3249 Commits (54970444ce9baf3154a3775c81280ca97d59c83d)

Author	SHA1	Message	Date
zhaoyuchen2018	54970444ce	Improve transpose performance with tile sm copy, test=develop (#22311) ... * Refine code, fix select tile error,test=develop * Refine element type and some comments, test=develop * Refine comments and gpu utils, test=develop * Remove some useless condition * Refine floor and ceil, test=develop * refine for loop. test=develop Signed-off-by: zhaoyuchen <zhaoyuchen01@baidu.com>	5 years ago
guofei	3a59a7a11f	Make assign op support LoDTensorArray and modify while_loop API (#22309) ... This PR makes assign op support LoDTensorArray and enable the loop_vars in while_loop to support tuple or list.	5 years ago
liu zhengxi	14b6133b45	Implement InferencePassTest for testing precision of inference passes (#22387) ... * add InterencePassTest for testing precision of inference passes, test=develop	5 years ago
Guo Sheng	e7bbad6c49	Fix the leaving out of rnn_memory_helper_grad's output vars. test=develop (#22499)	5 years ago
Wilber	870f465887	fix test_fusion_seqpool_concat lod level between compile and runtime (#22488)	5 years ago
Huihuang Zheng	903039a3c6	Add Simple Framework for Transforming Dygraph to Static Graph (#22491) ... This PR provides very basic and simple framework for transforming Dygraph to Static Graph. API names, final outputs are not determined yet. Feel free to modify or add class/function/type when you think the framework is not extendable for you.	5 years ago
Yiqun Liu	dcfb603897	Enable the detection of subgraph composed of grad ops (#21223) ... * Add the first implememtation of fusion_group op #19621 (#3) * Add the dynamic load of nvrtc, and support runtime compiling of CUDA kernel using nvrtc. test=develop * Call CUDA driver api to launch the kernel compiled by nvrtc. test=develop * Disable for mac and windows. test=develop * Refine the codes to support manually specified num_threads and workload_per_thread. test=develop * Refine the CUDA kernel to support large dims. test=develop * Add DeviceCodePool to manage all device codes. * Add the first implementation fusion_group op. * Add unit-test for fusion_group op. * Add the check of result. * Add the check of nvrtc in unit-test. test=develop * Add comment to explain the inputs, outputs and features of fusion_group op. test=develop * Disable fusion_group op for mac and windows. test=develop * Make the compiling of device code return status instead of hanging up. test=develop * Add the check of whether there is CUDA driver library, and do not core dump when failing to call the CUDA driver API. * Unify fusion_group_op's input and output names. test=develop * Add the check of CUDA driver library in unittest. test=develop * Enable generating code for a given subgraph. #21126 (#4) * Enable generating code for a given subgraph. * Support sorting the subgraph. * Remove the rearange of expressions because we use the sorted subgraph directly. * Enable generating code for a subgraph which is composed of grad ops. * Use expression information to check the accuracy in unittest. * Separate load and store from computation expressions. test=develop * Improve the loading statements in generated codes. test=develop * Remove unused arguments from formal list. test=develop * Enable the detection of subgraph of grad ops. * Generate code for detected subgraph in fusion_group_pass. * Add an option in BuildStrategy to enable fusion_group_pass and add unittest. test=develop * Fix a bug when checking whether the shape of all inputs are the same. * Add debug information. * Remove subgraph_detector from inference/analysis to the common framework/ir directory. (#5) test=develop * Call subgraph_detector in fusion_group pass. test=develop * Disable fusion_group when WITH_GPU is OFF. test=develop * Refine all PADDLE_ENFORCE message. test=develop * Fix the case that some inputs are not defined in grad ops, and set op_role for fused op. test=develop * Follow review comments. test=develop	5 years ago
Aurelius84	50af6b5d79	polish no_grad_set of gradient and append_backward (#22440) ... * polish backward api doc test=develop, test=document_preview, test=document_fix * polish backward api doc test=develop, test=document_preview, test=document_fix * no_grad supports set of Variable test=develop, test=document_preview * polish sample code of append_backward test=develop, test=document_preview * modify assert into Raise TypeError test=develop,test=document_preview * fix unittest failed test=develop * rm useless file test=develop * polish en doc test=develop * polish code of no_grad_set test=develop * polish code of no_grad_set test=develop	5 years ago
Aurelius84	c2f39431aa	add skip_check_grad_ci of var_conv_2d (#22451) ... * add skip_check_grad_ci of var_conv_2d test=develop * modify check_shape_white_list test=develop	5 years ago
Wilber	7bc4b09500	add WITH_NCCL option for cmake. (#22384) ... cmake选项中添加了WITH_NCCL，显示指定是否编译NCCL的部分代码，WITH_NCCL默认打开，但如果WITH_GPU为OFF，则关闭WITH_NCCL 添加了PADDLE_WITH_NCCL定义 单机单卡能够关闭NCCL编译，多卡的话需要默认打开NCCL，如果关闭NCCL，则只能使用单卡 Co-authored-by: 石晓伟 <39303645+Shixiaowei02@users.noreply.github.com>	5 years ago
Bai Yifan	c8b90d8f9a	fix deformable_conv small cases, test=develop (#22441)	5 years ago
Leo Chen	822e5b36ec	Support int16 for Tensor (#22423) ... * add int16 support, test=develop * add test, test=develop * fix typo, test=develop * fix dtype error in slice, test=develop	5 years ago
tangwei12	7e2665c58b	fix bug with half (#22378) ... * fix bug with half communicator	5 years ago
liu zhengxi	0404e7a985	Update the precision of pad, pad2d, pad_constant_like's unit tests from fp32 to fp64 (#22394) ... * update the ut precision of pad pad2d pad_constant_like from fp32 to fp64, test=develop	5 years ago
xujiaqi01	371f377bea	add GeneralRoleMaker (#22295) ... * add GeneralRoleMaker which is for general usage * test=develop	5 years ago
lidanqing	ade5022681	[UT Coverage]Improve sum_mkldnn_op line coverage (#22275)	5 years ago
Wojciech Uss	92462e948d	improve elementwise_add_mkldnn_op test code coverage (#22359)	5 years ago
wangchaochaohu	e6ca512af1	refine convert type for numpy type (#22386)	5 years ago
lilong12	6d325a9470	enable test_dist_se_resnext_nccl.py, test=develop (#22067)	5 years ago
石晓伟	a392b77785	add no_grad_set value check in op_test, test=develop (#22370) ... * add no_grad_set value check in op_test, test=develop * update ops list, test=develop	5 years ago
gongweibao	ad2bc0c364	Fix a distribution bug and cleanup some not need logs. (#22381)	5 years ago
juncaipeng	7b0692a610	remove skip_check in test_activation_mkldnn_op, test=develop (#22376) ... * remove skip_check in test_activation_mkldnn_op, test=develop	5 years ago
whs	5f655d2cef	Refine unitest im2seq op. (#22372)	5 years ago
Zeng Jinle	039bb5050c	Polish backward.py to prune more ops (#22246) ... * polish backward prune, test=develop * fix control flow op bug, test=develop * add some unittests, test=develop * fix unittest args, test=develop * follow huihuang's comments, test=develop	5 years ago
Zeng Jinle	26da700ab2	Fix static graph dataloader early reset bug (#22365) ... * fix dataloader early reset bug, test=develop * change implementation, test=develop * fix ut, test=develop	5 years ago
zhupengyang	c6f888e5a5	update unittest accuracy to float64 for relu, prelu, maxout (#22273)	5 years ago
juncaipeng	325f0722bd	skip all mkldnn op test (#22363)	5 years ago
Chengmo	a22aaf7078	test=develop, fix test dist fleet geo unittest (#22287) ... fix timeout of test_dist_fleet_geo	5 years ago
Yiqun Liu	b7cac50b64	Implement a common python unittest to test the ir passes. (#22209) ... * Implement a common python unittest to test the ir passes. test=develop * Save the results in np.array and support to startup on CPU. test=develop * Fix the unittest. test=develop * Add check_program to check whether the optimized program is different from the origin one. test=develop * Remove the inferface all_ops. test=develop * Add exception test in pass_test. test=develop	5 years ago
songyouwei	4dec15999f	Allow Layer attr shadow (#22333) ... * allow sublayer or param shadow attrs * add unittest test=develop * change remove fn name test=develop	5 years ago
zhupengyang	cf0e264e63	all cases use large shape and skip special cases (#22332)	5 years ago
tangwei12	82bc814a57	integrated HALF_ASYNC to communicator (#21869) ... * add half_async in the communicator * fix DistributedStrategy	5 years ago
hong	00c0139e6e	add learning rate api of optimizer (#22080) ... * add learning rate api; test=develop * fix uni test converage; test=develop * fix travis ci error; test=develop * fix comment; test=develop * fix example error; test=develop * polish the api description, test=develop Co-authored-by: zhongpu <2013000149@qq.com>	5 years ago
zhupengyang	cb5e0d20c2	skip cases with small shape (#22318)	5 years ago
hong	008ea1ae2e	Fix test layers 单测随机挂问题 (#22278) ... * fix test_layers compare static graph and dygraph result; test=develop * fix test_layers random error; test=develop	5 years ago
Chen Weihang	35efbe6d95	Speeding up dygraph DataLoader with multiprocessing (#21762) ... * add multiprocess for dygraph data loader, test=develop * polish code & add safe gurad, test=develop * refactor dygraph dataloader & add signal handler, test=develop * fix member initializer compile error on ci, test=develop * fix member initializer compile error one more, test=develop * remove useless config, test=develop * skip windows incompatible problem, test=develop * add unittest for coverage, test=coverage * add more exception unittest case, test=develop * deal with signal handler coverage, test=develop * polish code & add signal handler tests, test=develop * deal with coverage ci problem, test=develop * split data loader test & coverage ci fix, test=develop * remove test_imperative_data_loader_with_exception, test=develop * remove singal process except test case, test=develop * add exception tests again & remove sample list test, test=develop * split normal and exception unittests to diff class, test=develop * polish doc for use_multiprocess effect in static mode, test=develop	5 years ago
zhupengyang	7ab44afbcc	check op tests' shape should be 100+ (#22047)	5 years ago
hong	737334989b	State dict do not count data parallel layers (#22169) ... * DataParallel state dict don't include _layers.; test=develop * add unitest of data parallel; test=develop * add load state test; test=develop	5 years ago
Bai Yifan	faba4b116a	Remove disable flag in test_fsp_op.py (#22171) ... * fix fsp_op, test=develop * fix fsp grad op maker, test=develop * update op_use_default_grad_op_maker.spec, test=develop	5 years ago
juncaipeng	67e9247f4c	Make every op in a line to avoid conflict in the future (#22274)	5 years ago
songyouwei	4bf2ccaa52	fix save_dygraph & save with nonexistent dir (#22266) ... * fix save_dygraph with nonexistent dir test=develop * minor fix test=develop * fix unittest test=develop * fix static save test=develop	5 years ago
zhongpu	9a7245de0e	fix optest of check_dygraph=False, test=develop (#22283)	5 years ago
danleifeng	f56e07492c	edit useless code;test=develop (#22276)	5 years ago
zhongpu	e31f8a2f5e	add CI check for sequence ops' unittests (#22221)	5 years ago
danleifeng	f5262865c0	change select_gpus into absolute values in launch.py (#22031)	5 years ago
zhupengyang	2b619493e2	all cases use large shape (#22241) ... enhanced ops: concat, nearest_interp, deformable_conv_v1, sequence_conv, transpose2, conv2d	5 years ago
Chen Weihang	efcdeb512f	Add all_parameters api for Program (#22180) ... * add all_parameters for Program, test=develop * refine unittest, test=develop * add example print result, test=develop	5 years ago
liym27	d90c121534	Revert "open test_parallel_executor_fetch_feed for PR_Windows_CI. test=develop (#22020)" (#22253) ... This reverts commit c7b03d308c.	5 years ago
juncaipeng	05ee05e248	no fp64 check for mkldnn op test, test=develop (#22217)	5 years ago
Zhen Wang	f2522e91c4	fix the type error caused by setting bool attr in OpDesc. test=develop (#22257)	5 years ago