Commit Graph

2005 Commits (1943119fc5f98f6b552ebb6d180346b9c27adb8e)

Author SHA1 Message Date
Xin Pan 61491ce250 clean
6 years ago
Xin Pan ce7e503cbe refactor to avoid scope.
6 years ago
Qiyang Min 0238a3bb4f
Merge pull request #14972 from velconia/accelerate_lstm
6 years ago
Houjiang Chen 242d3c71a6
Merge pull request #15031 from hjchen2/develop
6 years ago
Xin Pan 71a4a8e981
Merge pull request #15071 from wopeizl/revert/15035
6 years ago
Qiao Longfei 3b294e2e2e Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into multithread-sparse-adam
6 years ago
sneaxiy c4ce2e7b21 merge develop, solve conflict
6 years ago
minqiyang 8ed0233924 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into accelerate_ddpg
6 years ago
Zeng Jinle 9c6a0203e2
Merge pull request #15073 from sneaxiy/add_scope_pool
6 years ago
sneaxiy b56aca82e9 merge develop
6 years ago
sneaxiy ee83ce75bf try to fix py35 compile error
6 years ago
sneaxiy 3e917a934a add scope_pool
6 years ago
Yancey1989 1a4f79a7de fix unittest test=develop
6 years ago
Yancey1989 86bb583881 Merge branch 'develop' of github.com:PaddlePaddle/Paddle into parallel_graph_mode
6 years ago
Yancey1989 495e73d766 enable gc
6 years ago
Yancey1989 28cdfbc2b0 delete comment code
6 years ago
Yancey1989 845bfd5807 cleanup code
6 years ago
peizhilin 2388d0e7d6 Revert "cherry-pick the #12759"
6 years ago
peizhilin 01c00b07dd fix test issues on windows
6 years ago
peizhilin 1e7f83e60a add cuda dso support for windows
6 years ago
tangwei12 dc8eca826e
code style fix, test=develop (#15045)
6 years ago
Yancey1989 41a64f6a2a Merge branch 'develop' of github.com:PaddlePaddle/Paddle into parallel_graph_mode
6 years ago
nhzlx a6aa8ea771 faster rcnn input is presistable. (fix it in paddle-trt)
6 years ago
hjchen2 956cf92145 Fix conv_elementwise_add2_act pass
6 years ago
Tao Luo 69659f4ae2
Merge pull request #15037 from jianhang-liu/fix/abnormal_stack_op_time
6 years ago
sneaxiy 179acc60b3 fix conflict with develop
6 years ago
wopeizl 09bd8fa67a
Merge pull request #15035 from wopeizl/debug/improvement1
6 years ago
sneaxiy dde3afe7b7 Merge develop
6 years ago
dongdaxiang 2df1d80767 Merge branch 'add_timer' of https://github.com/guru4elephant/Paddle into add_timer
6 years ago
Wu Yi 856f0da0fe
Fp16 training (#14992)
6 years ago
Brian Liu e821b12f57 Fix issue which cause abnormal CPU usage in stack op
6 years ago
chengduo b9fb03cf54
Move GetTensor to tensor_util (#15011)
6 years ago
nhzlx 73b47df1f4 Merge branch 'develop' of https://github.com/paddlepaddle/paddle into add_affine_channel_fuse
6 years ago
nhzlx ce3782c193 add affine_channel fuse.
6 years ago
peizhilin 7f6d8acecb cherry-pick the #12759
6 years ago
sneaxiy 3a2afbf02e polish code
6 years ago
tensor-tang 05d1121b22
Merge pull request #14802 from mozga-intel/mozga-intel/fill_constant_operator_ngraph
6 years ago
tensor-tang 9d4f1d468a
Merge pull request #14804 from mozga-intel/mozga-intel/top_k_operator_ngraph
6 years ago
sneaxiy 68d91cd594 add copy ctor
6 years ago
dongdaxiang 3b3cb4ea55 Merge branch 'add_timer' of https://github.com/guru4elephant/Paddle into add_timer
6 years ago
sneaxiy e02f67eff7 rewrite unsafe_cast
6 years ago
minqiyang 68b86d6665 Change default value to align with the original react
6 years ago
dongdaxiang 2dee8f6cd5 add TrainFilesWithTimer in async_executor
6 years ago
dongdaxiang d434fcbaa6 add TrainFilesWithTimer in async_executor
6 years ago
minqiyang 250e893745 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into accelerate_ddpg
6 years ago
Xin Pan 103f08f50e
Merge pull request #14910 from panyx0718/clean3
6 years ago
Zeng Jinle 0021b05b19
Merge pull request #14993 from sneaxiy/fix_check_lod
6 years ago
chengduo 79bd6dfa18
[Feature] Add Temporary Allocator (#14875)
6 years ago
sneaxiy a30c5373eb use std::is_sorted
6 years ago
minqiyang 8149a07a41 Fix wait stream two times bug
6 years ago
sneaxiy b8051e7927 merge develop
6 years ago
Tao Luo df1e4e2f10 fix check_lod
6 years ago
minqiyang 0a4b6fc056 Remove unnessesary code
6 years ago
minqiyang 53619a79b4 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into accelerate_lstm
6 years ago
minqiyang 6fabbd8fb8 Polish code and remove spin lock
6 years ago
Zeng Jinle 95cbe07c40
Merge pull request #14836 from sneaxiy/feature/py_func
6 years ago
mozga-intel 7048caf9a0 Enable top_k operator for a ngraph
6 years ago
mozga-intel ecfa68ecaa Enable fill_constant operator for a ngraph
6 years ago
sneaxiy 600f6d8272 polish code
6 years ago
sneaxiy 7f6e513b1f fix mac ci bug
6 years ago
sneaxiy c1f7e54f62 merge develop
6 years ago
typhoonzero da87f7a698 Revert "[Feature] Fp16 training for resnet50 (#14850)"
6 years ago
sneaxiy 89b9d86d9d fix windows compile bug
6 years ago
Qiao Longfei d76bda50c4 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into multithread-sparse-adam
6 years ago
sneaxiy 490eb9061f polish infer shape of py_func op
6 years ago
Xin Pan 969ad966ba all converted
6 years ago
Xin Pan a872eb90c2
Merge pull request #14959 from panyx0718/clean2
6 years ago
sneaxiy 13429c3e9f clean code, remove void registration
6 years ago
chengduo 550e7e410b
Code Clean parallel_executor.py (#14849)
6 years ago
Wu Yi 3d750f9c5a
[Feature] Fp16 training for resnet50 (#14850)
6 years ago
minqiyang 679d1a9e0b Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into accelerate_lstm
6 years ago
sneaxiy 83ac85158a polish code
6 years ago
sneaxiy 045dc12766 merge develop
6 years ago
sneaxiy ce4a26ddad clean code
6 years ago
Tomasz Patejko e9eee0de6a MKLDNN residual connection fuse: fixing accuracy problem (#14874)
6 years ago
sneaxiy 53f6c6991a polish code
6 years ago
sneaxiy 74a8e6b032 merge develop
6 years ago
Xin Pan 1fe3ac352a move more and fix while
6 years ago
sneaxiy ae6f46a1a9 rewrite variable type
6 years ago
Xin Pan 9ef8a76873 convert more
6 years ago
Xin Pan 876993887b convert more interface to avoid scope
6 years ago
Xin Pan 8c19f0bfe3 fix
6 years ago
mozga-intel 9035bb81fe Enable mul operator for a ngraph engine (#14801)
6 years ago
Xin Pan 4dd61e7260 convert GetInputVarPtrs and GetOutputVarPtrs
6 years ago
Xin Pan 52d3903a12 fix
6 years ago
Xin Pan 0e0983cc1d convert more infer shape
6 years ago
Xin Pan 62eb43ba98 convert more
6 years ago
Xin Pan dfcf746ea1
Merge pull request #14904 from panyx0718/clean2
6 years ago
Qiao Longfei 3f3a84b6dc Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into multithread-sparse-adam
6 years ago
sneaxiy a500dfa579 rewrite ddim
6 years ago
Zeng Jinle 16c244bc3f
Merge pull request #14928 from sneaxiy/fix_gc
6 years ago
Xin Pan f897bd16c0 clean
6 years ago
Xin Pan 70981f5d79 clean
6 years ago
colourful-tree 44ad2f4479
Merge pull request #14873 from colourful-tree/develop
6 years ago
minqiyang 69642000dc Hide KeyHasher
6 years ago
Zhaolong Xing a9fb34fad8
Merge pull request #14903 from NHZlX/add_conv_elementwise_pass
6 years ago
dzhwinter 7cd24b1318
add ir memory optimize. (#14530)
6 years ago
Xin Pan fb8ae30331 fix
6 years ago
guru4elephant a79a3ea2f0
Merge branch 'develop' into develop
6 years ago
wopeizl 0f085f0a5a
Merge pull request #14892 from wopeizl/windows/port3
6 years ago
Yancey1989 06936a2ff5 fix 1gpu test=develop
6 years ago
sneaxiy c631412eab fix gc bug
6 years ago
Xin Pan eaf8ba35b5 change input
6 years ago
Xin Pan 840e6729e2 inject context
6 years ago
Xin Pan bbff0df320 try cache variables
6 years ago
Xin Pan 52bc4ee75a delay infer scope
6 years ago
Yancey1989 d3a4da5cf6 fix comment test=develop
6 years ago
Yancey1989 49870f507d delete unused code test=develop
6 years ago
Qiao Longfei 3bd54ed769 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into multithread-sparse-adam
6 years ago
minqiyang 27a0d6c2dc Polish code
6 years ago
minqiyang aa41ee75a1 Accelerate PADDLE_ENFORCE
6 years ago
nhzlx fcc93d96d5 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_conv_elementwise_pass
6 years ago
Yancey1989 a7d6b1f921 code cleanup test=develop
6 years ago
minqiyang 728e7e88fb Use xxHash as scope's hash algorithm
6 years ago
Yancey1989 a760a550b0 Merge branch 'develop' of github.com:PaddlePaddle/Paddle into parallel_graph_mode
6 years ago
Yancey1989 fd144954ed redefine api test=develop
6 years ago
minqiyang 81651fca45 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into accelerate_ddpg
6 years ago
Yu Yang bacf1d2399 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/tensor_type
6 years ago
Yu Yang e439257ef7 Fix include style
6 years ago
nhzlx c0c9fcd9c7 add source file
6 years ago
dongdaxiang 4c0a769d1d avoid clock time in WIN32 mode
6 years ago
dongdaxiang 66522046ad remove clock time in WIN32 mode
6 years ago
dongdaxiang f2b92d77b5 remove clock time in WIN32 mode
6 years ago
nhzlx 4e4a777243 add conv+elementwiseadd pass
6 years ago
gongweibao 0b1c7d838c
Add brpc serialization support. (#11430)
6 years ago
Yan Chunwei a985949be9
Fea/fuse conv elementwise add fuse (#14669)
6 years ago
Yancey1989 4a4ccac1d0 update by comment test=develop
6 years ago
Yu Yang 04a570b463 Fix ut
6 years ago
heqiaozhi 09d669ba40 fix static_cast to const_cast
6 years ago
peizhilin 23dec78772 fix script issue
6 years ago
heqiaozhi bd1c1724aa add ps_instance doc
6 years ago
Yu Yang 62b5772ffb Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/tensor_type
6 years ago
Qiao Longfei fac8702269 adam support multithread
6 years ago
Wang Guibao e213050223
Fix multi-threading bug with WItH_MKL=ON (#14882)
6 years ago
Yancey1989 4f304eaa6f fix unittest test=develop
6 years ago
dongdaxiang d839bd0dd4 simple commit
6 years ago
minqiyang 3bb9b0cac2 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into accelerate_ddpg
6 years ago
minqiyang a81495d6f4 Fix code
6 years ago
minqiyang ad6ae0b071 1. Add SpinLock
6 years ago
Yancey1989 c722b1dcb6 Merge branch 'develop' of github.com:PaddlePaddle/Paddle into parallel_graph_mode
6 years ago
Yu Yang 4ecdb6f486 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/tensor_type
6 years ago
sneaxiy f6741df462 merge develop
6 years ago
Zeng Jinle 1b564bc49a
Merge pull request #14670 from sneaxiy/refactor_eager_deletion
6 years ago
dongdaxiang c9b799896e fix tag in async_executor
6 years ago
heqiaozhi f81957a753 refine cmake for pslib & pre_define
6 years ago
sneaxiy 2328bee1cc fix Windows compile bug
6 years ago
Yancey1989 23eb8c4299 fix ci test=develop
6 years ago
heqiaozhi 2912d5311b fix code style bug & change pslib.cmake & change Cmakelist adapt pslib
6 years ago
dongdaxiang c59cdf3a24 refine executor_thread_worker.h and executor_thread_worker.cc code style
6 years ago
heqiaozhi 3c01cdeff0 refine executor_thread_worker.cc & executor_thread_worker.h code style
6 years ago
dongdaxiang c71279bc69 refine code style for async_executor.h and async_executor.cc
6 years ago
Yu Yang 162637b64a Fix ngraph compile
6 years ago
dongdaxiang 33ee5cad61 format code style of executor_thread_worker.cc
6 years ago
Yu Yang 194e66f785 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/tensor_type
6 years ago
heqiaozhi 7bd16e3afa fix some bug & add log
6 years ago
Xin Pan 30aad88449
Merge pull request #14839 from panyx0718/infer3
6 years ago
sneaxiy db2daefe50 merge develop
6 years ago
Yancey1989 106e285236 add unittest for parllelgraph mode test=develop
6 years ago
minqiyang a61eb543f5 Add RWLock to Scope
6 years ago
sneaxiy e82772f425 fix cmake conflict
6 years ago
sneaxiy 0c554a59fa merge develop
6 years ago
minqiyang 1b61021cb3 Polish code
6 years ago
minqiyang b75bd29c3a Remove debug info
6 years ago
minqiyang 7a43e51703 Add gperf tools
6 years ago
Yu Yang c00e07cda0 Fix distribute compile
6 years ago
tensor-tang d846fb03a8
Merge pull request #14807 from baojun-nervana/ngraph_follow_up
6 years ago
sneaxiy ca84c2ca8f merge develop
6 years ago
sneaxiy e240ba2918 implement backward
6 years ago
sneaxiy 06f8aa5b97 remove while_op support temporarily
6 years ago
Yu Yang 81520a24cf Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/refine_eigen_tensor
6 years ago
Yu Yang 9bd70a1e04 Change tensor uses proto::VarType::type
6 years ago
Yu Yang 8175983ef9
Merge pull request #14814 from reyoung/feature/gprof
6 years ago
heqiaozhi 10ed9e0a6e download & run & instance
6 years ago
Yancey1989 5cc83f79bf update by comment
6 years ago
Yancey1989 82726402be exception safe
6 years ago
minqiyang 570338699b Add debug info
6 years ago
Yu Yang 9f8d86858e Revert data_type
6 years ago
sneaxiy 7923042365 merge develop
6 years ago
Tao Luo 3243b45bb2
Merge pull request #14651 from luotao1/has_attr
6 years ago
Xin Pan 1735022a1b fix clang
6 years ago
heqiaozhi 016a068775 stop server
6 years ago
Tao Luo 067ed70f2d add HasProtoAttr function in op_desc.h, clean node.h
6 years ago
heqiaozhi 8e3fe2d735 add skip op
6 years ago
Yancey1989 79082c9459 fix pyreader failed
6 years ago
minqiyang 202b2f1fa7 Move the beta pow scale calculation into Adam Op
6 years ago
Yu Yang b22d638d8f Speed up SizeOfType
6 years ago
Xin Pan 4a716a70c6
Merge pull request #14811 from panyx0718/revert-14782-revert-14398-imperative
6 years ago
Brian Liu 9623b45f40 Remove unnecessary MKLDNN reorder (#14799)
6 years ago
Yancey1989 2dda19f756 Merge branch 'develop' of github.com:PaddlePaddle/Paddle into parallel_graph_mode
6 years ago
Tao Luo e99597d35c Merge branch 'develop' into luotao1-has_attr
6 years ago
Yu Yang f0c0bf328d Add gperftools supports for PE
6 years ago
Xin Pan 748549b2e3 Revert "Merge pull request #14798 from PaddlePaddle/revert-14786-revert-14782-revert-14398-imperative"
6 years ago
bingyanghuang 943ad4781f One possible solution to add flexibility for mkldnn placement pass (#14768)
6 years ago
baojun-nervana fddbd87c0a Rename argument
6 years ago
baojun-nervana 22ac2133e4 Rename class
6 years ago
baojun-nervana bfde5e10ce Move ngraph compile control to cmake
6 years ago
sneaxiy 2c6159a151 fix unittest
6 years ago
Xin Pan c049fa7cf7
Revert "Revert "Revert "Imperative"""
6 years ago
gongweibao f1fb64b17f
Add reduce sparse tensor feature. (#14757)
6 years ago
sneaxiy eb8252466b polish code
6 years ago
Tao Luo c83d5b7a16
Merge pull request #14709 from yihuaxu/develop_4f71a6ee2_conv3d_bias_fusion_mkldnn_impl
6 years ago
Yancey1989 47740ace28 fix performance
6 years ago
Yancey1989 220db4f334 clean code
6 years ago
Yancey1989 cb8a24be14 clean code
6 years ago
heqiaozhi 575ae7c6c3 refine pslib inferface & fix some bugs
6 years ago
Yihua Xu 3821fc3950 Merge branch 'develop' into develop_4f71a6ee2_conv3d_bias_fusion_mkldnn_impl
6 years ago
Yihua Xu 240d974ac5 Clean Code
6 years ago
Xin Pan 2538ef64f1
Revert "Revert "Imperative""
6 years ago
guru4elephant b82a44ea85
Merge pull request #14778 from wangguibao/async_executor_bugfix
6 years ago
sneaxiy 8095fb5e68 fix code bug in CPU compilation
6 years ago
sneaxiy 387bac46b5 refine code
6 years ago
Tao Luo cf66133857
Merge pull request #14734 from luotao1/memory_load
6 years ago
Xin Pan 6217f42ab7
Revert "Imperative"
6 years ago
Tao Luo 743cb840f1 update with comments
6 years ago
Yancey1989 c9de6f1b05 init parallel graph mode
6 years ago
wangguibao 5a2cd4505b AsyncExecutor bugfix: Tensor to LoDTensor
6 years ago
wangguibao 5f98d80039 AsyncExecutor bugfix: Tensor change to LoDTensor
6 years ago
Tao Luo 923b18877e Merge branch 'develop' into memory_load
6 years ago
Tao Luo 405b2486db support loading from memory
6 years ago
Xin Pan b52f5d2870
Merge pull request #14398 from panyx0718/imperative
6 years ago
heqiaozhi d3ca359e44 config init & adapt to interface
6 years ago
Xin Pan 73b4d1aa72
Merge pull request #14742 from panyx0718/infer2
6 years ago
Xin Pan 82d68281c0 follow comments
6 years ago
Xin Pan 41c28d54c6 allow customize kernel selection
6 years ago
Xin Pan 439af8d50a
Merge pull request #14717 from panyx0718/infer
6 years ago
sneaxiy d0c8b9b9b3 remove timeout unittest
6 years ago
heqiaozhi 419506f510 refine for compile pslib.so
6 years ago
Wu Yi 29d9fb53fc
[Feature] multi process multi gpu dist training, boost v100 performance by 20% (#14661)
6 years ago
sneaxiy e694d0c2e4 fix while_op eager deletion bug
6 years ago
Xin Pan dc458b1482
Merge pull request #14713 from panyx0718/api
6 years ago
heqiaozhi 2301abc481 cc libaray add pslib
6 years ago
sneaxiy 35a2578426 fix bug
6 years ago
zhang wenhui abbe382e1e
Revert "Add EstiminateFlops"
6 years ago
Xin Pan 0591ba96ec fix hack
6 years ago
sneaxiy 64ad051b9a merge develop
6 years ago
sneaxiy c47c451a00 fix bug
6 years ago
heqiaozhi a77fa67bbd async_thread_trainer & libmct & pslib.cmake
6 years ago
Tao Luo 3437e17713 Merge branch 'has_attr' of https://github.com/luotao1/Paddle into luotao1-has_attr
6 years ago
Xin Pan bcf36d8401 add more files to protected file list
6 years ago
Tao Luo 0e3048db43
Merge pull request #14659 from luotao1/update_pass
6 years ago
Xin Pan 7e0801d4ed
Merge pull request #14441 from baojun-nervana/intel/ngraph_op
6 years ago
Xin Pan b80fe8264a polish
6 years ago
Yihua Xu 64e261c6cd Implement the fusion of convolution and bias for mkldnn
6 years ago
Tao Luo 8d6984eb9b change OpHasAttr to RuntimeHasAttr, add some comments
6 years ago
Tao Luo a6ac42669c Merge branch 'develop' into update_pass
6 years ago
Yu Yang 0f0e197914
Merge pull request #14666 from reyoung/feature/estiminate_flops
6 years ago
Xin Pan aeb74af54c allow operator to run imperatively
6 years ago
Xin Pan b1f6fda5e5 run forward
6 years ago
barrierye 08233beed7 add the comment for CheckFile function. test=develop
6 years ago
barrierye d62a3dd72d add the comment for CheckFile function. test=develop
6 years ago