Layer::FullConnectedLayer Class Reference

#include <full_connected.hpp>

Inheritance diagram for Layer::FullConnectedLayer:

Public Member Functions
	FullConnectedLayer (size_t n, size_t m)
	~FullConnectedLayer ()
void	init_standard ()
void	init (Real mu, Real sigma)
Vector	feed_forward (Vector x) override
void	init_nabla () override
Vector	back_propagation (Vector e) override
void	update (Real eta) override
Public Member Functions inherited from Layer::Layer
	Layer (size_t n, size_t m)
	~Layer ()
size_t	get_input_size () const
size_t	get_output_size () const
Vector	get_output () const
virtual Vector	feed_forward (Vector x)=0
virtual void	init_nabla ()=0
virtual Vector	back_propagation (Vector e)=0
virtual void	update (Real eta)=0

Public Attributes
Vector	b
Vector	w
Vector	nabla_b
Vector	nabla_w
Public Attributes inherited from Layer::Layer
string	name
size_t	n
size_t	m
Vector	x
Vector	y
Vector	d

Constructor & Destructor Documentation

FullConnectedLayer()

Layer::FullConnectedLayer::FullConnectedLayer	(	size_t	n,
		size_t	m
	)

~FullConnectedLayer()

Layer::FullConnectedLayer::~FullConnectedLayer

(

)

Member Function Documentation

back_propagation()

Vector Layer::FullConnectedLayer::back_propagation ( Vector  e )

overridevirtual

Apply back propagation algorithm on the delta output vector d. Used the input vector stored in x_in_ref during feedforward. Return a reference to the computed (and stored) input delta vector. Nabla vectors must be computed here.

Implements Layer::Layer.

feed_forward()

Vector Layer::FullConnectedLayer::feed_forward ( Vector  x )

overridevirtual

Apply the layer to the input vector x. Vectors x_in_ref and x_out must be updated in consequence. Return a reference to x_out.

Implements Layer::Layer.

init()

void Layer::FullConnectedLayer::init	(	Real	mu,
		Real	sigma
	)

init_nabla()

void Layer::FullConnectedLayer::init_nabla ( )

overridevirtual

Initialize nabla vectors which are used during gradient descent.

Implements Layer::Layer.

init_standard()

void Layer::FullConnectedLayer::init_standard

(

)

update()

void Layer::FullConnectedLayer::update ( Real  eta )

overridevirtual

Update layer parameters using gradient descent algorithm with learning rate eta.

Implements Layer::Layer.

Member Data Documentation

b

Vector Layer::FullConnectedLayer::b

nabla_b

Vector Layer::FullConnectedLayer::nabla_b

nabla_w

Vector Layer::FullConnectedLayer::nabla_w

w

Vector Layer::FullConnectedLayer::w

---

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

• src/layers/full_connected.hpp
• src/layers/full_connected.cpp