29 Commits (8e02870ce2780c662802caf8ce0c5fd34191d25b)

Author	SHA1	Message	Date
dzhwinter	5ad1aef051	"cudnn operators change to cudnn kernel" (#6660) ... * "unified operators" * "add CUDNN register" * "add use cudnn attribute" * "add attribute" * "test conv tranpose op" * "remove duplicated attr" * "fix op test" * "add attribute to set cudnn" * "add more log" * "need layout op register support" * "add more log" * "change GetExpectedKernelType " * "fix Get attr in conv_op" * "fix CI" * "fix tests" * "removed kernel priority fallback" * "fix CI" * "fix stack pointer bug" * "refine buggy interface" * "add const cast to save life" * "fix get_output_with_grad" * "fix op test with dataformat" * ""fix pooling * "fix pooling test" * "fix CI" * "fix with_gpu error" * "add transform needed functional check" * "fix unpack list error" * "comment out parallel.do temporary" * "fix CI" * "fix compile doc error" * "make threshold larger"	7 years ago
Yiqun Liu	b5fda2723f	Port WarpCTC Operator (#5107) ... * Add Seq2BatchFunctor, which will be used in WarpCTCOp. * Implement WrapCTCFunctor and WrapCTCKernel. * Add unittest of warpctc_op. * Modify the check_output inferface in python unittest framework to allow check a subset of outputs. * Use absolute offset lod in warpctc_op and related functors. * Refine the comments of warpctc_op. * The new python unittest supports checking a subset of the outputs, so revoke the previous change. * Rename the transform from LoDTensor to Tensor with shape [max_sequence_length, num_sequences, sequence_width] to PaddingSequenceFunctor. * Update to the newest codes. * Rename the PaddingSequenceFunctor to PaddingLoDTensorFunctor and remove the computation of dimensions out of the functos.	7 years ago
Luo Tao	761b329793	unify the indentation of license	7 years ago
dzhwinter	735eba2976	Feature/operator run place (#6783) ... * "change operator interface" * "move devicepool to device_context" * "fix operator test" * "fix op_registry Run interface" * "net op passed. Need to fix nccl multi-Context" * "add nccl group function" * "add nccl group function" * "fix gpu count exceed 32 error" * "fix recurrent op, nccl op" * "change the other operators interface with Place" * "fix typo" * "fix pybind" * "fix device in python side" * "fix pybind failed" * "add init for test" * "fix CI"	7 years ago
Yu Yang	d5cab4f07c	Fix compile on CUDA9.1 & MacOS (#6642)	7 years ago
Yu Yang	f291abfc53	Add HasCUDNN to detect if CUDNN is installed or not (#6349) ... * Add HasCUDNN to detect if CUDNN is installed or not * Fix CI	7 years ago
武毅	4ecbab42d8	Fix compile on cudnn7 (#5982) ... * fix compile on cudnn7 * update * update * make silent	7 years ago
Qiao Longfei	c9172c1cb3	Make enforce target (#5889) ... * make enforce a target and dependent on nccl when gpu is enabled * add some more dependency	7 years ago
dangqingqing	524ccba4fe	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into cmake_speed	8 years ago
dangqingqing	f5e367655e	Use G++ to compile some cu operators.	8 years ago
emailweixu	2378679a9e	Fix a dead lock bug for dyload/nccl.h when nccl lib cannot be loaded (#5533) ... It caused by a bug of std::call_once described in https://stackoverflow.com/questions/41717579/stdcall-once-hangs-on-second-call-after-callable-threw-on-first-call. It is likely caused by a deeper bug of pthread_once, which is discussed in https://patchwork.ozlabs.org/patch/482350/	8 years ago
Qiao Longfei	56b723c40d	Cudnn batch norm op (#5067) ... * init cudnn batch norm op * rename batch_norm_cudnn_op.cc batch_norm_op.cu * correct name style * add ExtractNCWHD, simplify code * fix ExtractNCWHD * use CUDNN_ENFORCE instead of PADDLE_ENFORCE	8 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
Markus Kliegl	164898277c	MatMul operator (#4856) ... * initial matmul operator Similar to np.matmul, but also has transpose_X and transpose_Y flags, and only supports tensors from rank 1 to 3 inclusive. For GPU, uses cublas?gemmStridedBatched. For CPU, uses cblas_?gemm_batch if available via MKL; otherwise a simple serial implementation that loops over the batch dimension is employed for now.	8 years ago
dangqingqing	207132226c	Add unit testing for cuDNN wrapper.	8 years ago
qijun	95df4700d4	Merge remote-tracking branch 'baidu/develop' into port_blas	8 years ago
Yu Yang	d299528829	Add curandGenerateNormal to curand.h	8 years ago
qijun	688c43b104	format code	8 years ago
Yu Yang	4591110201	Fix gaussian_random_op compile error ... * Should always use `dynload::` for cuda function. * Fix cublas.h without DSO load.	8 years ago
qijun	5f1081d83d	fix bug in dynload	8 years ago
liaogang	b58725bd51	Add cpplint for *.h and cuda *.cu	8 years ago
liaogang	f812de2cce	ENH: unify PADDLE_ENFORCE	8 years ago
qijun	4e918377d0	fix bug in dynload	8 years ago
qijun	14d2c3990f	split device_context	8 years ago
qijun	69d76812ae	fix cublas dynload bug	8 years ago
Yu Yang	a046605307	Refine CUDA Related libraries	8 years ago
qijun	e6fcdd47e5	fix wrong including header-file in files in paddle/platform/dynload dir	8 years ago
qijun	3567ea6d7c	move to dynload directory	8 years ago