Neuron Class Reference

Represents a single neuron in the neural network. More...

#include <NeuralNetwork.hpp>

Public Member Functions
	Neuron (unsigned numOutputs, unsigned index)
	Constructs a Neuron. More...
void	setOutputVal (double val)
	Sets the output value of the neuron. More...
double	getOutputVal () const
	Gets the output value of the neuron. More...
void	feedForward (const std::vector< Neuron > &prevLayer)
	Feeds forward the input values to the next layer. More...
void	calcOutputGradients (double targetVal)
	Calculates the output gradients for the output layer. More...
void	calcHiddenGradients (const std::vector< Neuron > &nextLayer)
	Calculates the hidden gradients for hidden layers. More...
void	updateInputWeights (std::vector< Neuron > &prevLayer)
	Updates the input weights for the neuron. More...

Detailed Description

Represents a single neuron in the neural network.

Constructor & Destructor Documentation

Neuron()

Neuron::Neuron	(	unsigned	numOutputs,
		unsigned	index
	)

Constructs a Neuron.

Parameters

numOutputs	The number of outputs from this neuron.
index	The index of this neuron in its layer.

Member Function Documentation

calcHiddenGradients()

void Neuron::calcHiddenGradients

(

const std::vector< Neuron > &

nextLayer

)

Calculates the hidden gradients for hidden layers.

Parameters

nextLayer

The next layer of neurons.

calcOutputGradients()

void Neuron::calcOutputGradients

(

double

targetVal

)

Calculates the output gradients for the output layer.

Parameters

targetVal

The target value.

feedForward()

void Neuron::feedForward

(

const std::vector< Neuron > &

prevLayer

)

Feeds forward the input values to the next layer.

Parameters

prevLayer

The previous layer of neurons.

getOutputVal()

double Neuron::getOutputVal

(

)

const

Gets the output value of the neuron.

Returns

The output value.

setOutputVal()

void Neuron::setOutputVal

(

double

val

)

Sets the output value of the neuron.

Parameters

val	The value to set.

updateInputWeights()

void Neuron::updateInputWeights

(

std::vector< Neuron > &

prevLayer

)

Updates the input weights for the neuron.

Parameters

prevLayer

The previous layer of neurons.

---

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

• ml_library_include/ml/neural_network/NeuralNetwork.hpp