7 Commits (2ffa3a8bf6a7cb0e3d5e1ac211417c234ab04f04)

Author	SHA1	Message	Date
xzl	2ffa3a8bf6	rename op to depthwise_conv2d, more efficient	7 years ago
xzl	aece290599	add depthwise gpu forward, backward, test, interface	7 years ago
zlx	3772d27dfb	add depthwise conv forward	7 years ago
Luo Tao	761b329793	unify the indentation of license	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	c359e39b59	add double type kernel	8 years ago
dangqingqing	f5e367655e	Use G++ to compile some cu operators.	8 years ago