MultiTaskKernel< InputTypeT > Class Template Reference
Special kernel function for multi-task and transfer learning. More...
#include <shark/Models/Kernels/MultiTaskKernel.h>
Inheritance diagram for MultiTaskKernel< InputTypeT >:
Public Types | |
| typedef AbstractKernelFunction< InputTypeT > | InputKernelType |
Public Member Functions | |
| MultiTaskKernel (InputKernelType *inputkernel, DiscreteKernel *taskkernel) | |
| Constructor. More... | |
| std::string | name () const |
| From INameable: return the class name. More... | |
Detailed Description
template<class InputTypeT>
class MultiTaskKernel< InputTypeT >
Special kernel function for multi-task and transfer learning.
- This class is a convenience wrapper for the product of an input kernel and a kernel on tasks. It also encapsulates the projection from multi-task learning data (see class MultiTaskSample) to inputs and task indices.
Definition at line 306 of file MultiTaskKernel.h.
Member Typedef Documentation
§ InputKernelType
template<class InputTypeT>
| typedef AbstractKernelFunction<InputTypeT> MultiTaskKernel< InputTypeT >::InputKernelType |
Definition at line 314 of file MultiTaskKernel.h.
Constructor & Destructor Documentation
§ MultiTaskKernel()
template<class InputTypeT>
|
inline |
Constructor.
- Parameters
-
inputkernel kernel on inputs taskkernel kernel on task indices
Definition at line 319 of file MultiTaskKernel.h.
Member Function Documentation
§ name()
template<class InputTypeT>
|
inline |
From INameable: return the class name.
Definition at line 327 of file MultiTaskKernel.h.
The documentation for this class was generated from the following file:
- include/shark/Models/Kernels/MultiTaskKernel.h