Commit Graph

48 Commits (e402c0ec7d813264d76841fc4972ebc631f7696e)

Author SHA1 Message Date
peizhilin cd562f8fb7 disable the parallel mode for adam op on windows test=develop
6 years ago
Qiao Longfei 8c516a24e5 remote min_row_size_to_use_multithread in adam interface test=develop
6 years ago
Qiao Longfei 9b4fe283e1 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into multithread-sparse-adam
6 years ago
Qiao Longfei 44b300556d change min_row_size_to_use_multithread to parameter of adam
7 years ago
Qiao Longfei 87b4eb1da4 change min_param_size_to_use_multithread to min_row_size_to_use_multithread
7 years ago
Qiao Longfei 17b1b660fc Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into multithread-sparse-adam
7 years ago
Qiao Longfei c15270c5b2 optimize multi thread adam
7 years ago
Qiao Longfei 0e747e8d02 change the limit of thead num
7 years ago
Qiao Longfei dfe85fb358 fix build
7 years ago
Qiao Longfei f057bbd1d1 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix-adam-tmp-var
7 years ago
Qiao Longfei f1c973b014 adam op should not create tmp var in compute
7 years ago
Qiao Longfei d161215332 optimize adam multi thread
7 years ago
Qiao Longfei 7a58ad5c79 lazy mode have higher priority then multithread
7 years ago
Qiao Longfei d0572bf02e add log for lazy mode test=develop
7 years ago
Qiao Longfei 1177b0bc84 update multi thread adam
7 years ago
Qiao Longfei 3b294e2e2e Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into multithread-sparse-adam
7 years ago
sneaxiy dde3afe7b7 Merge develop
7 years ago
Wu Yi 856f0da0fe
Fp16 training (#14992)
7 years ago
minqiyang 8b6b0da062 Use adam_update
7 years ago
minqiyang f4e7a47381 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into accelerate_adam
7 years ago
sneaxiy c1f7e54f62 merge develop
7 years ago
typhoonzero da87f7a698 Revert "[Feature] Fp16 training for resnet50 (#14850)"
7 years ago
Qiao Longfei d76bda50c4 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into multithread-sparse-adam
7 years ago
minqiyang 6467eda534 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into accelerate_adam
7 years ago
Wu Yi 3d750f9c5a
[Feature] Fp16 training for resnet50 (#14850)
7 years ago
sneaxiy 74a8e6b032 merge develop
7 years ago
sneaxiy ae6f46a1a9 rewrite variable type
7 years ago
Qiao Longfei e2d56561e7
Merge pull request #14889 from jacquesqiao/optimize-adam
7 years ago
Qiao Longfei e0df9f2346 merge lazy mode
7 years ago
Qiao Longfei 8936c7913b add log test=develop
7 years ago
Qiao Longfei 59cf96ec18 add log
7 years ago
Qiao Longfei fe3995d335 refine code test=develop
7 years ago
Qiao Longfei fd152289fa clean for range in test=develop
7 years ago
Qiao Longfei 1141db8114 update test_adam_op
7 years ago
Qiao Longfei 3bd54ed769 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into multithread-sparse-adam
7 years ago
Qiao Longfei 96604fda10 fix gpu data
7 years ago
minqiyang 24eb8f038c Fix bug
7 years ago
Qiao Longfei fcde2b2725 add ForRangeIn
7 years ago
minqiyang 5fea8cd478 Add sorted_result parameter to SelectedRows Functor
7 years ago
minqiyang da796dfe05 Remove BinarySearch from Adam Op
7 years ago
Qiao Longfei c624417c6f change sparse mode to lazy mode
7 years ago
Qiao Longfei fac8702269 adam support multithread
7 years ago
Qiao Longfei 3dc29b3905 change sparse_update to adam_update
7 years ago
Qiao Longfei fc6ec6bd14 add sparse mode adam
7 years ago
Yu Yang 9bd70a1e04 Change tensor uses proto::VarType::type
7 years ago
Yu Yang 7604b1ad51 Fix Eigen macro when using GPU
7 years ago
minqiyang 53433d7f2e Revert the changes of VLOG
7 years ago
Wu Yi a2d9b34417
Refine operator cmake (#14413)
7 years ago