Paddle

Commit Graph

Author	SHA1	Message	Date
lilong12	0205e9f84e	remove the send/recv of tensor size (#31460 ) * remove the send/recv of tensor size, but users have to specify the shape of the received var explicitly.	5 years ago
liuyuhui	4a8b8b4547	[Kunlun] add gen_bkcl_id_op, support multi XPU cards training using multiprocess (#30858 )	5 years ago
WangXi	31ed9c9eed	Fleet distributed strategy support pure fp16 (#30754 )	5 years ago
lilong12	8126a41d73	fix the bug of all_reduce pipeline gradient multiple times (#30437 ) * update, test=develop	5 years ago
hutuxian	9fec1618d2	Ascend Framework Part3: Ascend Parser (#30391 )	5 years ago
JZ-LIANG	75936d838f	Recompute Offload (#30233 )	5 years ago
Chengmo	d479ae1725	【Paddle.Fleet】Support local save sparse param (#30175 ) * add save tensor support Co-authored-by: seiriosPlus <tangwei12@baidu.com>	5 years ago
Chengmo	528e03fc08	【Paddle.Fleet】Fix tensor table (#30075 ) * add tensor table	5 years ago
WangXi	ab04997846	[fleet] combine amp and gradient merge, test=develop (#30086 )	5 years ago
lilong12	01950ceb42	fix the bug in pipeline data parallelism (#29731 ) * update, test=develop	5 years ago
tangwei12	032414ca2a	[Feature] one ps (3/4) (#29604 ) * oneps (3/4) Co-authored-by: MrChengmo <cmchengmo@163.com> Co-authored-by: malin10 <malin10@baidu.com> Co-authored-by: chengmo <chengmo@baidu.com>	5 years ago
WangXi	9cbcc6cadc	fleet sync build strategy, test=develop (#29732 )	5 years ago
JZ-LIANG	d33d468f02	[Sharding] add hybrid-dp feature (#29518 ) * Sharding add hybrid-dp feature * update sharding in distributed_strategy * update sharding unitest * revise code format for sharding	5 years ago
WangXi	0c2a51d240	optimizer amp, all use fp16 communication, overlap last comm and compute (#28957 )	5 years ago
JZ-LIANG	0dadacc4eb	[sharding] doc, api, bug fixed (#28983 ) * add lars to fleet meta optimizer * add lamb to proto * add lamb to fleet meta optimizer * fixed syntax bug * fixed syntax bug * fixed syntax error in lamb, add config setter of lamb in distributed_strategy * trigger unitest to rerun * add new unitest func for lamb * revise unitest for lars and lamb * revise dgc meta unitest * revise lars document in distribute_strategy * revise lars lamb document in distributed_strategy.py * revise lars lamb document in distributed_strategy.py * add weight decay exclude logic to lars * restore optimzier.py * restore optimizer.py as develop except lars * add epsilon and exclude fn to distributed_sttrategy * add lars epsilon * revise unitest for fleet lars and lamb * revise lars lamb unitest for CI coverage * revise lars argument api * revise lars argument api * revise lars argument api * revise api doc of lars * fix op role * add sharding save and add_sync_comm_for_test function * add comm_analyse to utlis * revise sharding_utils * add sharding saving unittest * revise sharding utils for unittest * revise sharding en doc * update sharding utils api * add doc for sharding * fixed bug in sharding var size count * update varsize count in sharding * fix sharding num_nccl_comm * Revert "fix sharding num_nccl_comm" This reverts commit d51587c15e9323acf226ddd36154275f0d1daf76.	5 years ago
WangXi	e931c7baf9	Fix multi nccl comm & wait server ready (#28663 )	5 years ago
Leo Chen	3815d7aa40	Upgrade string literals to raw string (#28989 ) * upgrade comment string to raw string * fix string in * fix string with ' ' * revert update on comments * upgrade only necessary * fix sample code checker * fix comments with '''	5 years ago
lilong12	f77a78cdee	enable pipeline to run with Executor.run() (#28373 ) * update, test=develop	5 years ago
JZ-LIANG	5a9f6889c1	[Sharding] add new features (#28568 ) * add lars to fleet meta optimizer * add lamb to proto * add lamb to fleet meta optimizer * fixed syntax bug * fixed syntax bug * fixed syntax error in lamb, add config setter of lamb in distributed_strategy * trigger unitest to rerun * add new unitest func for lamb * revise unitest for lars and lamb * revise dgc meta unitest * revise lars document in distribute_strategy * revise lars lamb document in distributed_strategy.py * revise lars lamb document in distributed_strategy.py * add weight decay exclude logic to lars * restore optimzier.py * restore optimizer.py as develop except lars * add epsilon and exclude fn to distributed_sttrategy * add lars epsilon * revise unitest for fleet lars and lamb * revise lars lamb unitest for CI coverage * revise lars argument api * revise lars argument api * revise lars argument api * revise api doc of lars * fix op role * add sharding save and add_sync_comm_for_test function * add comm_analyse to utlis * revise sharding_utils * add sharding saving unittest * revise sharding utils for unittest	5 years ago
Chengmo	4dc8c44ba1	【Paddle.Fleet】Fix fleetrun heter (#28252 ) * fix fleetrun heter ps on paddlecloud	5 years ago
mapingshuo	81244fbfab	add sharding strategy in fleet(#27900 ) * add sharding	6 years ago
WangXi	fb641c915e	【paddle.fleet】fleet add _get_applied_meta_list and _get_applied_graph_list (#27952 )	6 years ago
Chengmo	328cb289ed	【paddle.fleet】fix sparse load (#27680 ) * add sparse tensor load method	6 years ago
mapingshuo	8d2cb14f98	support gradient merge with recompute, test=develop (#27834 ) * support gradient merge with recompute, test=develop test=develop	6 years ago
Chengmo	c5f2802d56	【paddle.fleet】Update fleetrun & ps-heter (#27472 ) * refine fleetrun.ps_launch * update fleet run for multi device support * ps_graph support ps-gpu * fix heter save * add heter save unittest * fix unittest & simple code * update fleetrun * fix fleetrun * fix launch barrier * fix role maker * add paddlecloud rolemaker unittest * rename heter_worker_device_guard	6 years ago
WangXi	0a1862d1d2	fleet combine amp dgc recompute meta optimizer (#27643 )	6 years ago
WangXi	e550fc02ae	fleet2.0 add fp16 grad compression (#27480 )	6 years ago
tangwei12	d6b54de467	【paddle.fleet】Fix/role maker api fix (#27326 ) * fix fleet util and gloo * fix worker endpoints * fix * fix UT * fix gloo * fix gloo * update gloo * update gloo * update gloo * update gloo * update gloo * fix gloo wrapper for hdfs * add file gloo and UT * fix UT * fix UT * fix UT * hide public method of RoleMaker * fix UT * GPU fleetrun support gloo * parameterserver fleetrun support gloo * add UT * add UT * fix UT * fix get server endpoint * fix get server endpoint * fix UT * hide public method of rolemaker * hide public method of rolemaker * hide public method of rolemaker * Update test_fleet_rolemaker_new.py * hide public method of rolemaker * hide public method of rolemaker	6 years ago
ShenLiang	746a8ded29	fix comment of adaptive lsgd (#27362 )	6 years ago
ShenLiang	54b81fa32c	add adaptivelsgd in meta_optimizer (#27289 ) * add adaptivelsgd * Todo fix the code to avoid the conflict.	6 years ago
mapingshuo	9dedafa0df	fix strategy, test=develop (#27323 ) * fix strategy, test=develop * fix can_apply	6 years ago
ShenLiang	2b6a5793fe	remove auto mode from localsgd optimizer (#27237 ) * rm auto from localsgd	6 years ago
123malin	60c3ef3ab8	【paddle.fleet】parameter_server_optimizer support auto_strategy (#27181 ) * parameter_server_optimizer support auto_strategy	6 years ago
JZ-LIANG	5d039f4086	modified the implement of Lars optimizer (#26733 ) add lars to fleet meta optimizer	6 years ago
123malin	f2d68d3ed5	【paddle.fleet】parameter_server_optimizer support auto_strategy (#26838 ) * test=develop, add ps auto	6 years ago
ShenLiang	aca450f6fb	fix the localsgd optimizer (#27094 ) * fix the localsgd	6 years ago
Dong Daxiang	0443b480b8	【paddle.fleet】add auto parallel L1 implementations (#27090 ) * add auto parallel L1 implementation test=develop	6 years ago
Chengmo	7f2aa2db3c	【paddle.fleet】Support Heter Parameter Server (#25998 ) * Support Heter Parameter Server	6 years ago
Dong Daxiang	994217ea05	【paddle.fleet】fix api documents (#26777 ) * fix api document	6 years ago
Dong Daxiang	83cd185947	【paddle.fleet】Meta from optimizer (#26392 ) * consider the combination of different strategies to work together	6 years ago
mapingshuo	cd48bdad31	add feature to fleet2.0 role_maker, distribute_strategy, test=develop (#26267 ) * add feature to fleet2.0 role_maker, distribute_strategy, test=develop	6 years ago
Dong Daxiang	4ec51e0205	【paddle.fleet】Clear disable (#26334 ) * add check approval test=develop	6 years ago
Yi Liu	3b2c580a66	【paddle.fleet】make fleet_localsgd_meta_optimizer work (#26213 ) * make fleet_localsgd_meta_optimizer work * fix bug in localsgd meta optimizer	6 years ago
Dong Daxiang	50a5bcfc9d	【paddle.fleet】paddle.fleet -> paddle.distributed.fleet. (#26186 ) * move paddle.fleet to paddle.distributed.fleet	6 years ago

44 Commits (test_benchmark_ci)