Fast sigmoidal function, which does not need to compute an exponential function. More...
#include <shark/Models/NeuronLayers.h>
Public Types | |
typedef EmptyState | State |
Public Member Functions | |
template<class Arg > | |
void | evalInPlace (Arg &arg) const |
template<class Arg > | |
void | evalInPlace (Arg &arg, State &) const |
template<class Output , class Derivative > | |
void | multiplyDerivative (Output const &output, Derivative &der, State const &state) const |
Fast sigmoidal function, which does not need to compute an exponential function.
It is defined as
\[ f(x)=\frac x {1+|x|}\]
it's derivative can be computed as
\[ f'(x)= (1 - |f(x)|)^2 \]
Definition at line 92 of file NeuronLayers.h.
Definition at line 93 of file NeuronLayers.h.
|
inline |
Definition at line 95 of file NeuronLayers.h.
|
inline |
Definition at line 100 of file NeuronLayers.h.
References shark::TanhNeuron::evalInPlace().
|
inline |
Definition at line 105 of file NeuronLayers.h.