8 Commits (608ebece366e24257fea70bff8cab007473f10b4)

Author	SHA1	Message	Date
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`	8 years ago
dangqingqing	2673657684	Move RowwiseAdd functor to math_funcion and Add ColwiseSum functor.	8 years ago
dangqingqing	f5e367655e	Use G++ to compile some cu operators.	8 years ago
dangqingqing	0c70bd28aa	Enable initial hidden state and cell state in LSTM Operator.	8 years ago
Yu Yang	65906ef1d0	Several Enhancement	8 years ago
dangqingqing	2a8dbd130d	LSTM Operator forward implementation.	8 years ago
dangqingqing	3cace73701	Add lstm implementation.	8 years ago
dangqingqing	8728b3cce2	Add LSTM Operators.	8 years ago