6 Commits (8823a12ef85de1f56bc6a378d402602dfe04782c)

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
Yang Yang(Tony)	5682916c47	add acknowledgement of lstm_unit_op (#5293)	8 years ago
Yu Yang	f60f0eae11	Using double precision to stablize lstm gradient check	8 years ago
Yu Yang	3a5693e0a8	Add Skeleton of Double support	8 years ago
zchen0211	757c32f20b	lstm unit gpu	8 years ago
zchen0211	2b10d322e5	lstm kernels	8 years ago