Table of Contents  ▸  List of Commands  ▸  Neural Networks  ▸  nn_parameters

◀  nn_normalize    |    nn_patchcolumn2d  ▶

nn_parameters

Arguments:

• out,width>0,_height>0,_depth>0,_spectrum>0,_learning_mode,_weight_decay>=0,_initialization

Description:

Add a parameters layer (4D grid of learnable parameters) to the current network.

learning_mode can be { 0:No learning | 1:Weights }.
initialization can be { 0:Zero | 1:Identity | 2:Lecun-initialization | 3:He-initialization 4:Residual He-initialization }.

Default values:

height=depth=spectrum=1, learning_mode=1, weight_decay=0 and initialization=3.

G'MIC is an open-source software distributed under the CeCILL free software licenses (LGPL-like and/or
GPL-compatible). Copyrights (C) Since July 2008, David Tschumperlé - GREYC UMR CNRS 6072, Image Team.