3706 Commits (065b68b6ca53b3eb140a9f3ebe95b8cdd856fef4)

Author	SHA1	Message	Date
dzhwinter	225c11a91f	polish cudnn related code and fix bug. (#15164) ... * staged. * polish code * polish code. test=develop * polish code. test=develop * api change. test=develop * fix default value. test=develop * fix default value. test=develop	7 years ago
Yiqun Liu	454f4f2140	Rewrite is_empty op to avoid unnecessary data transform. (#15509) ... * Rewrite is_empty op to avoid unnecessary data transform. test=develop * Add the implementation of InferShape and InferVarType for is_empty op. test=develop * Rewrite is_empty op to avoid directly inherit OperatorBase. test=develop	7 years ago
xiaolil1	6724be2b0d	INT8 Pool kernel Key Creation Optimization. (#15883) ... * Optimize key creation of INT8 pool kernel to improve the peformance of ResNet-50 and MobileNet, especially for latency. test=develop * Optimize key creation of pool fp32 grad. test=develop	7 years ago
xiaoli.liu@intel.com	c4187dbd7c	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dequantize-reuse	7 years ago
Tao Luo	ba90e05281	Merge pull request #15917 from jczaja/prv-tensor-mkldnn-ops ... [MKL-DNN] Adjusting ops to Tensor modifications	7 years ago
Qiao Longfei	02425b2f64	fix compile	7 years ago
baojun-nervana	e4ab40a7b9	added concat op test=develop	7 years ago
colourful-tree	7d8f639883	Merge pull request #15902 from colourful-tree/new_develop ... remove mkldnn & fix commit	7 years ago
Tao Luo	effec86600	Merge pull request #15913 from liangan1/func_coverage ... Enable function coverage for U8/S8 ConvMKLDNNOpKernel	7 years ago
tensor-tang	8bc6381546	fix jitcodekey and refine test ... test=develop	7 years ago
tensor-tang	7044cfa7c7	add sgd jitcode and op test ... test=develop	7 years ago
tensor-tang	8e04133719	add benchmark and mkl sgd implement ... test=develop	7 years ago
tensor-tang	07efdb5139	Merge remote-tracking branch 'ups/develop' into jit/sgd	7 years ago
Jacek Czaja	c63f6b2039	- MKL-DNN pooling updated to set_prim_desc ... - MKLDNN ops revisited - disabled softmax modifications - disabled elementwise_add - reverted LRN modifications - reverted SUM primitive - Partial reviing of softmax - Enable softmax - Softmax changes - LRN is back - LRN partially disabled - LRN is back - LRN fix - compilation fixes - Sum fixed(hopefully) - Enabling (partially) elementwise_add - Fixes to elemenwise_add - Lint fixes quantize fix - compilation fix test=develop Disabling pooling - Disabled quantize op test=develop	7 years ago
qingqing01	8e439ccfff	Fix bug in fake_quantize_op and add more unit testing (#15912)	7 years ago
qingqing01	f4846bf3dc	loosly check in the InferShape of cross_entropy_op. (#15863) ... * loosly check in cross_entropy_op when soft_label is True * Add Runtime assertion in backward infer_shape check. * Skip InferShape check when un-know the input dimensions	7 years ago
Qiao Longfei	49f2f4f91d	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add-communicator	7 years ago
Qiao Longfei	f768fbf715	support multi graph ... test=develop	7 years ago
Yihua Xu	7396788694	Optimize gelu operation with mkl erf. ... test=develop	7 years ago
xiaoli.liu@intel.com	70759d181b	Optimize INT8 DeQuantize Op with primitive reuse. ... test=develop	7 years ago
Yiqun Liu	f4634d76d7	Optimize the CUDA implementation of sequence_expand op by reduce the times of copying lod data from CPU to GPU. (#15493) ... * Optimize the CUDA implementation of sequence_expand op by reduce the times of copying lod data from CPU to GPU. test=develop * Refine the op benchmark to support setting lod in config. test=develop	7 years ago
guomingz	630c1e8317	This PR improve performance of prior_box op about 1.25x faster on CPU. (#15909) ... * This PR improve performance of prior_box op about 1.25x faster on CPU. * Test Env:SKX 8180 with fake data on 28 threads(bs=1). * The below table shows the ~25% improvement which generated by [eval_tp_fake_data.py](https://github.com/PaddlePaddle/Paddle/issues/15618#issuecomment-464613976). | Type |Event | Calls | Total | Min. | Max. | Ave. | Ratio.| | ---------------- | ------------------ | ---- | ------- | -------- | -------- | ------------ | -------- | | w/ optimization | thread0::prior_box | 6000 | 921.201 | 0.110572 | 0.383402 | **0.153533** | 0.084585 | | w/o optimization | thread0::prior_box | 6000 | 1151.85 | 0.102276 | 0.426702 | **0.191976** | 0.103337 | test=develop * Fix the style issue. test=develop	7 years ago
Tao Luo	9c05421c97	Merge pull request #15914 from Sand3r-/mgallus/mkldnn-sum-code-reuse ... Refactor MKL-DNN Sum to use reference version on fallback	7 years ago
chengduo	7ca8553d4e	Add alloc_continuous_space_op (#15900) ... * add alloc_continuous_space_op test=develop * Polish code test=develop * follow comment test=develop	7 years ago
baojun-nervana	2ffacdebc2	Update ngraph version to v0.14 test=develop	7 years ago
Qiao Longfei	cf0511f21e	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add-async-ssa-graph-executor ... test=develop	7 years ago
Michal Gallus	6ebe9877bb	Improve code reuse at MKL-DNN sum ... test=develop	7 years ago
Qiao Longfei	b8491bfd4e	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add-communicator	7 years ago
liangan1	4acc522087	Enable function coverage for U8/S8 ConvMKLDNNOpKernel ... test=develop	7 years ago
Xin Pan	44e7fcddc5	Merge pull request #15844 from panyx0718/infer ... add per kernel config and remove const_cast.	7 years ago
Jacek Czaja	dec9cf53c8	[MKL-DNN] MKL-DNN specific Tensor modification (#15429) ... * - Implemented draft of primitive desc keeping in Tensor test=develop - TransposeMKLDNNHandler::AcquireSrcMemory was reimplemented - Added nchw and nc formats setting for sake of compatiblity Fixed unit tests - Worakaround to problem with 5D data in conv - Added 3D and 1D MKL-DNN formats for name handles for tensor test=develop - Fix to UTs test=develop - Conv fp32 op was updated Cosmetic fixes test=develop - tensor mkldnn cosmetics test=develop - Moved most of mkl-dnn specific code from Tensor to mkl-dnn utils * - Lint fixes test=develop * - setting prim dec in Tensor , sets also layout to kMKLDNN test=develop * - Moved creation of prim desc totally out of Tensor test=develop * - Cosmetic fixes adter review test=develop	7 years ago
heqiaozhi	08c96d1b48	remove mkldnn & fix commit ... test=develop	7 years ago
Xin Pan	5dd281f738	polish ... test=develop	7 years ago
heqiaozhi	fab09ac0b8	Merge branch 'new_develop' of https://github.com/colourful-tree/Paddle into new_develop	7 years ago
heqiaozhi	da4f5a2f18	remove mkl & fix commit ... test=develop	7 years ago
colourful-tree	f2d6473ef8	Merge branch 'develop' into new_develop	7 years ago
heqiaozhi	04f876f5bc	remove mkl & fix commit	7 years ago
dengkaipeng	373cfb0ccf	use kernel size in global_pooling. test=develop	7 years ago
dengkaipeng	60305196b8	fix spell mistakes. test=develop	7 years ago
Tao Luo	8a7efc78f1	Merge pull request #15882 from sfraczek/unique_ptr_dereference ... Change *(smart_ptr.get()) -> *smart_ptr	7 years ago
tensor-tang	a0c37662b9	enable sgd jitkernel refer code and test ... test=develop	7 years ago
xuezhong	1dad36f6aa	Merge pull request #15609 from xuezhong/add_sample_logits_op ... add sample_logits and sampled_softmax_with_cross_entropy op	7 years ago
Kaipeng Deng	9e524a7b51	Merge pull request #15870 from heavengate/fix_adaptive_pool_doc ... fix adaptive pool doc.test=develop	7 years ago
dengkaipeng	14df92fe8f	fix spell error. test=develop	7 years ago
dengkaipeng	144016fcfc	fix adaptive_pool and yolov3_loss. test=develop	7 years ago
Sylwester Fraczek	74672d1aff	Change *(smart_ptr.get()) -> *smart_ptr ... reason: dereferencing smart pointer is the same as the underlying pointer test=develop	7 years ago
tensor-tang	ee2321debd	Revert 15770 develop a6910f900 gelu mkl opt (#15872) ... * Revert "Optimze Gelu with MKL Erf function (#15770)" This reverts commit 676995c86c. * test=develop	7 years ago
xuezhong	81870723c6	Merge pull request #15605 from xuezhong/fix_bug_for_lstmp ... Fix bug for lstmp	7 years ago
dengkaipeng	eb65b4e47d	\frac -> \frac. test=develop	7 years ago
dengkaipeng	8167588f14	add blank after math::. test=develop	7 years ago
dengkaipeng	d9ec605873	use math:: instead of 29. test=develop	7 years ago
dengkaipeng	19292ac6a1	fix adaptive pool doc.test=develop	7 years ago
Yiqun Liu	7d96c74ab2	Initialize the benchmark tester for operator. (#15772) ... * Initialize the benchmark tester for operator. test=develop * Rearrange the codes. test=develop	7 years ago
Yihua Xu	676995c86c	Optimze Gelu with MKL Erf function (#15770) ... * Optimize for gelu operator * Set up the low accuracy mode of MKL ERF function. test=develop * Only enable MKLML ERF when OS is linux * Use the speical mklml version included vmsErf function to verify gelu mkl kernel. test=develop * Add the CUDA macro to avoid NVCC's compile issue. test=develop * Add the TODO comments for mklml library modification. test=develop * Clean Code test=develop * Add the comment of marco for NVCC compiler. test=develop	7 years ago
mozga-intel	5d132ecf83	Auto-cmake generator, auto-fill map (#15402) ... test=develop	7 years ago
Krzysztof Binias	1578c60bdd	Add new ut and remove unnecessary code ... test=develop	7 years ago
Xin Pan	5eb87506bc	add per kernel config and remove const_cast. ... test=develop	7 years ago
Qiao Longfei	31a05d3efd	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add-async-ssa-graph-executor ... test=develop	7 years ago
Dun	a83e470405	Profiler refine and add CUDA runtime api tracer (#15301) ... * refine profiler && add runtime tracer * test=develop * test=develop * test=develop * test=develop * test=develop * test=develop * test=develop * test=develop * fix bug && test=develop * add thread id map && test=develop * test=develop * testing * bug fix * remove cuda event && refine code && test=develop * test=develop * test=develop * test=develop * fix windows temp file && test=develop * test=develop * fix windows bug && test=develop * fix start up issue && test=develop * code polish && test=develop * remove unused code && test=develop * add some cupti cbid && test=develop * add FLAGS_multiple_of_cupti_buffer_size && test=develop * fix compile error && test=develop * add keyword && test=develop * fix && test=develop * code polish && test=develop	7 years ago
Qiao Longfei	cc71e89499	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add-async-ssa-graph-executor ... test=develop	7 years ago
mozga-intel	13ec2d331b	Enable momentum operator for a ngraph engine (#15673) ... * Enable momentum operator for a ngraph engine test=develop * Update tests test=develop * Unnecessary line of the code as intended was removed test=develop	7 years ago
xuezhong	eb7bc3e7ea	remove non-ascii charactor ... test=develop	7 years ago
tensor-tang	e1c707fe9c	fix warnings (#15790) ... * fix warnings test=develop * fix enforce test test=develop	7 years ago
xuezhong	d328660304	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_bug_for_lstmp	7 years ago
xuezhong	f2262d7336	update comment ... test=develop	7 years ago
Tao Luo	6402424f7a	Merge pull request #15773 from chengduoZH/fix_shape_api_doc ... Fix shape api doc	7 years ago
xuezhong	d12252e6a6	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_sample_logits_op ... test=develop	7 years ago
xuezhong	c5360a3f6b	refine code	7 years ago
tensor-tang	5aea2cd2e0	Merge pull request #15652 from tensor-tang/refine/pyramiddnn ... refine fused emb seq pool	7 years ago
mozga-intel	df23a6f894	Enable cross_entropy operator for a ngraph engine (#15674) ... * Enable cross_entropy operator for a ngraph engine test=develop * Update tests test=develop * Added PADDLE_ENFORCE for the batch_norm operator test=develop * Update the message about which format are supported right now test=develop	7 years ago
Yiqun Liu	56a5039e24	Correct the doc in Python API (#15725) ... * Correct the comment in control_flow.py. * Correct the argument list of ops. test=develop * Update API.spec. test=develop * Skip op_callstack attr for all op apis. test=develop * Remove use_mkldnn and is_test from python api. test=develop * Remove use_mkldnn and is_test from op_proto_maker and hard-coding them in python when generating doc string. test=develop	7 years ago
baojun	72061b0ac0	Add ngraph op coverage (#15721)	7 years ago
chengduozh	d79d2f686c	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_shape_api_doc ... test=develop	7 years ago
xuezhong	4424021623	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_sample_logits_op	7 years ago
Yihua Xu	685a20ef56	Add JIT CRF_decoding and Layer_norm unit-test (#15699) ... * Add the CRFDecoding and LayerNorm's test case test=develop * Fix the size checking issue test=develop * Remove the remnant code test=develop * Add TestAllImpls and double support test=develop * Clean Code test=develop * Add benchmark test for LayerNorm & CRFDecoding test=develop	7 years ago
tensor-tang	75fc792d40	fix when table width larger than 64 ... test=develop	7 years ago
tensor-tang	40402d5e68	add emb seqpool jitcode ... test=develop	7 years ago
tensor-tang	2ccbcb157d	Merge remote-tracking branch 'ups/develop' into refine/pyramiddnn	7 years ago
chengduozh	3ce12b1b8e	fix shape api doc ... test=develop	7 years ago
Dun	5e6834d891	inplace group_norm (#15754) ... * inplace group * test=develop	7 years ago
Hongyu Liu	8c0292dead	Merge pull request #15717 from phlrain/fix_leak ... Fix lstm possible leak	7 years ago
Tao Luo	4da291c6a3	Merge pull request #15726 from qingqing01/fix_api_doc ... Fix row_conv doc	7 years ago
Dun	e4b9fcdbd2	More restrict check load_combine_op. (#15479) ... * fix && test=develop * fix && test=develop * test=develop	7 years ago
qingqing01	48a5cccbcd	Fix debug mode in prior_box_op (#15702) ... * Fix debug mode in prior_box_op * Refine code	7 years ago
Dang Qingqing	2868232556	Fix row_conv doc ... test=develop	7 years ago
tensor-tang	a3a3d3d861	add embseqpool jitkernel mkl impl and use it ... test=develop	7 years ago
tensor-tang	15da2f9a0d	add embseqpool jitkernel refer code, test and benchmark ... test=develop	7 years ago
tensor-tang	c2ccf14590	Merge remote-tracking branch 'ups/develop' into refine/pyramiddnn	7 years ago
qingqing01	abcefe7211	Fix debug mode in fake_quantize_op (#15693) ... * Fix debug mode in fake_quantize_op * Remove template specialization	7 years ago
liuhongyu	029be5fda9	fix lstmp bug; test=develop	7 years ago
liuhongyu	393fa6021e	set lstm lstmp unsed pointer to nullptr; test=develop	7 years ago
liuhongyu	869f00ffc6	set lstm lstmp unsed pointer to null	7 years ago
jerrywgz	6f11f35abe	Merge pull request #15703 from jerrywgz/enhance_expand_op ... support multiple var types for expand op	7 years ago
Tao Luo	3086502522	Merge pull request #15704 from Sand3r-/mgallus/old-fc-mkldnn-branch-fix-develop ... Fix old FC backward weights descriptor creation	7 years ago
baojun	c47e258ea4	Add ngraph sum, sigmoid, relu_grad and tanh_grad op (#15642) ... * Added ngraph sum op test=develop * Added sigmoid, relu_grad and tanh_grad test=develop * remove duplicates test=develop	7 years ago
tensor-tang	33d0cebbff	Merge pull request #15695 from tensor-tang/fix/name ... fix jitcode name, use after free	7 years ago
Michal Gallus	7a8eff36a6	Fix old FC backward weights descriptor creation ... test=develop	7 years ago
chengduo	ad61e1b22c	fix potential bug (#15688) ... test=develop	7 years ago
dzhwinter	f9ac88e1a0	Merge pull request #15694 from liuwei1031/fix_security_issue ... Fix security issue	7 years ago
jerrywgz	8fc0fc314a	support multiple var types for expand op, test=develop	7 years ago