5 Commits (8894c67d7168dfa5f5dc8e57ec2b5c60f24e368c)

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`	7 years ago
zhouxiao-coder	496583a165	update data types and test cases	7 years ago
zhouxiao-coder	e68a217f34	Add optional inputs and outputs to enable updating;Add weight to match original implementation	7 years ago
zhouxiao-coder	9b0f092853	change input data type	7 years ago
zhouxiao-coder	c22f7fcd17	add positive_negative_pair_op evaluator	7 years ago