204 Commits (5528ec13212a1171c8274769b3b6e1842cc6861f)

Author	SHA1	Message	Date
Luo Tao	761b329793	unify the indentation of license	7 years ago
gongweibao	8d1ad97b3d	Add log to `InitParam` `GetParameter` `SendGrad` and etc. (#5162) ... * add logs and fix a bug * fix break buf * modify path bugs * fix by comments * fix by comments * add batch * add float32tostring * add pb support * moidfy gotpaht * compile ok * add proto * delete not need * add proto * add empty proto * clean not need * clean not need * modify deps * fix by comments and update depend * fix compile error * fix loop bugs	7 years ago
gongweibao	8c9119afcd	add logs and fix a bug (#5074) ... add logs and fix a python path bug	7 years ago
Helin Wang	00e2dcf37a	Fix according to comments	7 years ago
Helin Wang	32c92640f0	Fix pserver checkpoint ... The pserver checkpoint before failed because the MD5 checksum is calculated incorrectly. Now changed to CRC32 checksum.	7 years ago
Helin Wang	fc57c09dc9	add detailed log for the pserver	7 years ago
Helin Wang	60238a1bfb	Go master, pserver, trainer: switch to log15, away from logrus	7 years ago
Helin Wang	f28b4d6805	Fix parameter server checkpoint serialization	7 years ago
武毅	40f3e0c194	fix_fault_tolerant_dist_lock (#4888)	7 years ago
武毅	0c72649afc	Fix gometalinter versioning (#4832) ... * fix gometalinter versioning * stop gometalinter	7 years ago
Helin Wang	05176bd1bb	master server will wait etcd forever	8 years ago
Helin Wang	5270585e10	fix according to comment	8 years ago
Helin Wang	da7a1f2f6c	master client: retry connecting to etcd	8 years ago
武毅	886e66a5ff	golang pserver use OptimizerConfig.proto (#3358) ... * golang pserver optimizer config for user * update * update * update * update * update by comments * fix errors * fix errors	8 years ago
Helin Wang	f64539bef9	use random port for embed etcd to avoid port collision	8 years ago
Helin Wang	b8461c79fc	implement init parameters selection with etcd	8 years ago
helinwang	544c7db780	Merge pull request #3223 from helinwang/master_timeout ... Master persist more states to etcd, schedule pending timeout after lo…	8 years ago
Helin Wang	01a62511b4	add curPass into log, remove JobTasks	8 years ago
Helin Wang	10794cf4de	Master persist more states to etcd, schedule pending timeout after load pending state.	8 years ago
Helin Wang	33fb8d7abf	fix according to comments	8 years ago
Helin Wang	5ce7703ce8	fix test by not triggering save checkpoint when not intended(change save duration to 1 hour)	8 years ago
Helin Wang	2ee418db78	fix pserver save / load checkpoint	8 years ago
Yancey	ec9d4d527e	Add start_record interface (#3128) ... * add start_record interface * call master client in reader * update * add demo code in comments * update comments * delete unittest for recordio reader	8 years ago
Yancey	53ea896996	Add master server unit test (#3086) ... * add master server unit test * fix comments * use t.Log * fix travis can not fetch git repo * fix git repo	8 years ago
Helin Wang	6fab04f4e1	fix vet shadow report	8 years ago
Helin Wang	54eac40f64	fix according to comments	8 years ago
Helin Wang	42fe3e88c7	gracefully shutdown pserver, fix gometalinter errors	8 years ago
Helin Wang	cb5c7526e5	shutdown master server gracefully	8 years ago
武毅	c10121e13c	[Done] Sync master client between passes and fix recordio split (#2948) ... * fix recordio split and task passes * update for pre commit * update * update, still need to sync client wait for pass end. * able to sync passes for task dispatching * update to comment * update * fix yapf check * why local pre-commit fails? version is the same * fix race condition * update * fix race condition * this still have duplicate problem in unit test * update * update * update by comment * update	8 years ago
武毅	39af255959	Fix new optimizer lr (#3074) ... * default learning rate, temperary fix * update	8 years ago
Helin Wang	c67d8276b7	fix according to comments	8 years ago
Helin Wang	3ff0a9fbb1	Implement distributed training save model, improve master.NewClient interface	8 years ago
helinwang	bea4056531	Merge pull request #2997 from helinwang/checkpoint ... do not do log.Errorln when checkpoint is not found (which is normal)	8 years ago
Helin Wang	a46198e5b1	fix client discover pserver context cancelled ... It's already fixed by Wuyi's PR, but his PR may take some time to merge, but I want to get this change in ASAP.	8 years ago
Helin Wang	a7e69d949f	do not do log.Errorln when checkpoint is not found (which is normal)	8 years ago
dongzhihong	e1e7309789	boring copyright	8 years ago
Helin Wang	25e57949cc	add more linters, fix errors found by them.	8 years ago
Helin Wang	5d7bccb2a3	fix golint errors	8 years ago
Helin Wang	2b1cac4113	Handle all unchecked errors ... Unchecked errors could be handled by: cd go; gometalinter --vendor --disable-all --enable errcheck $(glide nv)	8 years ago
Yancey	83f263e6ec	Fix fetch record from master failed (#2848) ... Fix fetch record from master	8 years ago
武毅	23b8346072	Fault tolerant distributed training, just work version, with etcd (#2849) ... * using etcd as fault tolerant training * update * workable version, ft not tested * small fix * update * remove TODO	8 years ago
Helin Wang	9eb9b2c29c	fix race condition in test	8 years ago
Helin Wang	777a5cca91	Client test: concurrently init param. Concurrently send grad and get param	8 years ago
Helin Wang	11660eab0e	Fix optimizer parameter buffer allocation size. ... The buffer allocation size should be number of bytes, not number of floats.	8 years ago
dongzhihong	7cfcda5f4f	"fix checkpoint pointer"	8 years ago
dongzhihong	46c704ecf0	"fix init error"	8 years ago
gangliao	e6e2bf45e5	Merge pull request #2832 from helinwang/go_cmake ... go_binary: remove hardcoded library link path, add pserver client test	8 years ago
Yancey	19bfb8a1f2	PServer recovery from checkpoint (#2741) ... * Server recovery from checkpoint	8 years ago
Helin Wang	b04986da9f	add pserver client test	8 years ago
Helin Wang	2231b92a89	go_binary: remove hardcoded library link path	8 years ago