Paddle

Commit Graph

Author	SHA1	Message	Date
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
翟飞跃	802ea50956	fix spelling errors (#17941 ) * fix spelling errors; test=develop * Update API.spec update md5 * Update API.spec * change the order of api;test=develop	6 years ago
qingqing01	80d2e66f9e	Update backward appending stragety to support double backward and fix some bug. (#18104 ) * Update backward.py: - If there is no input grad var in all outputs of previous ops, do not append this op into graph. - Only apply this stragety when double backward. * Update some double backward op. * Update sum_op to judge whether a tensor is empty by numel or IsInitialized().	6 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	6 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.	6 years ago
Zeng Jinle	0c335dcd2c	Make conv cudnn workspace size configurable (#17036 ) * make_conv_cudnn_ws_size_configurable, test=develop * change std::max to std::min test=develop	6 years ago
tink2123	e0f7bf4f2f	polish the code test=develop	6 years ago
tink2123	ffe81af073	modified infer shape test=develop	6 years ago
sneaxiy	7000ec85d9	fix some op grad maker fix ctest eager deletion disable bug test=develop	6 years ago
Wojciech Uss	2579ade45f	Add cpu_quantize_pass for C-API quantization (#16127 ) * Add cpu_quantize_pass for C-API quantization test=develop * add cpu_quantize_pass test * fix lint: add include memory unorderd_map and unordered_set test=develop * fuse_relu 1 test=develop * tuned 2 without squash * fixes test=develop * remove unused vars test=develop * refactored test=develop * fix lint c-style cast -> C++ style cast test=develop * remove QuantMax and c style casts test=develop * last usage of QuantMax removed test=develop * Fix Analysis Predictor UT Check if memory_optimize_pass has already been added to the analysis config before adding a new one, so that it is not added multiple times. test=develop * change map to unordered_map fix the forgotten part of cpu_quantize_pass_tester.cc test=develop * removed quantized attribute * fixed cpu_quantize_pass_tester and op attr comments test=develop * removed redundant line test=debug * removed gmock test=develop * fix after merge	6 years ago
liangan1	4acc522087	Enable function coverage for U8/S8 ConvMKLDNNOpKernel test=develop	6 years ago
Xin Pan	5eb87506bc	add per kernel config and remove const_cast. test=develop	6 years ago
chengduo	ad61e1b22c	fix potential bug (#15688 ) test=develop	6 years ago
Dun	9f8f0fc2d3	Memory optimization of depthwise conv op and group norm op (#15313 ) * mem opt * test=develop * test=develop * test=develop * test=develop * test=develop * test=develop * test=develop * refine code test=develop * refine code test=develop * refine code test=develop * refine code test=develop * refine with cub test=develop * fix mkldnn test && remove comments && test=develop * polish code && test=develop * add only_forward test && test=develop	6 years ago
xiaolil1	bbc9336878	Enable basic MKL-DNN INT8 Conv OP (#15124 ) * Enable basic MKL-DNN INT8 Conv OP test=develop * Modify test case test=develop * Clean unittest code test=develop * Fix test test=develop * Modify test test=develop * Modify basic INT8 Conv test=develop	6 years ago
Yu Yang	bacf1d2399	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/tensor_type	6 years ago
Yan Chunwei	a985949be9	Fea/fuse conv elementwise add fuse (#14669 )	6 years ago
Yu Yang	9bd70a1e04	Change tensor uses proto::VarType::type test=develop	6 years ago
Yihua Xu	155328a488	Clean Code test=develop	6 years ago
Yihua Xu	65dbc7cca4	Merge branch 'develop' into develop_4f71a6ee2_conv3d_mkldnn_opt	6 years ago
Xin Pan	82d68281c0	follow comments test=develop	6 years ago
Xin Pan	41c28d54c6	allow customize kernel selection test=develop	6 years ago
Yihua Xu	669191c9cc	Implement conv3d with mkldnn library (test=develop)	6 years ago
qingqing01	fd7e643153	Convolution fusion operator. (#14449 ) * Convolution fusion operator. * Clean code test=develop	6 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	6 years ago
chengduo	6c6e638550	Add InferVarType for some op (#14201 ) * add_infer_var_type test=develop * InferVarTypeHelper-> VarTypeInferenceHelper test=develop * PassInputTypeAndDTypeOnOutput test=develop * follow comment test=develop	6 years ago
qingqing01	abe209234f	Exhaustive search for cuDNN conv. (#14286 ) * exhaustive search for cuDNN conv. * Refine code and add unit testing. * Fix model load in fluid/inference and unit testing in conv2d * Follow comments. * Fix compiling test=develop	6 years ago
qingqing01	db8c52da5e	Revert " Exhaustive search for cuDNN conv. (#14043 )" This reverts commit `ce7d9b0799`.	6 years ago
qingqing01	ce7d9b0799	Exhaustive search for cuDNN conv. (#14043 ) * exhaustive search for cuDNN conv. * Refine code and add unit testing. * Clean code * Fix model load in fluid/inference and unit testing in conv2d * Follow comments.	6 years ago
Xin Pan	8f2116d8fa	clean up after the changes have been stopped for so long. test=develop	6 years ago
Tomasz Patejko	4be45af1cc	MKLDNN conv + elementwise_add fusion: skip connection attribute renamed. Comments about patterns added. test=develop	6 years ago
Tomasz Patejko	b8e54ab5cc	MKLDNN conv + elementwise_add fusion: parameter name changed to ResidualData	6 years ago
Tomasz Patejko	41f3d78fdf	MKLDNN conv + elementwise_add fusion: output and elemwise param share data in conv primitive. Output is properly allocated	6 years ago
Tomasz Patejko	56528531ea	MKLDNN conv + elementwis_add fusion: initial work on passing eltwise data to conv primitive	6 years ago
Michał Gallus	8cbefd1a43	Fuse Conv+BN+SkipConnectionAdd+ReLU with transpiler temporarily (#13350 )	7 years ago
Jacek Czaja	dfbd1cc3c1	Merge pull request #13209 from Sand3r-/mgallus/conv-relu-fuse [MKLDNN] Fuse Conv+BatchNorm + ReLU	7 years ago
Michal Gallus	5d34ef61cb	Fuse MKLDNN's Conv + ReLU	7 years ago
Krzysztof Binias	1658958fe6	Reusing converted weights	7 years ago
Michał Gallus	cd32ddac12	Fuse Convolution and Eltwise Add into MKLDNN's Conv+Bias (#12669 ) * Fuse Convolution and Eltwise Add into Conv+Bias * Reduce bias branching at conv_mkldnn_op * Add MKLDNN build checks for Conv Bias * Conv-bias: check if bias input exist befor assignment * Conv-bias: Remove Bias dim check from infershape It was causing conv3d test to crash upon\ncalling HasInput(Bias)	7 years ago
tensor-tang	647c0eb7ec	Merge pull request #11099 from mozga-intel/mozga-intel/Conv_mkldnn_layout MKLDNN layout: Support for convolution operator	7 years ago
dzhwinter	bfa3fd6f15	add inplace attribute to op_proto_maker (#10665 ) * "add inplace attribute" * "register inplace attribute" * "change se-next model for memory-reuse" * "fix typo" * repick * fix merge conflict * "fix stupid error"	7 years ago
mozga-intel	9908d3cfbc	MKLDNN layout: Support for convolution operator	7 years ago
mozga-intel	3ff9ba0e6b	Mkldnn layout (#11040 ) * Add MKLDNN layout support in Paddle Add MKLDNN layout in Paddle so that MKLDNN friendly memory layout can be used in MKLDNN enabled OP kernel. Before this commit, NCHW is hardcode to be used in all MKLDNN op kernels. As a result, non-optimized execution path is selected in MKLDNN primitive which bring worse performance. Besides framework change, three MKLDNN OP kernels were updated for using new MKLDNN layout. They are conv/pool2d/batch_norm. Other MKLDNN OP kernels need be also updated in similar way to achieve best performance. * Add MKLDNN layout support in activation OP * Don't populate layout from input to output when kMKLDNN in * Refine pool mkldnn op kernel * MKLDNN layout * Remove the inferitance from tensor file * MKLDNN layout: refactoring * Remove additional #define to register new operator * Prepare mkldnn tests to work with layout	7 years ago
Yu Yang	0e78cb69fb	Clean OpProtoAndCheckerMaker Do not use ctor * Reduce line of codes. * We can use virtual function for Maker now. * The implementation does not care what maker holds, it is easier to refactor later.	7 years ago
Yang Yang(Tony)	e04c43d543	add semicolon to op registry (#10034 ) * script to add semicolon * fix typo	7 years ago
Yang Yang	ce7c2e86e5	script to fix all	7 years ago
Yi Wang	54316bdd7e	Update	7 years ago
Liu Yiqun	371c53f88c	Add profiling event in feed, fetch and load op.	7 years ago
Kexin Zhao	e4de5dc347	add conv2d fp16 support	7 years ago
pzelazko-intel	8c71adaa8c	MKLDNN conv2d kernel added (#8451 ) * MKLDNN conv2 OP kernel added * TODOs added * mkldnn conv2d OP refactor * CanCUDNNBeUsed and CanMKLDNNBeUsed moved	7 years ago

1 2

56 Commits (112cf850b7bc403b5ffea655898d3a0d3d4a4921)