#include <LogSoftmaxVariable.h>
Inheritance diagram for PLearn::LogSoftmaxVariable:
Public Member Functions | |
LogSoftmaxVariable () | |
Default constructor for pestilence. | |
LogSoftmaxVariable (Variable *input) | |
PLEARN_DECLARE_OBJECT (LogSoftmaxVariable) | |
virtual void | recomputeSize (int &l, int &w) const |
Recomputes the length l and width w that this variable should have, according to its parent variables. | |
virtual void | fprop () |
compute output given input | |
virtual void | bprop () |
virtual void | bbprop () |
compute an approximation to diag(d^2/dinput^2) given diag(d^2/doutput^2), with diag(d^2/dinput^2) ~=~ (doutput/dinput)' diag(d^2/doutput^2) (doutput/dinput) In particular: if 'C' depends on 'y' and 'y' depends on x ... | |
virtual void | symbolicBprop () |
compute a piece of new Var graph that represents the symbolic derivative of this Var | |
Private Types | |
typedef UnaryVariable | inherited |
|
Reimplemented from PLearn::UnaryVariable. Definition at line 54 of file LogSoftmaxVariable.h. Referenced by LogSoftmaxVariable(). |
|
Default constructor for pestilence.
Definition at line 58 of file LogSoftmaxVariable.h. |
|
Definition at line 55 of file LogSoftmaxVariable.cc. References inherited. |
|
compute an approximation to diag(d^2/dinput^2) given diag(d^2/doutput^2), with diag(d^2/dinput^2) ~=~ (doutput/dinput)' diag(d^2/doutput^2) (doutput/dinput) In particular: if 'C' depends on 'y' and 'y' depends on x ... d^2C/dx^2 = d^2C/dy^2 * (dy/dx)^2 + dC/dy * d^2y/dx^2 (diaghessian) (gradient) Reimplemented from PLearn::Variable. Definition at line 89 of file LogSoftmaxVariable.cc. References PLERROR. |
|
Implements PLearn::Variable. Definition at line 77 of file LogSoftmaxVariable.cc. References k, PLearn::Variable::nelems(), PLearn::safeexp(), and PLearn::sum(). |
|
compute output given input
Implements PLearn::Variable. Definition at line 70 of file LogSoftmaxVariable.cc. References PLearn::log_softmax(). |
|
|
|
Recomputes the length l and width w that this variable should have, according to its parent variables. This is used for ex. by sizeprop() The default version stupidly returns the current dimensions, so make sure to overload it in subclasses if this is not appropriate. Reimplemented from PLearn::Variable. Definition at line 60 of file LogSoftmaxVariable.cc. References PLearn::Var::length(), and PLearn::Var::width(). |
|
compute a piece of new Var graph that represents the symbolic derivative of this Var
Reimplemented from PLearn::Variable. Definition at line 96 of file LogSoftmaxVariable.cc. References PLERROR. |