Commit Graph

4619 Commits (32a15ed90c17a4865cf0d0e070b8ba09f307ebd6)

Author SHA1 Message Date
fengjiayi e84d3a7fbf Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_parallel_executor_tests
7 years ago
Wu Yi 8352f93809
Merge pull request #9933 from typhoonzero/add_dist_unit_test
7 years ago
Jacek Czaja de8094f57e Cosmetic fixes
7 years ago
Jacek Czaja acdf7cbd19 - Added EPS for softmax MKLDNN op
7 years ago
Luo Tao 81c47b21ef add type check and default scope
7 years ago
typhoonzero 69188e5981 fix ut
7 years ago
typhoonzero c6d7c2bdff Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into split_byref_op
7 years ago
Luo Tao 01b88f2177 Merge branch 'develop' into merge_bn
7 years ago
mozga-intel 6e7b883bdd Initial implementation of multiplication operator for MKLDNN
7 years ago
Tao Luo fee5b24cb4
Merge pull request #9985 from mozga-intel/mozga-intel/mkldnn_separate_tests
7 years ago
typhoonzero ed89b7b7e6 dist train use split_by_ref
7 years ago
typhoonzero 22f58382bc run dist train test single process
7 years ago
Yu Yang fc6f0be276
Merge pull request #9942 from reyoung/feature/tuning_pe_trans
7 years ago
JiayiFeng 1de9edee51 Follow comments
7 years ago
JiayiFeng 93276fd146 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_parallel_executor_tests
7 years ago
typhoonzero 0c6eef3e58 add split by ref test
7 years ago
JiayiFeng ab86fb11c4 complete parallel accuracy test
7 years ago
mozga-intel 5e13314dc1 Commit
7 years ago
Yu Yang 7286954337 Add more comments
7 years ago
tangwei12 e05f3eaf69 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into quick_help
7 years ago
Yu Yang a1d910b144 Fix line witdth
7 years ago
Yu Yang b4aaa00a8a Polish logic of ParallelExecutor
7 years ago
Yu Yang c0f4e0332c Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/simplify_delay_logic
7 years ago
wanglun 3fbe9c34f7 Fix typo in docstring (#9829)
7 years ago
typhoonzero d78cab1519 update
7 years ago
guosheng f086ccb7b6 Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into add-python-label-smooth
7 years ago
JiayiFeng 415460b557 stash
7 years ago
JiayiFeng e412b1aec7 Merge branch 'unify_executor_interface' into add_parallel_executor_tests
7 years ago
JiayiFeng 22df230ee4 rename 'feed_dict' in ParallelExecutor.run() to 'feed'
7 years ago
JiayiFeng fbd5cf604f stash
7 years ago
fengjiayi 5fc8326732 Add parallel accuracy test
7 years ago
guosheng d04073050e Add python wrapper for label smoothing
7 years ago
typhoonzero 745aacfc38 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_dist_unit_test
7 years ago
typhoonzero c4de344ce2 update
7 years ago
typhoonzero 2b05b010ff fix chunk evaluator name in metrics
7 years ago
typhoonzero 34cffe9169 add unitest file
7 years ago
typhoonzero 8c1b257596 add dist unitest data compare
7 years ago
Yu Yang d0342f12aa Simplify DelayOps Logic
7 years ago
dzhwinter e54f203c55 "move to a new PR"
7 years ago
fengjiayi 1b5a6351af make for_parallel default True
7 years ago
Xin Pan 2f38264078
Merge pull request #9905 from panyx0718/mem-opt
7 years ago
Xin Pan d4024a6ebd Polish memory_optimizer code with mode comments and less identions
7 years ago
Yu Yang e4cfe477db
Merge pull request #9898 from reyoung/feature/mix_cpu_gpu_op
7 years ago
Yancey 9bc0c23b17
Merge pull request #9901 from Yancey1989/fix_duplicate_lr_op
7 years ago
Luo Tao ec512cdcce add comment for branch network
7 years ago
fengjiayi c1bf06f9d9
Merge pull request #9289 from dzhwinter/speed/sequence_expand
7 years ago
Yu Yang ed2d7d7d26 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/mix_cpu_gpu_op
7 years ago
Yancey1989 253441b553 fix duplicate lr op after distribute transpiler
7 years ago
Yu Yang 482314e3b1 Add CRF unittest
7 years ago
Luo Tao f45818e7f9 create new varible in scope
7 years ago
Yancey 41a9146de3
Merge pull request #9787 from Yancey1989/random_selected_rows_value
7 years ago
Wu Yi 1bdc7261a5
Merge pull request #9578 from typhoonzero/threadpool_for_io
7 years ago
Luo Tao 6e735e1e6e Merge branch 'develop' into merge_bn
7 years ago
mozga-intel d24b5e060f The fully connected: the operator is removed when the MKLDNN flag is OFF
7 years ago
jiaozhenyu ddf5783644
Merge pull request #9812 from jshower/develop
7 years ago
tangwei12 a006c54c11 fix code style #9660
7 years ago
tangwei12 e5113da39c fix code style #9660
7 years ago
tangwei12 5ee67f15bf fix code style #9660
7 years ago
tangwei12 2dfd8841a4 make quick start code more simpler #9660
7 years ago
Qiao Longfei 4c55a6022a
Dist transpiler support prefetch (#9714)
7 years ago
typhoonzero 26cfc634b9 multi stream thread pool
7 years ago
Yancey1989 7132bbe6b7 update by comment
7 years ago
whs ef169eb996
Merge pull request #9459 from wanghaoshuang/fix_avg
7 years ago
Tao Luo 648908004e
Merge pull request #9849 from qingqing01/parallel_exe_test
7 years ago
Yang Yang(Tony) a64edbf14e
delete backward.cc related code on the python side (#9854)
7 years ago
Yang Yang(Tony) b26f505002
remove net op and cond_op (#9663)
7 years ago
Sharan Narang 09c1c692f3 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into sentiment_analysis
7 years ago
tangwei12 554cea6839 move most codes to uci_housing #9660
7 years ago
Dang Qingqing 9fd76c4f3a Improve test_parallel_executor.
7 years ago
Wu Yi 652cf43002
Merge pull request #9746 from typhoonzero/multigpumultinode
7 years ago
Tao Luo 2a3d49040c
Merge pull request #9786 from weixing02/initializer
7 years ago
Yancey1989 1aada35206 Merge branch 'develop' of github.com:PaddlePaddle/Paddle into random_selected_rows_value
7 years ago
dzhwinter d52fa26fda
Feature/metrics (#9791)
7 years ago
typhoonzero dfc60251bf Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into multigpumultinode
7 years ago
fengjiayi 90084a25d2
Merge pull request #9743 from JiayiFeng/modify_readers_to_fit_parallel_executor
7 years ago
Luo Tao 7815cdffba use clone method to flush in forth
7 years ago
dzhwinter 996b3e15fa Merge remote-tracking branch 'origin/develop' into speed/sequence_expand
7 years ago
JiayiFeng 8c1eb8693e update unittest
7 years ago
fengjiayi 0dacbbe1fe update multi_pass_reader unittest
7 years ago
JiayiFeng b1cc28dab3 update
7 years ago
JiayiFeng 273f4892b2 update recordio unittest
7 years ago
JiayiFeng 75a2e50581 Merge branch 'modify_dev' into modify_readers_to_fit_parallel_executor
7 years ago
JiayiFeng 72b5de05fe update unittest
7 years ago
Luo Tao 3f320c1345 Merge branch 'develop' into merge_bn
7 years ago
wanghaoshuang a7c6bf771c Change do_model_average_for_mean_and_var to boolean in batch_normal.
7 years ago
Tao Luo 0ad892a0d4
Merge pull request #9816 from luotao1/remove_op
7 years ago
typhoonzero ec6976810d Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into multigpumultinode
7 years ago
Sharan Narang fc346e3432 Add cutoff parameter to build_dict.
7 years ago
fengjiayi a84b81502c Remove Readers' HasNext()
7 years ago
typhoonzero ce08dc8751 have stream removed error
7 years ago
Luo Tao 5483258b7e fuse batch norm for conv operator with bias
7 years ago
jshower ad6ddf533c for ci
7 years ago
typhoonzero 0bf799a523 wip testing
7 years ago
JiayiFeng 284a213774 fix a name conflict
7 years ago
jshower 7c1434dd73 code style
7 years ago
weixing02 5ecfb62de7 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into initializer
7 years ago
chengduo e0babe7c3e
Merge pull request #9780 from chengduoZH/feature/fix_batch_size_is_littler_than_gpu_count
7 years ago
Luo Tao e7467d94a0 add remove_op, remove_var in Python end
7 years ago
weixing02 56b04e5b2c Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into initializer
7 years ago
weixing02 93940642ac Adjust
7 years ago
Yancey1989 cb7bbf426c revert uniform_random_op
7 years ago
Yancey1989 291aa23153 Merge branch 'random_selected_rows_value' of github.com:Yancey1989/Paddle into random_selected_rows_value
7 years ago
Yancey1989 1cc09c7bc5 Merge branch 'develop' of github.com:PaddlePaddle/Paddle into random_selected_rows_value
7 years ago
Yancey1989 3f6fc10b9f new op that init table value randomly
7 years ago
Luo Tao ea0cf6f382 rewrite inference_transpiler in Python end
7 years ago
jshower d9a5222385 code style
7 years ago
Wu Yi b44b6a4f00
Merge pull request #9798 from typhoonzero/fix_dist_transpiler_bug
7 years ago
JiayiFeng ee178d5aeb fix bugs
7 years ago
Luo Tao 16e31343d8 Merge branch 'develop' into merge_bn
7 years ago
jshower 9fe938cb2a Changing network configuration, avoid nan
7 years ago
weixing02 71081a9bb9 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into initializer
7 years ago
Kexin Zhao 0f38bb4593
add fp16 support to activation op (#9769)
7 years ago
Xin Pan 326b434ab5
Merge pull request #9699 from panyx0718/test
7 years ago
typhoonzero 92313a99c0 update
7 years ago
typhoonzero d02b17e597 fix dist transpiler bug
7 years ago
weixing02 919aa2ea0a Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into initializer
7 years ago
weixing02 cd41dca2ab Adjust
7 years ago
Luo Tao 17833d30e3 fuse batch norm for conv operator without bias
7 years ago
JiayiFeng fe48dfcb6e Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into modify_dev
7 years ago
Xin Pan bbbc3592db
Merge pull request #9735 from panyx0718/dist
7 years ago
JiayiFeng 759979998c Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into modify_dev
7 years ago
Yancey1989 f909ff1a36 update unit test
7 years ago
weixing02 0586b9e5c9 Add *Initializer to fluid api docs
7 years ago
fengjiayi 5416bac5d8 Make shared decorated readers' creater be only in main_program
7 years ago
chengduoZH 7e7611d067 when the number of samples of current batch is less than the count of devices, let it crash.
7 years ago
Yu Yang 31464f3425
Merge pull request #9772 from reyoung/feature/add_gitignores
7 years ago
Yu Yang a4bf635795 Add gitignores
7 years ago
JiayiFeng 3f90a583b4 update unittest
7 years ago
qiaolongfei 055662f9f4 add insert_op for block
7 years ago
qingqing01 2b7e5bd366
Support testing during training by ParallelExecutor. (#9738)
7 years ago
Xin Pan 75c9eb11c4 refine
7 years ago
fengjiayi 9b3c964341 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into modify_readers_to_fit_parallel_executor
7 years ago
Tao Luo ec89ed69ef
Merge pull request #9655 from kbinias/kbinias/mkldnn-activations-dim2
7 years ago
gongweibao c1b6692fe5
Fix debuger bugs. (#9705)
7 years ago
JiayiFeng 5ad2486905 fix errors
7 years ago
typhoonzero 4459fa6726 fix seq tag model dist transpile
7 years ago
fengjiayi fca9e8847d Update Readers Python API
7 years ago
fengjiayi 6be51f10ec Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into modify_readers_to_fit_parallel_executor
7 years ago
Xin Pan bdb47cd933 Add some comments for distribute_transpiler
7 years ago
fengjiayi 38ba7e5af4 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into update_reader
7 years ago
gongweibao 6651301b29
Modify comments of `sync_with_cpp`
7 years ago
Wu Yi 3682035f14
Merge pull request #9695 from emailweixu/round_robin
7 years ago
liuguoyi 9d95ccc887 fix pool and conv3d unittest when paddle is not build with gpu
7 years ago
lgone2000 6ba262572c fix test_conv2d_op when compile without cuda (#9698)
7 years ago
Xin Pan b138d29c38 Avoid init_p2p all the times
7 years ago
xuwei06 560d960b27 Fix a minor bug for distributed_spliter.round_robin
7 years ago
Kexin Zhao b2a1c9e8b7 Add float16 support to non-cudnn softmax op on GPU (#9686)
7 years ago
chengduo 3d3e5aeb9c
Merge pull request #9648 from chengduoZH/feature/expose_CUDAPinnedPlace_to_python
7 years ago
Xin Pan 0935f2cf05 shrink test_parallel_executor size.
7 years ago
Lei Wang 09b4a1a361 Build: generate all the build related files into one directory. (#9512)
7 years ago
qiaolongfei ff4208e676 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into sgd-support-update-selected-rows
7 years ago
chengduoZH 17842e33c3 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/expose_CUDAPinnedPlace_to_python
7 years ago
Xin Pan 043c23080c
Merge pull request #9637 from panyx0718/feed
7 years ago
chengduoZH 8e4e155c52 add PyCUDAPinnedTensorSetFromArray
7 years ago
Xin Pan 92e92ceba1 Add type check
7 years ago
Krzysztof Binias c4107748d0 Add support for dim equals 2 in activation functions
7 years ago
wanghaoshuang 2e40660e7a Fix some issues.
7 years ago
Tao Luo 5eb9cecc4a
Merge pull request #9607 from luotao1/remove_var
7 years ago
Wu Yi c00a5dec78
Merge pull request #9644 from typhoonzero/fix_condition_op_in_optimize
7 years ago
qingqing01 a84a580e65
Add CUDA kernel for prior_box_op. (#9553)
7 years ago
typhoonzero a16a872783 update
7 years ago
fengjiayi 6dcfd97a92 add docstring
7 years ago
typhoonzero e0b396e7ba update by comment
7 years ago
fengjiayi af24290123 Add 'buffer_size' api for open_files op
7 years ago
typhoonzero 39277e9282 fix transpiler condition op in optimize
7 years ago
Xin Pan 4bbfa9eccb Add feed to ParallelExecutor
7 years ago
Tao Luo a98a3fdc46
Merge pull request #9385 from mozga-intel/mozga/mkldnn-fc
7 years ago
Zhaolong Xing d02a68c447
Merge pull request #5768 from NHZlX/add_upsample_layer
7 years ago
wanghaoshuang 19db989ec2 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_avg
7 years ago
Luo Tao 09b53c086d add remove_var from c++ end
7 years ago
mozga-intel 32f8ac7d3b Remove additional message
7 years ago
mozga-intel 34a8084328 Added new fc files, register fc kernel
7 years ago
mozga-intel 2811ea4440 Implementation of MKLDNN FC
7 years ago
qiaolongfei cbfec1f7d6 diable test of sparse_parameter_sgd on GPU
7 years ago
Tao Luo 957104564f
Merge pull request #9600 from luotao1/sync_with_cpp
7 years ago
Xin Pan 49313d4066
Merge pull request #9548 from panyx0718/group_nccl_all_reduce
7 years ago
guosheng 3c370ee0d5 Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into add-python-pad
7 years ago
Luo Tao 103407aa60 refine sync_with_cpp when remove ops or remove vars
7 years ago
Xin Pan cf251eb8cf shrink test size
7 years ago
qiaolongfei af1d3f5bc0 remove print
7 years ago
qiaolongfei a994327fb1 add TestSGDOpOptimizeSelectedRows
7 years ago
JiayiFeng de2d82d6a9 fix a bug
7 years ago
Xin Pan b25a9c488d Reduce test size to avoid GPU memory issue.
7 years ago
Xin Pan b123ce88a1 Add enable/disable for delayed ops
7 years ago
Xin Pan 46f3a39e91 polish and add comments.
7 years ago
Xin Pan d0ac92531d Improve ParallelExecutor performance
7 years ago
qiaolongfei abb7deee39 optimize test_lookup_table_op.py
7 years ago
fengjiayi 649ae2700e fix bugs
7 years ago
dzhwinter c72450d24d "seperate test"
7 years ago
fengjiayi 44d5f42a7e update reader
7 years ago
qiaolongfei e64dda7e44 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into sgd-support-update-selected-rows
7 years ago
qiaolongfei b154f311b9 init TestSGDOpOptimizeSelectedRows
7 years ago
Qiao Longfei 37d9a72e9e
Merge pull request #9575 from jacquesqiao/lookup_table_support_SelectedRows_as_parameter
7 years ago
qiaolongfei 13ecb5e50a Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into lookup_table_support_SelectedRows_as_parameter
7 years ago
Helin Wang f801f743d1 fix the __all__ in dataset package
7 years ago
Helin Wang f837eee724 add paddle.v2.reader,dataset back for backward compatibility
7 years ago
guosheng 0b8534f2a4 Refine python wrapper for pad_op
7 years ago
qiaolongfei f02968bb21 disable test_recv_op
7 years ago
qiaolongfei 2be10ebe8a disable test_recv_op
7 years ago
qiaolongfei 7a6ffb6280 add TestLookupTableWIsSelectedRows
7 years ago
dzhwinter 62ba8723ac Merge remote-tracking branch 'origin/develop' into speed/sequence_expand
7 years ago
guosheng 629c9210d1 Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into add-python-pad
7 years ago
guosheng 4c8eef5739 Add python wrapper for pad_op
7 years ago
武毅 b9d8bbe4f2
Merge pull request #9558 from typhoonzero/fix_dist_transpile_one_trainer
7 years ago
Guo Sheng d908c3b2fa
Merge pull request #9008 from lcy-seso/enhance_reshape
7 years ago
typhoonzero 772cdfe196 fix single pserver error
7 years ago
dzhwinter 6cfc0c1497
"polish code" (#9318)
7 years ago
wanghaoshuang 61f4706b1d Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_avg
7 years ago
wanghaoshuang 9708b21f19 Refine average model option
7 years ago
Helin Wang bcf7c36b0b Make paddle.fluid no longer depends on paddle.v2
7 years ago
fengjiayi 95658767eb
Merge pull request #9428 from JiayiFeng/kernel_of_increment_op
7 years ago
Yancey 374f1ca3b7 Fix dist error with lr decay layer (#9489)
7 years ago
wanghaoshuang d1a7b47e04 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_avg
7 years ago
Yang Yu af230d9bef Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into cpp_parallel_executor
7 years ago
xzl 663f0358a6 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_upsample_layer
7 years ago
Yang Yu 34a440fa64 Revert "make append activation in place by default (#9417)"
7 years ago
dzhwinter a80bf702f3 Merge remote-tracking branch 'origin/develop' into speed/sequence_expand
7 years ago
dzhwinter b661fe1d76 "fix ci"
7 years ago
dzhwinter 8425c2c859
Speed/sequence op1 (#9217)
7 years ago
guosheng 5b8bb34470 Refine reshape_op by following comments.
7 years ago
chengduo 24100e1fb8
Merge pull request #9449 from chengduoZH/feature/add_cos
7 years ago
fengjiayi 869ef01d66 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into kernel_of_increment_op
7 years ago
JiayiFeng 52574733a6 Add KernelType switch for IncrementOp kernel
7 years ago
chengduoZH bdda08d9f2 add sin
7 years ago
JiayiFeng 01c5ca7364 fix bugs
7 years ago
Yang Yang(Tony) ce16400dae
make append activation in place by default (#9417)
7 years ago
JiayiFeng 917b205c1c Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into kernel_of_increment_op
7 years ago
chengduoZH 2e577379ca add cos
7 years ago
wanghaoshuang 764723d4f5 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_avg
7 years ago
wanghaoshuang e1290c4fd7 Make Average Model support for 'moving mean' and 'moving variance' of batch_normal op
7 years ago
Tao Luo 857a8997de
Merge pull request #9384 from luotao1/removeVar
7 years ago
xzl 4747b2c92d Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_upsample_layer
7 years ago
dzhwinter 0be1e09f2c "fix ci"
7 years ago
Yang Yu b0775588c0 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into cpp_parallel_executor
7 years ago
fengjiayi 055fb215a1 remove unnecessary 'force_cpu'
7 years ago
Yu Yang f2d29be784 Disable transformer
7 years ago
guosheng 09743b6117 Refine test_reshape_op
7 years ago
Yu Yang 5408854090 Disable model evaluation in unittests
7 years ago
Yang Yang f34f2d4026 make bn inplace in img_conv_group by default
7 years ago
Yang Yang 54a8c04fab add inplace attr to bn
7 years ago
guosheng 4bfbc59122 Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into enhance-ReshapeOp
7 years ago
guosheng c078ed4608 Enhance reshape_op by adding Input(Shape)
7 years ago
Luo Tao 7f4012247e adjust remove rule for variables
7 years ago
Yu Yang c42c4a6718 Add performance tests
7 years ago
Yu Yang 9af870854e Use heap variables
7 years ago
dzhwinter 5447046aee merge develop branch
7 years ago
qingqing01 123cf165fb
Set stop_gradient=True for some variables in SSD API. (#9396)
7 years ago
Tao Luo fe142d92ff
Merge pull request #9359 from wanglun/develop
7 years ago
Tao Luo 1b67bc022c
Merge pull request #9329 from tpatejko/tpatejko/mkldnn-lrn
7 years ago
Thuan Nguyen c7bf77d0e1 Add in is_copy attribute to SelectCase. (#9393)
7 years ago