|
cMHN 1.2
C++ library for learning MHNs with pRC
|
Class storing all relevant information for a regulator. More...
#include <regulator.hpp>
Public Member Functions | |
| ~Regulator ()=default | |
| Regulator (Regulator const &)=default | |
| Regulator (Regulator &&)=default | |
| Regulator & | operator= (Regulator const &) &=default |
| Regulator & | operator= (Regulator &&) &=default |
| Regulator ()=delete | |
| Regulator (auto const &score, auto const &grad, auto const &name, auto const &lambda=T(1e-2)) | |
| auto | score (pRC::Tensor< T, D, D > const &theta) const |
| auto | grad (pRC::Tensor< T, D, D > const &theta) const |
| auto & | lambda () |
| auto const | lambda () const |
| auto | name () const |
Class storing all relevant information for a regulator.
Member functions score(theta) and grad(theta) return the score and the gradient wrt the entries of the theta matrix. The member function name() accesses the name given to the regulator. The member function lambda() can be used to get and set the regularization strength.
Note that the lambda functions accessed through score() and grad(), as well as the regulators name accessed through name() can not be changed, only the regularization strength can be changed.
| T | type used for calculations |
| D | number of events of the dataset |
|
delete |
|
inlineexplicit |
|
inline |
|
inline |
|
inline |
|
inline |
|
inline |