53 Commits (3d1afe2e6540315f034fd397c7808de71ba494e1)

Author	SHA1	Message	Date
yuyang18	d1203e3822	Add types	7 years ago
yuyang18	4649f662e7	Follow comments & polish doc	7 years ago
yuyang18	ab210925b8	Add more docs	7 years ago
yuyang18	f97c5d4c47	Trainer documentation	7 years ago
yuyang18	958ab99ef8	Polish Non-Layer API	7 years ago
tangwei12	bf2c53ae0a	Merge branch 'develop' of github.com:PaddlePaddle/Paddle into new_api_about_cpkt	7 years ago
Jeff Wang	637827a5bc	Use for_test=True in the Fluid Trainer to clone the test program (#11323) ... * Use for_test=True in the Fluid Trainer to clone the test program * fix typo * Should do the samething to the inferencer	7 years ago
tangwei12	9e026a93cf	remove chief	7 years ago
tangwei12	7fbddaa64a	bug fix	7 years ago
tangwei12	cb7c1245b3	code optimized	7 years ago
tangwei12	2f44585e83	code optimized	7 years ago
tangwei12	53409a29d8	code optimized	7 years ago
tangwei12	6db240d78b	update trainer about epoch_id and step id	7 years ago
tangwei12	9735f25011	optimized	7 years ago
Siddharth Goyal	a4237171a5	Modify optimizer in new API to support more usecases (#11168) ... * Modify optimizer in new API to support more usecase * Modify CMake to include only modified examples	7 years ago
tangwei12	08e5f0ae48	rename need_load_checkpoint to get_latest_checkpoint_serial	7 years ago
tangwei12	7973d9b4b5	bug fix	7 years ago
tangwei12	46f2688f30	bug fix	7 years ago
tangwei12	bca4da4225	cancle only chief delete files	7 years ago
tangwei12	e44c278e60	bug fix about clean	7 years ago
tangwei12	b44ede8033	bug fix	7 years ago
tangwei12	d712af25dc	add distribute config	7 years ago
tangwei12	0deb6f90ba	annotation optimized and code style optimized	7 years ago
tangwei12	0211c5df0a	bug fix	7 years ago
tangwei12	9086043090	bug fix and optimize	7 years ago
tangwei12	486e1e337d	bug fix and optimize	7 years ago
tangwei12	ad9dfeb018	bug fix and optimize	7 years ago
tangwei12	5eea5db95f	optimized checkpoint and save_model	7 years ago
tangwei12	514b2427ed	add save/load persist_vars_without_grad	7 years ago
tangwei12	dca0b6d9cc	restore param_path	7 years ago
tangwei12	b044724db7	update fluid Train API param_path to checkpoint_config	7 years ago
Qiao Longfei	eb7d87545e	add trainer.stop and fix a bug for train_by_parallel_executor (#10762)	7 years ago
qiaolongfei	e8d24aa144	Inferencer support parallel_executor	7 years ago
yuyang18	f04754886b	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/trainer_by_pe	7 years ago
typhoonzero	a41a94f2ee	support nccl2 dist train in trainer	7 years ago
yuyang18	791af3a088	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/trainer_by_pe	7 years ago
Qiao Longfei	1c4bb5c83d	user need to set feed order for Trainer.train and Trainer.test (#10679)	7 years ago
daminglu	74ca73b80d	Update trainer api (#10674)	7 years ago
yuyang18	c4ad0dd084	Add fetch metrics	7 years ago
yuyang18	2a0205a5d9	Draft for train by parallel executor	7 years ago
Qiao Longfei	2a971f3084	Add inferencer infer (#10445) ... * add Inference.infer * optimize code * update no_test_word2vec_new_api.py * update trainer * split check_and_get_place * use inference_program to save inference model in Trainer * update demo * update save_inference_model * clean code	7 years ago
fengjiayi	ba57348f8f	trainer.test() (#10453) ... * a draft of trainer.test() * polish trainer.test() * polish trainer.test() * update code format * update * polish code * polish code * polish code * Make trainer.test follow the rule of returning [loss, metric, metric, ..]	7 years ago
Jeff Wang	889c919048	Use _prog_and_scope_guard to switch the scope (#10421)	7 years ago
Yancey	5b06944857	fix trainer import error on ce (#10448) ... * fix trainer import error on ce * fix setup.py.in	7 years ago
Yancey	bb3247e339	fix traner.py import error (#10442)	7 years ago
Jeff Wang	bd66eed50a	Trainer save load params (#10386) ... * Load/save the params from the params_path * Switch to use load_persistables and save_persistables * Instaed of setup the executor to run program and scope. Pass the program to the load_persistables	7 years ago
Helin Wang	8ee23da846	Fluid new API: dist train without modifying code ... Works with 1 trainer 1 pserver. 2 trainer 1 pserver will stuck at the end of first step, still investigating. The user only need to set envrionment variables to enable distributed training. run pserver: PADDLE_TRAINING_ROLE=PSERVER PADDLE_PSERVER_IPS=127.0.0.1 PADDLE_TRAINERS=2 PADDLE_CURRENT_IP=127.0.0.1 python no_test_word2vec_new_api.py run trainer: PADDLE_TRAINING_ROLE=TRAINER PADDLE_PSERVER_IPS=127.0.0.1 PADDLE_TRAINERS=2 PADDLE_TRAINER_ID=0 python no_test_word2vec_new_api.py	7 years ago
Helin Wang	a66052c6ff	improve trainer API ... - The trainer and inferencer will load params from disk if param_path argument is not None in their constructor. - Remove params.py, we will expose core.Scope to the user if needed (e.g., for GAN). Currently we will not expose it, unless we clearly know doing so can support GAN. - Add `save_params` to Trainer (a TODO item). - rename "network" to "program"	7 years ago
Yu Yang	1bb579a3f5	A naive trainer implementation	7 years ago
Helin Wang	b5dd215d46	improve comments	7 years ago