128 Commits (b6a26749dc1747d2378e4976366d18268841b74c)

Author	SHA1	Message	Date
tangwei12	bc5f0246a8	large scale kv speedup (#26510) ... * rename communicator meet->BatchesCounter * fix parame recv for sparse * geo sparse init from pserver * optimize init from pserver * add large scale optimizer fuse(SGD/ADAM) * rectification init_worker and exe.run startup program	5 years ago
danleifeng	0721767ba9	fix server_num bug;test=develop (#27442)	5 years ago
danleifeng	905e2346ac	add endpoints log;test=develop (#27439)	5 years ago
danleifeng	fc61efd736	fix port env bug(int);test=develop (#27405)	5 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	5 years ago
tangwei12	99626502f7	【paddle.fleet】gloo and util (#27213) ... * fix worker endpoints * fix gloo wrapper for hdfs * GPU fleetrun support gloo * parameterserver fleetrun support gloo * fix get server endpoint	5 years ago
123malin	f36b9a7f79	【Fleet2.0 Util】 add documents (#26698) ... * test=develop, util documents	5 years ago
danleifeng	8d05c00c67	fix paddle.fleet en-doc for apis in dynamic mode (#27354) ... * fix fleet dynamic-mode en-doc;test=develop	5 years ago
ShenLiang	746a8ded29	fix comment of adaptive lsgd (#27362)	5 years ago
gongweibao	11bcf0e21c	Cleanup redundant code files (#27319)	5 years ago
ShenLiang	54b81fa32c	add adaptivelsgd in meta_optimizer (#27289) ... * add adaptivelsgd * Todo fix the code to avoid the conflict.	5 years ago
yaoxuefeng	c67c391682	refine fleet dataset class api (#27133)	5 years ago
danleifeng	389a9a7e0e	fix ports conflict when use paddlecloud to launch analogue multi-nodes (#26191) ... * fix ports conflict when launching multi-nodes in paddlecloud;test=develop * add DISTRIBUTED_TRAINER_ENDPOINTS env for cloud;test=develop	5 years ago
mapingshuo	9dedafa0df	fix strategy, test=develop (#27323) ... * fix strategy, test=develop * fix can_apply	5 years ago
ShenLiang	2b6a5793fe	remove auto mode from localsgd optimizer (#27237) ... * rm auto from localsgd	5 years ago
123malin	60c3ef3ab8	【paddle.fleet】parameter_server_optimizer support auto_strategy (#27181) ... * parameter_server_optimizer support auto_strategy	5 years ago
JZ-LIANG	5d039f4086	modified the implement of Lars optimizer (#26733) ... add lars to fleet meta optimizer	5 years ago
Dong Daxiang	f7d08b7db8	【paddle.fleet】refine launch and distributed repr string for print (#27093) ... * refine launch and distributed repr string for print	5 years ago
123malin	f2d68d3ed5	【paddle.fleet】parameter_server_optimizer support auto_strategy (#26838) ... * test=develop, add ps auto	5 years ago
ShenLiang	aca450f6fb	fix the localsgd optimizer (#27094) ... * fix the localsgd	5 years ago
Dong Daxiang	0443b480b8	【paddle.fleet】add auto parallel L1 implementations (#27090) ... * add auto parallel L1 implementation test=develop	5 years ago
Chengmo	a72752263b	support heter-xpu-ps (#27018) ... support heter-xpu-ps	5 years ago
mapingshuo	9e4fe92303	fix strategy example (#26856) ... * fix doc * test=develop * test=develop * test=develop * test=develop * test=develop * test=develop * test=develop * test=develop * fix doc, test=develop * update localsgd doc test=develop * update localsgd doc test=develop * fix fleet dgc amp doc, test=develop * fix, test=develop * fix async configs Co-authored-by: liuyi05 <gavin1332@gmail.com> Co-authored-by: WangXi <wangxi16@baidu.com> Co-authored-by: seiriosPlus <tangwei12@baidu.com>	5 years ago
danleifeng	6b4ca0d7f1	【paddle.fleet】distributed_optimizer supports dygraph (#26541) ... paddle.distributed.fleet supports dynamic graph execution.	5 years ago
danleifeng	3a2a711681	【paddle.fleet】simplify fleetrun log infos (#26888) ... * print detailed and clear log infos; test=develop	5 years ago
danleifeng	e35ad3eee8	【paddle.fleet】support running python train.py for fleet tasks (#26249) ... * support running python train.py for fleet-task; test=develop	5 years ago
lilong12	030b298e82	fix sample codes in collective.py (#26787) ... * fix sample codes, test=develop	5 years ago
Chengmo	d0962abd20	supplement bug fix of parameter server (#26217) ... * fix fluid.embedding	5 years ago
Chen Weihang	28cb653145	Remove backend argument of init_parallel_env (#26773) ... * remove backend argument of init_parallel_env * remove keep name table in transformer * add cpu version check * add skip unittest for init_parallel_env * polish doc: remove func use & update example	5 years ago
tangwei12	9ded7565ec	【paddle.fleet】FleetAPI 2.0 (#26772) ... * add FleetAPI doc Co-authored-by: liuyuhui <liuyuhui@baidu.com>	5 years ago
Chengmo	7f2aa2db3c	【paddle.fleet】Support Heter Parameter Server (#25998) ... * Support Heter Parameter Server	5 years ago
Dong Daxiang	994217ea05	【paddle.fleet】fix api documents (#26777) ... * fix api document	5 years ago
Chen Weihang	31f422ae5e	Add interface to launch parallel dygraph by multiprocessing (#26044) ... * add dygraph parallel run interface * polish implement & unified env property name * add print config arg * refactor init_parallel_env function * Compatible with multiprocessing and launch modes * set default trainer start port * support run in python 2 * polish python2 support code * remove python2 support * refine launch import * polish dome design details * refactor api implemention & path * use new method _set_expected_place * add spawn unittest framework & mnist test * add more unittests & doc * fix unittest failed * polish english doc * self review and polish details * refactor code by reviewer's comments * fix unittest failed * fix parallel_env unittest * fix several typos * fix error introduced when fixing typos * add unpublic note for start_processes * polish details by xiaoguang's comment * verify correctly when spawn nprocs=-1 * refactor spawn & init_parallel_env design * polish doc details * open spawn unittests * try to fix doc compile error * try to fix unknown doc format error * add skip unittest when not gpu	5 years ago
Yi Liu	2024ef69a2	【paddle.fleet】add comments about localsgd in DistributedStrategy (#26709) ... * modified timeout value on windows and mac (#26690) * add Local SGD algorithm referenced paper test=develop	5 years ago
lilong12	1c68138327	[api 2.0] add collective op for cpu using gloo and paddle.distributed.* apis (#26552) ... add collective op for cpu using gloo and paddle.distributed.* apis	5 years ago
JZ-LIANG	958d7212c7	【paddle.fleet】Document refine lars & lamb (#26533)	5 years ago
Dong Daxiang	08d736ad78	【paddle.fleet】add cudnn related strategies to DistributedStrategy (#26598) ... * add cudnn related strategies to DistributedStrategy	5 years ago
WangXi	7ff197d3ba	Add fleet dgc amp doc, test=document_fix (#26608)	5 years ago
liuyuhui	66596bd20f	【paddle.fleet】solve the initial configuration about fleet and rolemaker (#26368) ... * solve the initial configuration about fleet and rolemaker Co-authored-by: seiriosPlus <tangwei12@baidu.com>	5 years ago
Dong Daxiang	7d3e46e1d5	【paddle.fleet】Document refine (#26526) ... * add documentation for DistributedStrategy	5 years ago
Dong Daxiang	83cd185947	【paddle.fleet】Meta from optimizer (#26392) ... * consider the combination of different strategies to work together	5 years ago
123malin	57d434df5d	add save/load for parameter server (#26235) ... * add save/load for parameter server	5 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	5 years ago
Dong Daxiang	4ec51e0205	【paddle.fleet】Clear disable (#26334) ... * add check approval test=develop	5 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	5 years ago
Qinghe JING	d549a9b1fe	【paddle.fleet】Set default value to strategy in distributed_optimizer (#26246) ... * set default value to strategy in distributed_optimizer test=develop	5 years ago
liuyuhui	935da32d25	【paddle.fleet】upgrade fleet: modify role_maker (#26038) ... * add unittest for paddlerolemaker with gloo	5 years ago
Dong Daxiang	50a5bcfc9d	【paddle.fleet】paddle.fleet -> paddle.distributed.fleet. (#26186) ... * move paddle.fleet to paddle.distributed.fleet	5 years ago
Yi Liu	f45f8363eb	records the offset of log when creating by paddle.distributed.launch (#25725) ... test=develop	5 years ago
gongweibao	80f1c50738	Fix typo in interface. (#24779)	5 years ago
Yi Liu	b0f18947f3	fix the compatibility of PY2 and PY3 in paddle.distributed.launch (#25304) ... * fix the compatibility of PY2 and PY3 in paddle.distributed.launch test=develop * only pull log of local rank 0 test=develop * log exception if UnicodeEncodeError occurs when pulling log in paddle.distributed.launch test=develop Co-authored-by: SunGaofeng <peakbee@gmail.com>	5 years ago
Yi Liu	5209b9a510	cat log to stdout when setting log_dir in launch (#25147) ... test=develop	5 years ago
mapingshuo	9388a6381c	fix popen error (#24767) ... replace join to terminate	5 years ago
zhangchunle	f62dfc6238	fs_wrapper add __all__ (#24335)	5 years ago
Kaipeng Deng	80cf3c3c4d	Refine DataLoader support multi-processing (#23107) ... * add DataLoader, Dataset, BatchSampler	5 years ago
gongweibao	63bfe0b946	Fix the default value bug of started port in launch.py. (#23531)	5 years ago
gongweibao	24a063f6ac	Add fleet checkpoint on local fs and remote fs(such as hdfs) for EDL (#22586)	5 years ago
gongweibao	4b40edf359	Use available ports instead of static ports. (#22553)	5 years ago
tianshuo78520a	d2ba91aad1	fix typo words (#22653)	5 years ago
gongweibao	ad2bc0c364	Fix a distribution bug and cleanup some not need logs. (#22381)	6 years ago
danleifeng	f5262865c0	change select_gpus into absolute values in launch.py (#22031)	6 years ago
danleifeng	3fe63d6780	add store_true to use_paddlecloud argument in launch.py (#21168)	6 years ago
Dong Daxiang	a6747a6ef1	add launch_ps module so that we can launch a parameter server trainin… (#20936) ... * add launch_ps module so that we can launch a parameter server training job 1) a user can specify worker_num and server_num 2) parameter server can be killed after all workers exit 3) unit test is added test=develop	6 years ago
WangXi	9d8ec42353	launch.py remove setting for nccl sync, test=develop (#20909)	6 years ago
WangXi	e78d7f57bb	Print the rank which trainer is error in launch.py, test=develop (#20838)	6 years ago
WangXi	8c2c8dc626	distribute.launch use poll to query subprocess (#19853) ... distribute.launch use poll to query subprocess	6 years ago
danleifeng	0865b5a9a0	distribute launch : add use_paddlecloud argument (#19273) ... distribute launch : add use_paddlecloud argument	6 years ago
gongweibao	86f0591175	Remove node_num function. (#19167) ... node_num is not needed for users, so remove them and fix the bugs about it!	6 years ago
guru4elephant	30562e371b	refine launch_ps and role_maker (#18795) ... refine launch_ps and role_maker	6 years ago
guru4elephant	70b03760fd	add parameter server launch (#18687) ... add parameter server launch so that a user can easily launch parameter server	6 years ago
gongweibao	c0a82748cf	Polish backwards optimizer dependency codes and use more default values. (#18255)	6 years ago
gongweibao	da9143c1cc	Polish codes of old prs. (#17938)	6 years ago
gongweibao	f3e5a5cf67	Unset https_proxy and http_proxy in our launch.py (#17915)	6 years ago
gongweibao	6a1df46991	Fine tuning launch.py (#17223)	6 years ago
chengduo	ca03f4989a	fix distributed launch.py (#17571) ... test=develop	6 years ago
Yan Xu	266444b8af	fix dist launch script test=develop (#17404)	6 years ago
Yan Xu	b4c3a6aa0b	[Imperative] implement imperative NCCLParallelContext (#16477) ... add NCCLParallelContext for parallel dygraph	6 years ago
Yan Xu	d424e5b4c9	add launch mp distributed job py module test=develop (#15620) ... * add launch mp distributed mode module test=develop * delete unused file test=develop * refine usage test=develop * refine usage test=develop * move distributed package test=develop * add to whl package test=develop	6 years ago