749 Commits (43a82d83fa60be0524a752a10e522429783cbf67)

Author	SHA1	Message	Date
Jacek Czaja	cef95ee30d	[MKL-DNN] Refactoring Softmax (#19312) ... * - First set of modifications - Compilation fixes - compilation fix - Another compilation fix - Moved AcquireSoftmaxPrimitiveDescriptor call into handler - MKL-DNN Softmax PD refactor test=develop - Compilation fix test=develop - another compilation fix - cosmetcis test=develop - Compilation fix - Fix to crash when softmax backward is created * - Fixes after review of softmax refactoring test=develop	6 years ago
Zeng Jinle	0a73f7202a	Add retry_allocator for gpu (#19409) ... * add retry_allocator for gpu, test=develop * follow chengduoZH's comments, test=develop * follow huihuang's comments,test=develop * change f,l in enforce.h to be file,line, test=develop * increase code coverage by adding unittests, test=develop * fix CMakeLists.txt, test=develop	6 years ago
Jacek Czaja	ecd9f330c9	[MKL-DNN] Fix to face model on AVX512 platforms (#19282) ... - Refactor step 1 - Compilation fix - Yet another compilation fix - Even more compilation fix - Lint fixes test=develop - Removed deprectaed PADDLE_ENFORCE occurance test=develop - Candidate fix to BN forward - Lint fixes test=develop - Refactoring in data_layout_transform - compilation fix - Another comppilation fix - Step further into darkness - Yet another compilation fix - Yet another compilation fix - missing header - compilation fix - Added MKLDNN -> Paddle conversion in fetch op test=develop - Compilation fix test=develop - Lint test=develop - Mul fix - Fix to MKLDNN MUL op and Elementwise MUL UT test=develop - Workaround for diffrent weights with groups representation Paddle vs MKL-DNN. test=develop - Candidate fix for 5D convolution with groups - Refactor of fix for conv3d and conv2d in fetch op test=develop - Compilation fix - Still same compilation fix - Compilation fix - Compilation fix - Reverted refactoring of fixes - Adapted test_conv2d_int8_mkldnn so it exects data in NCHW format not NHWC test=develop - minor fix in UT test=develop - Lint fixes test=develop	6 years ago
liuwei1031	d6cb1a4122	add dynamic C runtime support on windows, test=develop (#19502)	6 years ago
Zeng Jinle	c2c5b1b941	remove signal raise msg, test=develop (#19527)	6 years ago
Zeng Jinle	caf59d0f3f	Add signal message to stderr (#19421) ... * add signal message to stderr, test=develop * add unittests for ugly SignalHandle, test=develop	6 years ago
Yi Liu	efb05ba258	supports multiple NCCL communicators preserved in NCCLCommContext (#19407) ... * supports multiple NCCL communicators preserved in NCCLCommContext test=develop * add ut for c_comm_init_all operator and fix cuda resource release problem test=develop	6 years ago
wopeizl	b8aa37d529	save the callstack information to file when exception throws test=dev… (#19324) ... * save the callstack information to file when exception throws test=develop	6 years ago
Tao Luo	6527a7df67	replace part of PADDLE_ASSERT to PADDLE_ENFORCE (#19285) ... * replace part of PADDLE_ASSERT to PADDLE_ENFORCE test=develop * remove unused fallback_alloc_size_ * add unit-test of CUDAPinnedAllocator test=develop	6 years ago
Yihua Xu	b920395842	Use sparse matrix to implement fused emb_seq_pool operator (#19064) ... * Implement the operator with sprase matrix multiply * Update the URL of mklml library. test=develop * Disable MKLML implematation when using no-linux. test=develop * Ignore the deprecated status for windows test=develop	6 years ago
Zeng Jinle	91a0911ca3	Make PADDLE_ENFORCE_EQ support types that cannot be converted to std::string (#19243) ... * make PADDLE_ENFORCE_EQ support cannot to string types, test=develop * follow huihuang's comments, test=develop	6 years ago
Zeng Jinle	708bd9798d	move_flags_to_unified_files_for_management, test=develop (#19224)	6 years ago
Zeng Jinle	002f325dcd	add PADDLE_ENFORCE_CUDA_SUCCESS, test=develop (#19211)	6 years ago
Adam	b837689e97	Add generalized Conv+Activation MKLDNN fuse pass creation (#19072) ... test=develop	6 years ago
gongweibao	29d8781240	Polish fleet API to support cuda collective mode and nccl2 mode. (#18966) ... Polish fleet API to support cuda collective mode and nccl2 mode	6 years ago
wopeizl	80b7ef6fc8	add tensorrt support for windows (#19084) ... * add tensorrt support for windows	6 years ago
Zhang Ting	c2063217e7	optimize error message for "embedding" and "cross_entropy" OP (#18765) ... * optimize error message, test=develop * optimize error message, test=develop	6 years ago
liuwei1031	a43a763b54	fix warpctc.dll not found issue (#18761) ... * fix warpctc.dll not found issue, test=develop * revert the linux platform change, test=develop * delete warpctc_lib_path.h.in, test=develop * add SetPySitePackagePath function * fix warpctc.dylib not found issue on Mac, test=develop * improve the paddle lib path setting logic, test=develop * fix mac ci issue caused by test_warpctc_op unittest, test=develop * tweak code, test=develop	6 years ago
Zeng Jinle	08fa98f7cc	Fix gpu_info PADDLE_ENFORCE_GT when fraction_of_gpu_memory_to_use=1.0 (#18950) ... * fix gpu_info, test=develop * fix reserving gpu memory calculation bug, add fraction=1 unittest, test=develop * fix bug again for reserving size, test=develop	6 years ago
Jacek Czaja	5cf2d38594	- Removed passing X from FWD to GRAD via device context (#18911) ... test=develop - Extracted key generation from FWD and GRAD into separate function test=develop - Compilation fix test=develop - another compilation test=develop	6 years ago
Huihuang Zheng	ea6ee76fa9	GPU allocation uses fraction of available memory (#18896) ... GPU allocation uses fraction of available memory, also fix the GetUsed without lock	6 years ago
Jacek Czaja	cfcb96d2df	[MKL-DNN] Fix int8 performance regression (#18758) ... test=develop - optimization of TID to string test=develop	6 years ago
Huihuang Zheng	0d3f16f53e	Try to modify external gflags to solve CI compilation (#18872)	6 years ago
Huihuang Zheng	cfce4994cf	Merge cuda 9/10 dockerfile with root dockerfile (#18693) ... Also fix a dependency error which may cause compile error	6 years ago
lidanqing	9ecd8ee789	change ComputeINT8 to template version to remove checking dst_datatype code (#18756) ... * change INT8 to template so that checking dst_dt with if-else could be removed. CI will be enabled after fixing reviews * reverse user_residual_memory_p and user_bias_memory_p declaration scope test=develop	6 years ago
Jacek Czaja	95c1816ec0	[MKL-DNN] Extended LRN with reusing via Acquire API (#18675) ... test=develop - compileation fix - Yet another compilation fix - Even yet another compilation fix - Surprise! Again compilation fix - lint fixes test=develop - Fix to workspace acquire of LRN test=develop - Fix to hash of BWD LRN test=develop - fix to lrn BWD PD acquire test=develop - Fixing LRN PD creation test=develop - cosmetic fix in comment test=develop - Fixes after review test=develop	6 years ago
chengduo	fd3aad6cb3	Make fuse_optimizer_op_pass also work when the model contains sparse gradients. (#18664) ... * support sparse gradients test=develop	6 years ago
Jacek Czaja	0d8e6c9b8b	MKL-DNN upgrade to 0.20 (#18370) ... test=develop	6 years ago
zhouwei25	772e09560e	Optimize the content of error reporting information, print error code and official document web sites (#18671) ... optimize the error reporting information of cuda related API index on develop: 130ac17 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into develop	6 years ago
Zeng Jinle	ae58afc546	Feature/auto_growth_allocator (#18561) ... * feature/auto_growth_allocator, test=develop * add unittest of AlignedAllocator, test=develop * try to turn on auto_growth to test on CI, test=develop * fix segmentation fault in mixed_vector.h, test=develop * add unittests, test=develop	6 years ago
liuwei1031	759530966c	print out error code of cudaGetDeviceProperties if failed (#18643)	6 years ago
Jacek Czaja	71d883b8ef	[MKL-DNN] Reimplemented pool2d mkl-dnn to use Acquire API (#18585) ... * - Added partial draft of pooling acquire - Workspace support - compilation fix - Added draft of pooling backward reimplementation - Segfault fix - reverted 'any' for diff_dst crewation in pooling - Lint fixes test=develop - lint fixes test=develop - Further lint fixes test=develop * - Fixes after review test=develop * - Lint fixes test=develop * - Even more lint fixes test=develop	6 years ago
Tao Luo	076f833110	add config.SetMkldnnCacheCapacity api for mkldnn cache clear strategy (#18580) ... * add config.SetMkldnnCacheCapacity api for mkldnn cache clear strategy test=develop * enhance MkldnnPostReset test=develop * add comments for mkldnn_cache_capacity field test=develop	6 years ago
gongweibao	c0a82748cf	Polish backwards optimizer dependency codes and use more default values. (#18255)	6 years ago
Zeng Jinle	be24e5b391	Clean unused code of dim and place (#18565) ... * clean code of dim and place, test=develop * fix failed unittests, test=develop	6 years ago
Jacek Czaja	8869d7f735	Activations MKLDNN ops refactoring (#18191)	6 years ago
Jiabin Yang	667f88f9a6	Fix/gcc 4.8 ubt link error (#18558) ... * test=develop, fix docker with paddle nccl problem * test=develop, fix/gcc_4.8_ubt_link_error * test=develop, fix code format	6 years ago
Physher	0caa08ea40	Add mkldnn int8 mul-op kernel (#17834)	6 years ago
Tao Luo	fe32879d2a	add mkldnn shapeblob cache clear strategy (#18513) ... * add mkldnn shapeblob cache clear strategy test=develop * refine with comments test=develop * make cache clear strategy more safey test=develop * add lock for GetShapeBlobSize test=develop	6 years ago
chengduo	55baeceddb	Enhance execution error info (#18482) ... * enhance execution error info test=develop	6 years ago
Tao Luo	3f3112ceb0	add shape_blob for cache mkldnn primitive (#18454) ... test=develop	6 years ago
Leo Zhao	8f5fffca0a	rename mkldnn set/get_cur_thread_id() to set/get_cur_mkldnn_session_id() (#18453) ... * rename mkldnn set/get_cur_thread_id() to set/get_cur_mkldnn_session_id() test=develop * update session id definition and adjust logic for default behavior test=develop * reset logic in mkldnn reuse as most of cases work in default. test=develop	6 years ago
Yi Liu	a873fa84ce	supports collective training with programs (#18392) ... 1. Since allreduce op has 4 reduce types, We split these four reduce types into four ops 2. We also refined the collective op code, e.g. we separated the collective op kernel into CPUKernel and CUDAKernel, and remove the device specified DeviceContext parameter in template as we already knew the target DeviceContext 3. We remove the newly added Collective op role to reduce the complexity of program and graph analysis	6 years ago
Brian Liu	4bc2987d2f	Fix bug in quantize kernel which cause crash in vgg16/19 model (#17964) ... * Fix bug in quantize kernel which cause crash in vgg16/19 model test=develop * refine the code to reduce verbose code; test=develop * remove useless code; test=develop	6 years ago
Leo Zhao	681d3553f1	Fix potential mkldnn concat/pool/conv kernel issues (#18393) ... 1. some key generation method is not aligned with PR#17965 2. enlarge ptr lifetime to avoid memory release if SetBlob fails otherwise it will get core dump. test=develop	6 years ago
HaoRen	9931bc64f5	add dependecy of collective_helper (#18365) ... * add dependecy of collective_helper * test=develop fix dependecy of collective_helper	6 years ago
Michał Gallus	8409693272	Reset DeviceContext after quantization warmup (#18182) ... test=develop	6 years ago
HaoRen	b7128bac5f	supports collective communicated training (#18175) ... * fix prepare context redundant code problem, optimize executor by caching create_varaiables test=develop * supports collective training in executor * make fetch_list runable with variables, add more unittest for use_program_cache test=develop * fix comment test=develop * use unique name for nccl_id * supports output to stream in program_to_code * insert sync_comm_stream before regularization; add skip_op_callstack capability in program_to_code * set op role in collective training * add collective op role * remove orig file * add build optimizer by strategy * add collective strategy * refine collective strategy * add multi-process role maker * refine strategy building factory so that we can easily plugin more strategy * scale loss grad in collective sgd transpiler * add support for distributed fc * code format * revert some features for dist fc * add support for distributed fc training * fix prepare context redundant code problem, optimize executor by caching create_varaiables test=develop * supports collective training in executor * make fetch_list runable with variables, add more unittest for use_program_cache test=develop * use unique name for nccl_id * supports output to stream in program_to_code * insert sync_comm_stream before regularization; add skip_op_callstack capability in program_to_code * set op role in collective training * add collective op role * fix comment test=develop * remove orig file * add build optimizer by strategy * add collective strategy * refine collective strategy * add multi-process role maker * refine strategy building factory so that we can easily plugin more strategy * scale loss grad in collective sgd transpiler * add support for distributed fc * code format * revert some features for dist fc * add support for distributed fc training * test=develop add collective op unittest standard * test=develop remove the test_collective directory * test=develop remove the test_collective directory * remove slicegather test * code format for reducescatter * update attr of shard_index_op * Modify macro nccl_helper * remove test without distribute * macro collective_helper * marcro update * test=develop update support python3.5 * test=develop change gpu memory use to 0.1 when test * test=develop update ut equal func * test=develop set flags to 1.5 * test=develop fix pickle dumple py35 * test=develop fix divide in slice and add sync_comm_stream update atol and rtol to 1e-05 rm shard_index op and test modify read input from file to read from memory remove origin_program in framework and add i/o in c_sync_calc_stream * test=develop update unittest sync operator I/O	6 years ago
Jacek Czaja	c2efdfd5bc	[MKL-DNN] Extending reusing to Elementwise_add_mkldnn op (#18146) ... * - Reusing of reuder used in elementwise_add_mkldnn - Added MKL-DNN sum prim reusing test=develop - Compilation fixes test=develop - Yet another compilation fix test=develop - Yet another compilation fix test=develo - Yet another linking fix test=develop - Final compilation fix test=develop - lint fixes test=develop - Lint fixes test=develop * - Fixes after review test=develop	6 years ago
chengduo	4978db2c10	Remove nccl dep when the number of GPU is 1 (#18158) ... * remove nccl dep when the number of GPU is 1 test=develop	6 years ago