50 Commits (3f4aca618f4ef0b87881c4a7ae62af2490e35780)

Author	SHA1	Message	Date
Yan Chunwei	7796f65f89	fix inference on gpu out of mem (#14414) ... * fix inference on gpu out of mem the transfer logic in operator.cc will keep creating new scopes.	6 years ago
tensor-tang	94ab65d591	disable avx2 and avx512 flag ... test=develop	6 years ago
dzhwinter	b9fcf8e677	"configure" (#13539)	7 years ago
dzhwinter	dbd7896678	cmakelist windows (#12927) ... * picked pr * "fix ci"	7 years ago
minqiyang	d214dff13c	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_anakin_in_manylinux1	7 years ago
minqiyang	8f8560744a	Reorder the cmake import and add CUDNN_INCLUDE_DIR into anakin cmake module	7 years ago
Yihua Xu	084d4a9e9e	Optimize CRF Decoding with AVX/AVX2/AVX512F instruction (#12767) ... * Optimize CRF decoding with AVX/AVX2 instruction * Enable the AVX2 flags for compiling * Clean the code and decrease the count of multiply calculation * Add the support of AVX512 instruction to optimize CRF Decoding * Clean the code * Enable the AVX512f flags for compiling * Clean the code for the invaluable switch * Fixed the issue to check AVX512F status * Clean the code * Add some explanation of the key points	7 years ago
dzhwinter	00463fdfe3	cudnn windows support (#12757) ... * cudnn widndows * "add comment" * "windows support" * "fix cmake error"	7 years ago
minqiyang	bfe8d6fa66	Polish code	7 years ago
minqiyang	e10e0d4a5b	Fix anakin build problem in manylinux1 docker image	7 years ago
luotao1	413bf9d494	disable anakin when cuda < 8.0 or cudnn < 7.0	7 years ago
dzhwinter	d7873e1412	remove patchelf in windows (#12710) ... * remove patchelf in windowls * "follow comment"	7 years ago
luotao1	3373535b21	fix specific cudnn include and library path	7 years ago
Luo Tao	e8aa6d1283	add anakin compiler from github source code	7 years ago
gongweibao	66c91911cf	Improve brpccmake (#11842)	7 years ago
Wu Yi	34865f2de3	Trainer send term signal (#11220) ... * wip * use executor.complete to end trainer * fix build * fix build with distribute off * fix typo * fix cmake typo * fix build	7 years ago
gongweibao	d9de6b8621	Add brpc surpport. (#11263)	7 years ago
Luo Tao	e116129f03	rewrite unittest of trt_activation_op	7 years ago
Houjiang Chen	83f4e9e9a6	enable eigen multi-threads on mobile device (#10938)	7 years ago
Luo Tao	d4682247e1	auto find tensorrt library	7 years ago
sabreshao	45c988d86a	Demostration of cmake refine for HIP support. ... 1. Add option WITH_AMD_GPU. 2. Add cmake/hip.cmake for HIP toolchain. 3. Some external module such as eigen may need HIP port. 4. Add macro hip_library/hip_binary/hip_test to cmake/generic.cmake. 5. Add one HIP source concat.hip.cu as an example. Each .cu may have its corresponding .hip.cu.	7 years ago
Yu Yang	22b5c07a7d	Fix the compilation on CUDA 9.1/GCC 5.3 ... * Make CUPTI_LIB_PATH not passing by macro. * Add missing header	7 years ago
Xin Pan	b9ec24c6e9	Extend current profiler for timeline and more features.	7 years ago
qingqing01	24509f4af9	Fix the grammar in copyright. (#8403)	7 years ago
Kexin Zhao	4901184ee9	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into float16	7 years ago
Kexin Zhao	41bd1f9115	fix gpu test, clean code and add cmake	7 years ago
tensor-tang	8496eab45a	make mklml necessary when with_mkldnn	7 years ago
Yu Yang	94e741d6f0	Use external project for NCCL (#5028)	8 years ago
Yu Yang	43c6ff212e	Feature/nccl dso (#5001) ... * "add nccl enforce" * Dev * Update comment * Add nccl test * Follow comments	8 years ago
Yan Chunwei	843ed8e320	dynamic recurrent op forward c++ implentation (#4597)	8 years ago
Yi Wang	f985700abf	Resolve conflict	8 years ago
Yu Yang	84500f9487	Change `PADDLE_ONLY_CPU` to `PADDLE_WITH_GPU` ... By shell command ```bash sed -i 's#ifdef PADDLE_ONLY_CPU#ifndef PADDLE_WITH_GPU#g' `find ./paddle/ -name '*.h' -o -name '*.cc' -o -name '*.cpp' -o -name '*.c' -o -name '*.cu'` sed -i 's#ifndef PADDLE_ONLY_CPU#ifdef PADDLE_WITH_GPU#g' `find ./paddle/ -name '*.h' -o -name '*.cc' -o -name '*.cpp' -o -name '*.c' -o -name '*.cu'` ```	8 years ago
Yi Wang	5f51d0afc4	Add -D PADDLE_WITH_CUDA in cmake/configure.cmake	8 years ago
hedaoyuan	adcca2cc06	Add PADDLE_USE_EIGEN_FOR_BLAS macro.	8 years ago
liaogang	7a56d46a8a	Rename PROJ_ROOT to PADDLE_SOURCE_DIR and PROJ_BINARY_ROOT to PADDLE_BINARY_DIR	8 years ago
tensor-tang	e71976f221	remove global linker and exe from mkldnn iomp	8 years ago
tensor-tang	577bb4e346	rename mkllite to mklml	8 years ago
tensor-tang	f490d94210	separate MKL_LITE from MKLDNN	8 years ago
tensor-tang	89a4158038	enable MKLDNN library and MKL small package	8 years ago
Helin Wang	fbfbe93a78	cmake: do not run glide install every time.	8 years ago
yi.wu	9c853c269d	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into cmake_go_vendor	8 years ago
Helin Wang	38790c1c21	fix according to comment	8 years ago
qiaolongfei	9e13b68f01	refine code	8 years ago
qiaolongfei	d9aac1e13d	add WITH_Go to disable compile go to paddle	8 years ago
Liu Yiqun	ea5b265fdb	Fix the setting of simd flags in cmake when there is no avx and sse3 support.	8 years ago
Liu Yiqun	1b8564206f	Use native cross-compiling support for Android of cmake.	8 years ago
Liu Yiqun	38fa74edaa	Fix cmake error of failing to find UINT64_MAX.	8 years ago
Liu Yiqun	ccd3d0a42b	Modify cmake for cross-compiling on arm architecture.	8 years ago
liaogang	c24e94c8a4	Check python if system already equipped one	8 years ago
liaogang	9e7f2b8de8	Add system configure	8 years ago