Commit Graph

516 Commits (cc95a7516cb111f08914e712244258dacbbe8f20)

Author SHA1 Message Date
Zeng Jinle c5eeecca7c
Fix tensor_py.h (#17195)
6 years ago
Zeng Jinle 5dfe2ab9e8
Fix mem leak when converting Tensor to numpy array (#17182)
6 years ago
Yan Xu 0b07eef118
ParallelDyGraph with GPU collective mode (#16827)
6 years ago
guru4elephant 03d469ad98
Merge pull request #17005 from wopeizl/fix_ncclwrapper_win1
6 years ago
liuwei1031 a770ce0615
add doc for memory_optimize, test=develop (#17010)
6 years ago
qingqing01 ea42e431f8
Speed unit testing. (#16978)
6 years ago
wopeizl 51a0243a56 fix nccl wrapper on windows
6 years ago
Zeng Jinle 1202d3fc74
Refine model gpu memory (#16993)
6 years ago
guru4elephant bbc6c5714f
Merge pull request #16887 from guru4elephant/add_nccl_context_pybind
6 years ago
gongweibao cbdb8a17b1
Polish DGC code (#16818)
6 years ago
dongdaxiang 466d177d09 add pybind dependency
6 years ago
dongdaxiang 4aa6f679b5 add pybind dependency
6 years ago
dongdaxiang b091139049 add nccl wrapper for python API
6 years ago
Yiqun Liu 112f16143b
Add an option to enable the cache of expected kernel in train phase. (#16724)
6 years ago
chengduo 55b15db5af
Add unit test for fuse all_reduce ops (#16699)
6 years ago
Yiqun Liu 3fe8cb0dd7
Enable the runtime_context_cache pass in train phase (#16640)
6 years ago
guru4elephant 7d653f0aed
Merge pull request #16652 from xjqbest/dataset_merge_develop
6 years ago
xjqbest 6a57e8075a remove trainer_id in datafeed and dataset
6 years ago
Yan Xu b4c3a6aa0b
[Imperative] implement imperative NCCLParallelContext (#16477)
6 years ago
xjqbest 271b7147cc fix dataset bug
6 years ago
chengduo b75a69bad6
Add Stream for fetch op handle (#16600)
6 years ago
乔龙飞 Qiao Longfei 21622ca30b
Merge pull request #16172 from jacquesqiao/add-async-ssa-graph-executor-communicator
6 years ago
sneaxiy 10249c0b78 Merge develop
6 years ago
Qiao Longfei adf272bcec Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add-async-ssa-graph-executor-communicator
6 years ago
xjqbest 9b84e8e66b fix code style
6 years ago
xjqbest a99c8d0c29 fix client to client communication bug
6 years ago
sneaxiy 33473890f3 Merge develop
6 years ago
dongdaxiang 720647e17f rebase current develop and fix conflict
6 years ago
dongdaxiang 45eb6f0765 run pre-commit check files and fix code style problem
6 years ago
xjqbest e95cafd9a7 fix code style & add dataset testcase
6 years ago
xjqbest be74de2c61 fix code style & fix register bug & add release_memory
6 years ago
xujiaqi01 a5b1a0e12b support multi dataset && add init model && fix bug
6 years ago
dongdaxiang b7a202aa38 add distributed optimizer factory
6 years ago
dongdaxiang f612877797 add incubate for unified API
6 years ago
dongdaxiang 317eb0aad3 add incubate for unified API
6 years ago
xujiaqi01 ecfc7df913 add dataset factory && fix style
6 years ago
xujiaqi01 3cea00bd52 store memory data in Dataset && fix bug
6 years ago
dongdaxiang cc4def6ba5 fix some conflict for compilation
6 years ago
heqiaozhi 9bca1926c1 refactor & fix bug
6 years ago
xjqbest 2e9a836c6f add DataSet and InMemoryDataFeed, support load data into memory and shuffle data
6 years ago
dongdaxiang e36bbcc871 fix some typo and CMakefile.txt
6 years ago
xjqbest 824b84d185 add DataSet and InMemoryDataFeed, support load data into memory and shuffle data
6 years ago
dongdaxiang be757096da add pybind for fleet
6 years ago
Qiao Longfei d8974e6da0 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add-async-ssa-graph-executor-communicator
6 years ago
chengduo 1096746cbf
Fuse Adam And SGD ops (#15933)
6 years ago
sneaxiy 2c836ff914 check default grad maker
6 years ago
Zeng Jinle 69cb9792ea
Merge pull request #16506 from sneaxiy/revert-16424-fix_allocator_bug
6 years ago
chengduo ed61d67c73
Fix the interface of Pass::Apply (#16484)
6 years ago
Zeng Jinle 174d0d0b90 Revert "Fix allocator bug"
6 years ago
gongweibao eb83abeac3
Add DGC(Deep Gradient Compression) interface. (#15841)
6 years ago