Commit Graph

248 Commits (f8ed2c229e01863798543ba3cde922726e08696d)

Author SHA1 Message Date
Xin Pan f117feab0c modify comments
7 years ago
Xin Pan b43e49fa31 fix
7 years ago
Xin Pan 36c2a9af27 pass builder allow cutomize pass in python.
7 years ago
chengduo d402234ba8
Feature/op_fuse_pass (#12440)
7 years ago
Xin Pan 5ce77889f4 clean unused inference_optimize c++ implementation
7 years ago
Yancey1989 5ce1a960a5 move bcast op into pass
7 years ago
Xin Pan 0904f07d46 polish
7 years ago
Xin Pan 56a977d436 add test
7 years ago
Xin Pan 926e1077ca version
7 years ago
qingqing01 1f09bc320c
Support data type int8_t . (#12841)
7 years ago
Yu Yang c6af7201e9
Merge pull request #12692 from reyoung/feature/fast_executor
7 years ago
Qiao Longfei 653fad08f8
Optimize selected rows for dist lookup table with pthread rwlock (#12635)
7 years ago
yuyang18 d1d825ee02 Hide unnecessary API
7 years ago
minqiyang 1800fef142 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into port_pybind11
7 years ago
yuyang18 05cadf1b24 Add FastExecutor
7 years ago
Wu Yi 8b77448d5f
hide misc APIs (#12540)
7 years ago
minqiyang 6abe819f07 Fix pybind11 problem
7 years ago
minqiyang 559d36328c Apply 2to3 to current paddle main python code
7 years ago
Yancey1989 d4f51218ef Merge branch 'develop' of github.com:PaddlePaddle/Paddle into rpc_complete_interface
7 years ago
Yancey1989 efd5a84986 update executor interface
7 years ago
Yancey1989 7570d8e77c add rpc complete interface
7 years ago
Wei Xu 264e8305b0 Fixed unittests for WITH_GPU=OFF and WITH_DISTRIBUTE=OFF build
7 years ago
Yancey 0042ba93c8
Merge pull request #12127 from Yancey1989/enforce_rpc_timeout
7 years ago
yuyang18 4ff1bde5fb
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/hide_api_cont
7 years ago
Yu Yang ebe3b5e78a
Merge pull request #11853 from sneaxiy/complete_py_reader_python
7 years ago
Yancey1989 d14afcedeb polish function name
7 years ago
Yancey1989 4a91a14549 enforce rpc client timeout
7 years ago
Yancey1989 9de096bd58 Merge branch 'develop' of github.com:PaddlePaddle/Paddle into fix_cpu_pe
7 years ago
Yancey1989 1effba3312 fix pe with cpu place
7 years ago
sneaxiy e40905daed Merge branch 'complete_py_reader_python' of https://github.com/sneaxiy/Paddle into complete_py_reader_python
7 years ago
Yancey1989 bbd53131f0 update
7 years ago
sneaxiy df7ecbf5a1 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into complete_py_reader_python
7 years ago
yuyang18 b83ffda4aa
Try hide APIs
7 years ago
fengjiayi 5528f59900 Split ReInit() to Shutdown() and Start()
7 years ago
dzhwinter 4ed0b62476
Move fluid::framework::InitDevices into fluid::platform (#11757)
7 years ago
sneaxiy 739c330914 fix merge conflict
7 years ago
Yancey f7fd711e3f
Merge pull request #11868 from Yancey1989/dist_pass_barrier
7 years ago
sneaxiy 1b59220d50 complete python reader op python side
7 years ago
fengjiayi ff4317cee9 follow comments
7 years ago
Yancey1989 1366832a41 add dist pass barrier
7 years ago
sneaxiy d4d946db5a update blocking queue
7 years ago
sneaxiy 502faf62a9 complete_py_reader_cpp
7 years ago
sneaxiy 697ba4b13d Add Python array reader op
7 years ago
Xin Pan 32bfebfe38 disable the LODTensor warning for now
7 years ago
gongweibao 80f63642e6
Add comments to `set_lod`. (#11588)
7 years ago
Kexin Zhao 417fcf4f43
Modify Pybind LoDTensor API according to length-based LoD (#11106)
7 years ago
Wu Yi 34865f2de3
Trainer send term signal (#11220)
7 years ago
chengduoZH aadaadf735 replace use_event with use_cuda, because use_event means the program running with CUDA, so use_cuda maybe more intuitive.
7 years ago
yuyang18 d9af153232 SSA Graph Builder Factory
7 years ago
yuyang18 86a61c177f Add ScopeBufferedSSAGraphExecutor
7 years ago
Xin Pan 35d5b1b9e2
Merge pull request #11036 from panyx0718/dist_timeline
7 years ago
Xin Pan f14e579cc3 clean up
7 years ago
fengjiayi 9f81f1ce57 Add uint8 support in pybind.cc
7 years ago
yuyang18 7c777dd549 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/exec_strategy
8 years ago
yuyang18 08295f9877 Add build strategy
8 years ago
yuyang18 e5281b3c2d Clean code & add execution strategy
8 years ago
typhoonzero 928418a9ac Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into gen_nccl_id_op
8 years ago
chengduoZH 97cb5479ae change PE strategy
8 years ago
typhoonzero d9320dcd94 complete code
8 years ago
yangyaming 82571deb89 Change `customize_loss_grad` to `use_default_grad_scale`.
8 years ago
Yu Yang 7a395881d4 Add customize_loss_grad option to PE
8 years ago
dzhwinter 1fc9fe773d merge develop
8 years ago
Yiqun Liu 598035f985 Fix a bug in save_inference_model and prune when the program is initailized by load_inference_model (#10011)
8 years ago
dzhwinter 425a1e76a2 Merge remote-tracking branch 'origin/develop' into memory/activation_memory
8 years ago
Yu Yang b4aaa00a8a Polish logic of ParallelExecutor
8 years ago
Yu Yang 2ab12ca248 Add comments and clean code
8 years ago
Yu Yang e6bc7bb09a Tuning code
8 years ago
dzhwinter e54f203c55 "move to a new PR"
8 years ago
Yang Yang(Tony) a64edbf14e
delete backward.cc related code on the python side (#9854)
8 years ago
Yang Yang(Tony) b26f505002
remove net op and cond_op (#9663)
8 years ago
typhoonzero dfc60251bf Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into multigpumultinode
8 years ago
fengjiayi 90084a25d2
Merge pull request #9743 from JiayiFeng/modify_readers_to_fit_parallel_executor
8 years ago
fengjiayi a84b81502c Remove Readers' HasNext()
8 years ago
typhoonzero 0bf799a523 wip testing
8 years ago
Xin Pan 326b434ab5
Merge pull request #9699 from panyx0718/test
8 years ago
qingqing01 2b7e5bd366
Support testing during training by ParallelExecutor. (#9738)
8 years ago
Yi Wang 1543c4cf6a
Fix cpplint errors of paddle/fluid/pybind and add some tests (#9694)
8 years ago
Xin Pan b138d29c38 Avoid init_p2p all the times
8 years ago
lgone2000 517f619501 fix pybind.cc compile error (#9681)
8 years ago
chengduoZH 8e4e155c52 add PyCUDAPinnedTensorSetFromArray
8 years ago
chengduoZH 6af178356b expose CUDAPinnedPlace to Python
8 years ago
Xin Pan b123ce88a1 Add enable/disable for delayed ops
8 years ago
Yu Yang edfd741e3a Add simple python wrapper for ParallelExecutor
8 years ago
Yu Yang d42117e742 Set NumThreads
8 years ago
Yu Yang 9cb8f50302 Complete fetch op
8 years ago
Yu Yang 5e87cd7574 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into cpp_parallel_executor
8 years ago
Kexin Zhao e26f1123da
Add fp16 mul op support and bind paddle fp16 to numpy fp16 (#9017)
8 years ago
Yu Yang baef1124fb ParallelExecutor And dependency engine
8 years ago
Yu Yang fea43077f6 Refine
8 years ago
Yu Yang 72be7a6151 Complete RecordIO reader op
8 years ago
Xin Pan b9ec24c6e9 Extend current profiler for timeline and more features.
8 years ago
Yu Yang eebfb71d3f
Merge pull request #8524 from reyoung/feature/unique_name_guard
8 years ago
Yu Yang e8cb97b8a2 Moving unique_name to python
8 years ago
kavyasrinet 77ee8fb240
Exposing Channel to be used as a Variable and integrating with Fluid (#8486)
8 years ago
qingqing01 24509f4af9 Fix the grammar in copyright. (#8403)
8 years ago
Tao Luo b56f4a4ee2 move code from /paddle/string to /paddle/fluid/string (#8363)
8 years ago
Yi Wang fc374821dd Correct #include path
8 years ago
Yi Wang 90648f336d Move file to fluid/; Edit CMakeLists.txt
8 years ago