ebl::nn_layer_full Class Reference

#include <EblLayers.h>

Inheritance diagram for ebl::nn_layer_full:

List of all members.

Public Member Functions
	nn_layer_full (parameter &p, intg indim0, intg noutputs)
void	fprop (state_idx &in, state_idx &out)
	fprop from in to out
void	bprop (state_idx &in, state_idx &out)
	bprop
void	bbprop (state_idx &in, state_idx &out)
	bbprop
void	forget (forget_param_linear &fp)
	initialize the weights to random values
void	display_fprop (state_idx &in, state_idx &out, unsigned int &h0, unsigned int &w0, double zoom, bool show_out=false)
	display fprop at (h0, w0)
Public Attributes
linear_module_replicable	linear
	linear module for weight matrix
addc_module	adder
	bias vector
tanh_module	sigmoid
	the non-linear function
state_idx *	sum
	weighted sum

---

Detailed Description

a simple fully-connected neural net layer: linear + tanh non-linearity.

---

Constructor & Destructor Documentation

ebl::nn_layer_full::nn_layer_full	(	parameter &	p,
		intg	indim0,
		intg	noutputs
	)

constructor. Arguments are a pointer to a parameter in which the trainable weights will be appended, the number of inputs, and the number of outputs.

---

The documentation for this class was generated from the following files:

• /media/e/eblearn-trunk/libeblearn/include/EblLayers.h
• /media/e/eblearn-trunk/libeblearn/src/EblLayers.cpp