Configuration class that controls neural net "shrinkage" which is actually a scaling on the parameters of each of the updatable layers. More...
#include <shrink-nnet.h>
Public Member Functions | |
NnetShrinkConfig () | |
void | Register (OptionsItf *opts) |
Public Attributes | |
int32 | num_bfgs_iters |
BaseFloat | initial_step |
Configuration class that controls neural net "shrinkage" which is actually a scaling on the parameters of each of the updatable layers.
Definition at line 33 of file shrink-nnet.h.
|
inline |
Definition at line 41 of file shrink-nnet.h.
|
inline |
Definition at line 42 of file shrink-nnet.h.
References OptionsItf::Register(), and kaldi::nnet2::ShrinkNnet().
BaseFloat initial_step |
Definition at line 39 of file shrink-nnet.h.
Referenced by kaldi::nnet2::ShrinkNnet().
int32 num_bfgs_iters |
Definition at line 34 of file shrink-nnet.h.
Referenced by kaldi::nnet2::ShrinkNnet().