368 Commits (develop)

Author	SHA1	Message	Date
gongweibao	65bbf950ee	Add multi-ncclcomm and 2D ncclallreduce support. (#17263)	7 years ago
Michał Gallus	0c39b97b4e	[MKL-DNN] Add Fully Connected Op for inference only(#15226) ... * fuse mul and elementwise add to fc * Reimplement the FC forward operator * Fix FC MKLDNN integration by transposing weights * Add FC MKLDNN Pass test=develop * FC MKLDNN Pass: change memcpy to std::copy * Fix MKLDNN FC handling of mismatch input and weights dims * Lower tolerance for MKL-DNN in resnet50 test test=develop * Adjust FC to support MKLDNN Op placement test=develop * Adjust Placement Op to set use_mkldnn attribute for graph test=develop * MKLDNN FC: fix weights format so that gemm version is called test=develop * FC MKLDNN: Remove tolerance decrease from tester_helper * FC MKL-DNN: Refactor the code, change input reorder to weight reorder * MKL-DNN FC: Introduce operator caching test=develop * FC MKL-DNN: Fix the tensor type in ExpectedKernelType test=develop * FC MKL-DNN: fix style changes test=develop * FC MKL-DNN: fallback to native on non-supported dim sizes test=develop * FC MKLDNN: fix CMake paths test=develop * FC MKLDNN: Refine placement pass graph mkldnn attribute test=develop * Fix Transpiler error for fuse_conv_eltwise test=develop * Fix missing STL includes in files test=develop * FC MKL-DNN: Enable new output size computation Also, refine pass to comply with newest interface. test=develop * FC MKL-DNN: enable only when fc_mkldnn_pass is enabled * FC MKL-DNN: Allow Weights to use oi or io format * FC MKL-DNN: Adjust UT to work with correct dims test=develop * Enable MKL DEBUG for resnet50 analyzer test=develop * FC MKL-DNN: Improve Hashing function test=develop * FC MKL-DNN: Fix shape for fc weights in transpiler * FC MKL-DNN: Update input pointer in re-used fc primitive * Add log for not handling fc fuse for unsupported dims test=develop * FC MKL-DNN: Move transpose from pass to Op Kernel test=develop * FC MKL-DNN: Disable transpose in unit test test=develop * FC MKL-DNN: Remove fc_mkldnn_pass from default list * Correct Flag for fake data analyzer tests test=develop * FC MKL-DNN: Add comment about fc mkldnn pass disablement test=develop * FC MKL-DNN: Disable fc in int8 tests test=develop	7 years ago
Qiao Longfei	92e7d5d7cc	fix distribute doc test=develop (#17318) ... * fix distribute doc	7 years ago
Qiao Longfei	58f7695ab2	Async exe support communicator (#17386) ... Async exe support communicator	7 years ago
liuwei1031	f82e4d75e7	improve the doc of paddle.fluid.memory_optimize, test=develop (#17473) ... * improve the doc of paddle.fluid.memory_optimize, test=develop * fix typo, test=develop	7 years ago
liuwei1031	6a53fa95e7	improve the API Sample of DataFeeder, memory_optimize and release_memory (#17374) ... * improve the API Sample of DataFeeder, memory_optimize and release_memory, test=develop * update API.spec, test=develop, test=document_preview * tweak the code format of feed API, test=develop * update API.spec, test=develop * improve doc for DataFeeder and default_main_program, test=develop	7 years ago
tangwei12	7330cd639c	truncated_gaussian_random supported in distributed training, test=develop (#17091)	7 years ago
tangwei12	1a4a51db2b	Fleet unify distributed training (#16791) ... * implement distributed transpiler with fleet	7 years ago
Qiao Longfei	0608f8ca56	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add-async_sparse_param_update_recorder	7 years ago
Qiao Longfei	d640c6cfa9	fix pylint	7 years ago
Qiao Longfei	542b52fac3	fix trainer_id	7 years ago
Qiao Longfei	de65398cb8	update transpiler and listen and serv op	7 years ago
Qiao Longfei	fab1b54d99	Merge branch 'add-communicator' of ssh://github.com/jacquesqiao/Paddle into add-async-ssa-graph-executor-communicator	7 years ago
Xin Pan	0c277ac6e9	polish ... test=develop	7 years ago
Xin Pan	840cf780e4	add deprecation warning. ... test=develop	7 years ago
Qiao Longfei	b8491bfd4e	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add-communicator	7 years ago
tangwei12	971f3bc9b0	fix params with only 1 dim (#15828) ... * fix params with only 1 dim * test=develop	7 years ago
dzhwinter	84f067be94	update. test=develop ... test=develop	7 years ago
dzhwinter	d453b0dcf7	add details. test=develop	7 years ago
Qiao Longfei	8bda4ab213	parameter recv can run	7 years ago
Qiao Longfei	fbd186bd5d	complete recv op	7 years ago
Qiao Longfei	4356f186b4	complete parameter_send	7 years ago
dzhwinter	9c9ad7d40b	Merge remote-tracking branch 'origin/develop' into feature/ir_inplace_pass ... test=develop	7 years ago
dzhwinter	0a63234c85	follow comments. test=develop	7 years ago
dzhwinter	8b97a3a44f	rerun ci. test=develop	7 years ago
tangwei12	90df7ff378	transpiler.py code clean (#15555) ... * move var strusted to vars_distributed.py, add optimizer's block name, test=develop * rename optimzier's seems complex, revert it, test=develop * replace * with details, test=develop	7 years ago
dzhwinter	a26a6bc728	add flag. test=develop	7 years ago
gongweibao	fe8f28c957	Add GetVariableNoBarrier on brpc. (#15488)	7 years ago
Wu Yi	22db82c053	fix tangwei merge issue test=develop (#15506)	7 years ago
tangwei12	8b50ad80ff	checkpoint at distributed training (#14854) ... checkpoint for distributed training.	7 years ago
Qiao Longfei	810439a993	fix style test=develop	7 years ago
Qiao Longfei	b16e832d4d	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix-dist-sparse-decay	7 years ago
乔龙飞 Qiao Longfei	e1679b8847	Merge pull request #14893 from JiabinYang/feature/add_prefech_hs ... Feature/add prefech hs	7 years ago
Qiao Longfei	49cce3fd0e	fix dist sparse l2 decay ... test=develop	7 years ago
haowang101779990	66ea718452	en api improve format Dec 27 ... test=develop	7 years ago
sneaxiy	3e917a934a	add scope_pool ... add module cleanup test=develop	7 years ago
JiabinYang	3b7b2e1ded	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/add_prefech_hs	7 years ago
JiabinYang	b2f789c66d	add test transpiler dist test, test=develop	7 years ago
Xin Pan	c89a1fb287	Merge pull request #14879 from panyx0718/clean ... clean parallel do	7 years ago
dzhwinter	7cd24b1318	add ir memory optimize. (#14530) ... * follow comments. test=develop * Fix typo * fix compile error. test=develop * merge develop branch. test=develop * Remove set_equal * Polish code * Delete unused functions test=develop * polish code. test=develop * follow comment * polish code. * fix windows compile error. test=develop * fix op handle. * rerun ci. test=develop * rerun ci. test=develop * rerun macci. test=develop * polish code. test=develop * rewrite sort code. test=develop * remove unused code. test=develop * fix tests. test=develop * fix conflict. test=develop * follow comment. test=develop * merge develop branch. test=develop * fix tests. test=develop * remove ToTypeIndex. test=develop * rerun ci. test=develop	7 years ago
dzhwinter	acd4b75923	skip_opt_set support list (#14845) ... * test=develop * fix tests. test=develop	7 years ago
Xin Pan	47ea2534fb	clean parallel do ... test=develop	7 years ago
JiabinYang	c35fdf1581	Merge branch 'add_prefetch_in_nce' of https://github.com/seiriosPlus/Paddle into feature/add_prefech_hs	7 years ago
dzhwinter	00776b167a	fix memory opt skip set by name (#14774) ... * random failed. rerun ci. test=develop * windows failed. rerun ci. test=develop	7 years ago
gongweibao	f1fb64b17f	Add reduce sparse tensor feature. (#14757)	7 years ago
tangwei12	b653ed0516	add prefetch and remvoe selectedrows of bias	7 years ago
Wu Yi	aebc175cd4	add nccl2 dist tests (#14755) ... * add nccl2 dist tests test=develop * fix dist_base test=develop * fix tests test=develop * fix test on mac test=develop	7 years ago
tangwei12	4cb0100c8e	add prefetch in nce	7 years ago
Tink_Y	6d04a9cf47	fix api format and example (#14686) ... * fix api format and examples test=develop * Update executor.py test=develop * Update nn.py * Update nn.py test=develop * Update nn.py test=develop	8 years ago
Qiao Longfei	c8b0241da2	fix code style test=develop	8 years ago
Qiao Longfei	9589babe12	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into refactor-prefetch ... test=develop	8 years ago
Qiao Longfei	87e4edd2ea	fix grad_varname in remote prefetch	8 years ago
Qiao Longfei	d98c59fd2c	support none sliced variable	8 years ago
Qiao Longfei	af2f5fc824	fix some bugs	8 years ago
tangwei12	3639d99f99	Fix save and load lookup table/optimizer vars (#14301) ... * fix mkdir conflict * fix load/save lookup tables test=develop * add lookup_table_utils * fix load optimize vars on pserver * delete lookup table utils * fix save and load lookup tables * fix load optimizer var * fix load optimizer var, test=develop * fix python 3 style, test=develop * move lookup_table_utils to contrib utils	8 years ago
Qiao Longfei	2b6c0c09d6	add unit test	8 years ago
Qiao Longfei	cc6ef41dc3	update dist transpiler	8 years ago
Qiao Longfei	05c15a0867	Merge pull request #14467 from jacquesqiao/update-trainer-retry ... optimize distribute checkport	8 years ago
Qiao Longfei	fbc529db91	update test=develop	8 years ago
Qiao Longfei	98a0437d70	optimize distribute checkport ... test=develop	8 years ago
Sylwester Fraczek	8a1eeec579	add mkldnn prop_kind phase for inference-only case to pooling and activations (#14278) ... * add is_test to pooling and activations add prop_kind support for layers activation. conv and pooling add a pass that sets is_test to true add transpiler version of is_test pass test=develop * patch test and pass test=develop * add pass to analyzer.h test=develop * add is_test attr description & pass only on mkldnn in: activation_op.cc batch_norm_op.cc conv_op.cc dropout_op.cc lrn_op.cc pool_op.cc sequence_pool_op.cc softmax_op.cc * fix is_test handling for activation pool and conv * change description of is_test for all layers again * remove GetAttr(use_mkldnn) from pass * rename correct_mkldnn_test_phase to is_test and remove dependency on MKLDNN test=develop * review fix magic number * two if(..)s into one * Check is_test once and pass mkldnn forward prop kind * dereference shared_ptr with * (without get()) test=develop * add is_test_pass back test=develop	8 years ago
Qiao Longfei	080740b340	Merge pull request #14300 from jacquesqiao/dist-table-support-optimizer-regular ... dist table support other optimize and regular config	8 years ago
Qiao Longfei	04da1dcfb8	optimize import ... test=develop	8 years ago
Qiao Longfei	d03cbd1b8c	follow comment test=develop	8 years ago
typhoonzero	f3eafec19d	fix pserver weight decay multi inputs test=develop	8 years ago
Qiao Longfei	67050468e1	optimize code ... test=develop	8 years ago
Qiao Longfei	ffd5a832d8	fix code style	8 years ago
Qiao Longfei	3d8077e9fb	update optimizer	8 years ago
Qiao Longfei	fbcdb29d8c	fix import issue	8 years ago
Qiao Longfei	866d6bfe59	dist table support other optimize and regular config	8 years ago
Wu Yi	8fc05e0373	fix cpu build test=develop (#14260)	8 years ago
Wu Yi	306236c2c0	feature/DC asgd (#12722) ... * wip * add ref_by_trainer_id op * ready to test * fix ref inputs * refine rpc_op_handle * fix merge bug	8 years ago
tangwei12	d325e668b8	[1.1] Load vars on PSERVER (#14037) ... * fix dim0 in _load_slice_up_vars * fix dim0 in _load_slice_up_vars, fix innershape in delete_var_op * Revert "fix lookuptable in reduce strategy" This reverts commit 0e722c5 * add unit test for dist * add unit test for dist, test=develop * cancel revert, test=develop	8 years ago
Tao Luo	a7a3544f72	Merge pull request #14181 from sfraczek/sfraczek/transpiler-depthwise_conv-mkldnn-pass ... added transpiler pass for mkldnn depthwise_conv	8 years ago
Wu Yi	d51daede93	add ftrl support for dist train test=develop (#14176)	8 years ago
Qiao Longfei	f37bd03529	Merge pull request #14153 from jacquesqiao/fix-pserver-crash-when-no-parameter ... set en empty optimize block if pserver has no optimize block	8 years ago
Qiao Longfei	f3bbd3b43a	code style format ... test=develop	8 years ago
Qiao Longfei	add4b466d8	dist table only handle is_distributed table	8 years ago
Qiao Longfei	d78e8f23a6	code format ... test=develop	8 years ago
Sylwester Fraczek	ebd1d753ed	added transpiler pass for mkldnn depthwise_conv ... test=develop	8 years ago
Qiao Longfei	bf9764898d	add TestEmptyPserverOptimizeBlocks	8 years ago
Qiao Longfei	a11d4f300e	use len instead of size for python list	8 years ago
Qiao Longfei	fa84ba2350	set en empty optimize block if pserver has no optimize block	8 years ago
Xin Pan	eabb75f6b3	Merge pull request #14055 from dzhwinter/fix/mem_opt ... OrderSet default is mutableSet, not set.	8 years ago
Qiao Longfei	d26ff8cb2d	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into cpu-for-1.1-merge-with-shape	8 years ago
Wu Yi	26200f2e42	[1.1] [project] train imagenet using large batch size (#13766) ... * fix nccl2 lars dist support * put lars in momentum op * add tests lars * fix ci * fix cpu kernel * soft warning * remove lars in test_recognize_digits.py * move to another op * add file * update api.spec test=develop * update test=develop * fix api.spec test=develop * wip * wip, finish grad merge ops * wip, finish graph build * wip test running * work on 1 gpu * workable version * update * fix tests * fuse broadcast op * fix compile failed * refine * add batch merge test mnist * fix CI test=develop * fix build * use independent bn params for batch merge test=develop * update api.spec * follow comments and for test * wip * refine tests test=develop * follow comments test=develop * remove startup bn modify test=develop * follow comments test=develop * fix merge test=develop	8 years ago
dongzhihong	755c04df6e	rerun ci. test=develop	8 years ago
dongzhihong	c2bc700157	Merge remote-tracking branch 'origin/develop' into fix/mem_opt	8 years ago
Qiao Longfei	641369f92b	Merge branch 'dist-table-do-not-init-on-trainer' of ssh://github.com/jacquesqiao/Paddle into cpu-for-1.1-merge	8 years ago
Qiao Longfei	f1a3fb041b	Merge branch 'fix_lookuptable_in_reduce' of https://github.com/seiriosPlus/Paddle into cpu-for-1.1-merge	8 years ago
tangwei12	5ce3a32e06	Merge branch 'develop' into optimizer-prefetch	8 years ago
seiriosPlus	b6590b05fb	submit by tangwei12, test=develop	8 years ago
Qiao Longfei	93f173db7d	code format test=develop	8 years ago
Qiao Longfei	a13c788a04	fix a bug	8 years ago
Qiao Longfei	d52fcaf42e	replace table init op with fake init	8 years ago
Qiao Longfei	8fe90cbdb1	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dist-table-do-not-init-on-trainer	8 years ago
Qiao Longfei	4673fea551	trainer startup should not init table optimizer because it maybe large	8 years ago
dongzhihong	8ee3bdb66b	"recun ci. test=develop"	8 years ago
Xin Pan	4cd44c00c5	fix ... test=develop	8 years ago
Xin Pan	38cf553108	fix distributed ... test=develop	8 years ago