: https://svn.baidu.com/idl/trunk/paddle@1497 1ad973e4-5ce8-4261-8a94-b56d1f490c56
```
  24acf8d5
- F
  
  fixed build issue of double definition of atomicAdd on modern GPUs · f2aa2747
  由 Fan Yang 提交于 9月 01, 2016
  
  f2aa2747
- Y
  Fix style check error in api. · df826899
  由 yuyang18 提交于 8月 31, 2016
```
ISSUE=4611151



git-svn-id: https://svn.baidu.com/idl/trunk/paddle@1470 1ad973e4-5ce8-4261-8a94-b56d1f490c56
```
  df826899
31 8月, 2016 10 次提交

ISSUE=4607611 refine cluster scripts · cecdedea

由 wangyanfei01 提交于 8月 31, 2016

git-svn-id: https://svn.baidu.com/idl/trunk/paddle@1461 1ad973e4-5ce8-4261-8a94-b56d1f490c56

cecdedea

Add missing file for last commit · 9e83dac0

由 yuyang18 提交于 8月 31, 2016

also rename hdfs_data to make it internally.

ISSUE=4604505


git-svn-id: https://svn.baidu.com/idl/trunk/paddle@1451 1ad973e4-5ce8-4261-8a94-b56d1f490c56

9e83dac0

Add PyDataProvider2 DataConverter to swig api. · d8f30da7

由 yuyang18 提交于 8月 31, 2016

* fix recommendation prediction also.

ISSUE=4561941



git-svn-id: https://svn.baidu.com/idl/trunk/paddle@1449 1ad973e4-5ce8-4261-8a94-b56d1f490c56

d8f30da7

bug fix for multi cpu machine training · 8fe4a338

由 hedaoyuan 提交于 8月 31, 2016

ISSUE=4602845

git-svn-id: https://svn.baidu.com/idl/trunk/paddle@1448 1ad973e4-5ce8-4261-8a94-b56d1f490c56

8fe4a338

add some source code comments · 66be6fed

由 qijun 提交于 8月 31, 2016

ISSUE=4592951

git-svn-id: https://svn.baidu.com/idl/trunk/paddle@1447 1ad973e4-5ce8-4261-8a94-b56d1f490c56

66be6fed

Fix name typos · 890ca39b

由 lipeng17 提交于 8月 31, 2016

ISSUE=4600925


git-svn-id: https://svn.baidu.com/idl/trunk/paddle@1443 1ad973e4-5ce8-4261-8a94-b56d1f490c56

890ca39b

Add missing file for last commit · 76b71d10

由 yuyang18 提交于 8月 31, 2016

ISSUE=4600307


git-svn-id: https://svn.baidu.com/idl/trunk/paddle@1439 1ad973e4-5ce8-4261-8a94-b56d1f490c56

76b71d10

Y
Add pillow to docker image, reduce docker image compile time. · 9af628c1
由 yuyang18 提交于 8月 31, 2016
```
ISSUE=4600057



git-svn-id: https://svn.baidu.com/idl/trunk/paddle@1438 1ad973e4-5ce8-4261-8a94-b56d1f490c56
```
9af628c1

Fix ubuntu image prefix · a46749b3

由 yuyang18 提交于 8月 30, 2016

ISSUE=4599563


git-svn-id: https://svn.baidu.com/idl/trunk/paddle@1435 1ad973e4-5ce8-4261-8a94-b56d1f490c56

a46749b3

Try Fix docker hub build. · 291ff97c

由 yuyang18 提交于 8月 30, 2016

ISSUE=4599521


git-svn-id: https://svn.baidu.com/idl/trunk/paddle@1434 1ad973e4-5ce8-4261-8a94-b56d1f490c56

291ff97c

30 8月, 2016 3 次提交

multi-gpu training with sparse input. · 2afe6609

由 hedaoyuan 提交于 8月 30, 2016

ISSUE=4586833

git-svn-id: https://svn.baidu.com/idl/trunk/paddle@1429 1ad973e4-5ce8-4261-8a94-b56d1f490c56

2afe6609

Bug fix for image classification demo and ResNet model zoo when using CPU. · 7857d7f7

由 dangqingqing 提交于 8月 30, 2016

But need to check consistency of CPU and GPU later for conv layer.
ISSUE=4592155

git-svn-id: https://svn.baidu.com/idl/trunk/paddle@1427 1ad973e4-5ce8-4261-8a94-b56d1f490c56

7857d7f7

Try to fix hub.docker.com build error. · 57109591

由 yuyang18 提交于 8月 30, 2016

Maybe is disk usage issue.

ISSUE=4596347


git-svn-id: https://svn.baidu.com/idl/trunk/paddle@1426 1ad973e4-5ce8-4261-8a94-b56d1f490c56

57109591

BaiXuePrincess / Paddle 与 Fork 源项目一致

BaiXuePrincess / Paddle
与 Fork 源项目一致