4275 Commits (0600b370ea714f7793a15f6756ec376161ad3d5c)

Author	SHA1	Message	Date
tensor-tang	0600b370ea	[CPU] refine softmax op fwd on CPU (#17522) ... * refine softmax fwd test=develop * fix compile issue wih gpu test=develop * add value clip to avoid exp	7 years ago
mozga-intel	035771512d	Enable elementwise min operator for ngraph (#17521)	7 years ago
jerrywgz	c1aae8b8d2	Fix GetExpectedKernelType in Concat op (#17459) ... * fix concat op vartype check, test=develop	7 years ago
Qiao Longfei	58f7695ab2	Async exe support communicator (#17386) ... Async exe support communicator	7 years ago
mozga-intel	109b5aed5a	[NGraph] Enable reshape operator test=develop (#17512)	7 years ago
Krzysztof Binias	43d15b9d96	Enable square operator for the nGraph Bridge. (#17551) ... test=develop	7 years ago
Sevin F. Varoglu	f86f49e779	[NGraph] add increment op to ngraph engine (#16929) ... * add increment op to ngraph engine test=develop * fix style errors test=develop	7 years ago
baojun	8923612b10	NGraph enable parse serialized graph test=develop (#17453)	7 years ago
guomingz	2281ebf0f3	Enable the convolution/relu6(bounded_relu) fusion for FP32 on Intel platform. (#17130) ... * Relu6 is the bottleneck op for Mobilenet-v2. As the mkldnn supports the conv/relu6 fusion, we implement it fusion via cpass way. Due to the int8 enabling for this fusion will be supported in MKLDNN v0.20, so this PR is focused on the fp32 optimization. Below table shows the benchmark(FPS) which measured on skx-8180(28 cores) Batch size | with fusion | without fusion -- | -- | -- 1 | 214.7 | 53.4 50 | 1219.727 | 137.280 test=develop * Fix the format issue test=develop * Add the missing nolint comments. test=develop * Fix the typos. test=develop * Register the conv_brelu_mkldnn_fuse_pass for the MKLDNN engine. test=develop * Adjust the indentation. test=develop * Add the test_conv_brelu_mkldnn_fuse_pass case. test=develop * Slightly update the code per Baidu comments. Let the parameter definition embedded into the code. That's will make the code easy to understand. test=develop	7 years ago
Yibing Liu	f9796b1249	Add LAMB Optimizer support (#17489) ... * Add LAMB optimizer * Expose LAMB Optimizer's APIs test=develop, test=document_preview * Cleanup code & doc test=develop, test=document_preview * Update lamb optimizer's formula test=develop	7 years ago
mozga-intel	99ab57123c	Enabled ngraph elementwise max operator (#17517)	7 years ago
Tao Luo	3d19f44a89	remove unused SERIAL compiler option (#17500) ... test=develop	7 years ago
mozga-intel	1eb151752e	Enable abs operator for a ngraph test=develop (#17436)	7 years ago
liuwei1031	ba70cc499e	fix security bugs : (#17464) ... http://newicafe.baidu.com:80/issue/PaddleSec-33/show?from=page http://newicafe.baidu.com:80/issue/PaddleSec-28/show?from=page http://newicafe.baidu.com:80/issue/PaddleSec-25/show?from=page http://newicafe.baidu.com:80/issue/PaddleSec-24/show?from=page http://newicafe.baidu.com:80/issue/PaddleSec-21/show?from=page http://newicafe.baidu.com:80/issue/PaddleSec-20/show?from=page test=develop	7 years ago
Zhaolong Xing	ff7f911b4d	add quant_dequant_moving_avg_max_abs op (#17480) ... * add quant_dequant_moving_avg_max_abs op test=develop * add more note for quantdequant op test=develop	7 years ago
Qiao Longfei	287de41c04	Optimize communicator flags (#17494) ... * optimize communicator flag * change flags in init py test=develop	7 years ago
lvmengsi	10b23a72c1	Double backward elementwise div (#17416) ... * double backward, elementwise_div * fix dx empty. test=develop * bug fix (#17392) fix secure bug * Eanble stack operator for a Ngraph, test=develop (#17406) * fix sqrt_grad_grad unittest. test=develop (#17410) * fix sqrt_grad_grad unittest. test=develop * disable sqrt_grad_grad unittest. test=develop * test=develop, fix unittest * test=develop, fix unittest * test=develop, fix unittest * test=develop, fix bug * fix unittest. test=develop * fix unittest dx. test=develop * tmp fix! for test... test=develop * reduce tmp, test=develop * test=develop, reduce tmp * fix broadcast unittest. test=develop * fix format. test=develop * refine code. test=develop * refine code. test=develop * refine GetDoubleGradSafeTensor. test=develop * fix format. test=develop	7 years ago
Zeng Jinle	3d4e8268c6	fix recurrent fwd bug when no backward and scope clear (#17460)	7 years ago
lvmengsi	977e9fcb27	support elementwise_sub double backward (#17476) ... add elementwise_sub_grad_grad op for backward of backward calculation	7 years ago
chengduo	5a6ab38013	Add record event And remove CSP (#17447) ... * add record_event test=develop * remove csp test=develop	7 years ago
Yan Xu	0217555530	polish parallel dygraph code (#17164) ... * add var grad hook test=develop	7 years ago
Bai Yifan	3a9ae28d32	fix assert,test=develop (#17445)	7 years ago
zhaoyuchen2018	b02f2aff04	Add conditional compile for gru opt (#17368) ... * improve gru unit performance. refine code test=develop Signed-off-by: zhaoyuchen <zhaoyuchen01@baidu.com> * Add conditional compile for gru opt Not enable gru opt if compute ability < 700 test=develop Signed-off-by: zhaoyuchen <zhaoyuchen01@baidu.com> * refine code. test=develop Signed-off-by: zhaoyuchen <zhaoyuchen01@baidu.com>	7 years ago
Zeng Jinle	712bfb17cb	fix recurrent_op,test=develop (#17433)	7 years ago
mozga-intel	6ee6700fac	Eanble stack operator for a Ngraph, test=develop (#17406)	7 years ago
Krzysztof Binias	0823a7bc8b	Optimize the sequence padding op (#17403) ... test=develop	7 years ago
baojun	1ce7b45b9e	NGraph Added fill_zeros_like op test=develop (#17295)	7 years ago
baojun	910196524d	NGraph Added dropout and dropout_grad to ngraph test=develop (#17320)	7 years ago
mozga-intel	b189480734	Ngraph Enable gather operator test=develop (#17296)	7 years ago
lvmengsi	4ef631013c	Double backward sqrt (#17387) ... * double backward sqrt * refine unittest. test=develop * refine test. test=develop * remove alpha in unittest. test=develop	7 years ago
lvmengsi	5d1ac41b00	Double backward reduce mean (#17372) ... * test=develop, double backward reduce_mean * add comment. test=develop * fix format. test=develop * rename GradGrad -> DoubleGrad. test=develop * fix op_use_default_grad_op_maker.spec. test=develop	7 years ago
jerrywgz	0cae5a36b6	enhance generate mask labels, test=develop (#17380)	7 years ago
Kaipeng Deng	bd9bef5a4e	add elementwise_add_grad_grad op (#17366) ... * add elementwise_add_grad_grad op. test=develop * use defined GradMaker. test=develop	7 years ago
jerrywgz	1c6d064627	add collect fpn proposals op,test=develop (#16074) ... * add collect fpn proposals op,test=develop	7 years ago
Kaipeng Deng	60be66e2c0	support fc_op double grad (#17317) ... * add double grad for mul_op. test=develop * fix format. test=develop * fix format. test=develop * fix format. test=develop * refine code. test=develop * remove setzero. test=develop * fix dx/dy init bug. test=develop * fix format. test=develop	7 years ago
liuwei1031	0863599323	Fix the uninitialized gru_value.output_value. (#17197) ... test=develop	7 years ago
Yihua Xu	218d8d8f73	Optimize the computing kernel of sequence_reverse operator (#17349) ... * Optimize the computing kernel of sequence_reverse operator. test=develop * Clean code test=develop * Fix for cpplint syntax checking. test=develop * Fix the compile warning issue. test=develop	7 years ago
Yiqun Liu	dcda20233c	Optimize the elementwise op using eigen (#15494) ... * Optimize the elementwise op with CUDA kernels. test=develop * Support setting of attr in op config file. test=develop * Add the support the setting dtype and initializer in config. test=develop * Save workspace. * Add initializer "zeros". test=develop * Fix compiling error. * Support the use of existed file to initailize tensor in op_tester. * Use eigen to optimize the elementwise_add/mul for the case that x and y have the same dims. test=develop	7 years ago
Kaipeng Deng	8bae8590ac	add double grad for elementwise_mul op (#17255) ... * add double grad for elementwise_mul. test=develop * remove comment. test=develop * fix grad sum. test=develop * fix for axis expand. test=develop * add test for axis expand. test=develop	7 years ago
Kaipeng Deng	11d3a38f25	add double grad for square op (#17173) ... * add double grad for square. test=develop * formax code. test=develop * fix for grad sum. test=develop * refine shape. test=develop * refine extract. test=develop	7 years ago
zhoukunsheng	d4b67e1692	Add Where Op(#16793)	7 years ago
zhoukunsheng	1bfff02047	Add Diag Op(#17027)	7 years ago
zhaoyuchen2018	8a2caacdbc	improve gru unit performance. (#16338) ... refine code fuse cublas calling and kernels into one cuda kernel. test=develop Signed-off-by: zhaoyuchen <zhaoyuchen01@baidu.com>	7 years ago
qingqing01	e32c9888f5	Double backward of conv2d. (#17211) ... * Add conv2d_grad_grad_op * Extracte the cuDNN conv algo searching code in conv_cudnn_helper.h. - Now use it in conv2d_grad_grad. - Will simply the searching code in conv2d and conv2d_grad in next PR. * Enhance and fix bug in unit testing of gradient_checker. * Support to fetch empty variables，return None in Python.	7 years ago
Zeng Jinle	fff270eacd	follow comments,test=develop (#17273)	7 years ago
zhoukunsheng	4292bd8687	Mod floordiv (#17251) ... * test=develop add elementwise_mod and elementwise_floordiv, fix equation problem in elementwise_mod	7 years ago
xiaoting	9ed4aaada4	modified formula for Lrn (#17281) ... * modified formula for lrn test=develop * modified api.spec test=develop	7 years ago
zhaoyuchen2018	792443ef23	Refine elementwise kernel. (#16952) ... * Refine elementwise kernel. Add a simple cuda kernel if grad x and y both exist Use 2D block cuda kernel to do broadcast. test=develop Signed-off-by: zhaoyuchen <zhaoyuchen01@baidu.com> * refine code. test=develop Signed-off-by: zhaoyuchen <zhaoyuchen01@baidu.com> * refine code. test=develop Signed-off-by: zhaoyuchen <zhaoyuchen01@baidu.com>	7 years ago
Yiqun Liu	6b84688ba2	Optimize the cuda implementation of sum_op (#17283) ... * Optimize the cuda implementation of sum_op, which add two lod_tensors inplace. test=develop * Use eigen to add to tensors. test=develop	7 years ago
chengduo	db5e74ab95	update assert (#17282) ... test=develop	7 years ago