5 Commits (f58fe6d3edb78c96547d91bf9da1492d0709b63d)

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
QI JUN	d7bf372d26	support adagrad sparse update (#5272) ... * adam sparse support * fix gpu build error * fix ci * fix ci * fix adagrad sparse update bug * fix gpu build error	7 years ago
Kexin Zhao	78f4c803f3	change learning rate and fix format	7 years ago
Kexin Zhao	d1de7ec630	Change learning rate from attribute to input tensor	7 years ago
Kexin Zhao	1ac654a69f	Implementing the Adagrad optimizer step operator	7 years ago