21 Commits (8894c67d7168dfa5f5dc8e57ec2b5c60f24e368c)

Author	SHA1	Message	Date
QI JUN	7333df8510	fix pool_op bug (#7879)	7 years ago
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
Yu Yang	e445b3ff20	Move framework.proto to proto namespace (#6718) ... * Move framework.proto to proto namespace * Fix compile * Fix compile * Fix Compile	7 years ago
QI JUN	61ec0b9516	Refine device context (#6433) ... There are mainly following fixes: - take `DeviceContext` as the template parameter of math functors and OpKernel instead of `Place` - remove `eigen_device` interface in base class `DeviceContext` - remove `GetEigenDevice` interface in `ExecutionContext` and base class `DeviceContext` - remove unused `platform::EigenDeviceConverter` - rename `REGISTER_OP_GPU_KERNEL` to `REGISTER_OP_CUDA_KERNEL` - rename `USE_GPU_ONLY_OP` to `USE_CUDA_ONLY_OP`	7 years ago
chengduoZH	2160497786	fix pooling functor parameter order	7 years ago
chengduoZH	cdf5e87104	fix attr name	7 years ago
chengduoZH	6bdf5c1417	fix bug	7 years ago
chengduoZH	0ab012cf7f	fix doc	7 years ago
Yu Yang	73a8b78a72	Correct OpWithKernel's infershape (#4847) ... They are public now	7 years ago
Yu Yang	75d0c79015	Change Name convention of operator attributes (#4807) ... * Change dataType to data_type Follow PEP8 * Change name_convention to fit PEP8	7 years ago
chengduoZH	6db476ed89	Separate the declarations and implementation of the PoolOp and PoolMaker class in order to reuse in pool_cudnn	7 years ago
chengduoZH	df59889984	remove conflict	8 years ago
chengduoZH	e1e3859e88	remove custom attr checker and fix code format	8 years ago
chengduoZH	4b06d8db91	fix globalPooling type (int => bool)	8 years ago
chengduoZH	dfc8d3c1c1	Fix (According to the review)	8 years ago
chengduoZH	f6e69d7412	fix maxpool backward functor	8 years ago
chengduoZH	b72854389e	Fix (According to the review)	8 years ago
chengduoZH	0417e4e4bf	fix framework::LoDTensor => Tensor	8 years ago
chengduoZH	84a2512b90	fix parameter name and function define	8 years ago
chengduoZH	3416f5e0f8	fix function define	8 years ago
chengduoZH	a8477381ac	Add pooling2d(max, ave) and pooling3d(max, ave) Op	8 years ago