inc\classes\help\classes\NeuralNetwork.php

Show:

inherited

Classes
- \HelpClasses\NeuralNetwork

Table of Contents

Class

Package: Temply-Account\Helpers

\HelpClasses\NeuralNetwork
jump to top

Package: Temply-Account\Helpers

Multi-layer Neural Network in PHP

Loosely based on source code by Phil Brierley, that was translated into PHP by 'dspink' in sep 2005

Algorithm was obtained from the excellent introductory book "Artificial Intelligence - a guide to intelligent systems" by Michael Negnevitsky (ISBN 0-201-71159-1)

Example: learning the 'XOR'-function // Create a new neural network with 3 input neurons, // 4 hidden neurons, and 1 output neuron $n = new NeuralNetwork(3, 4, 1); $n->setVerbose(false);


// Add test-data to the network. In this case,
// we want the network to learn the 'XOR'-function
$n->addTestData(array (-1, -1, 1), array (-1));
$n->addTestData(array (-1,  1, 1), array ( 1));
$n->addTestData(array ( 1, -1, 1), array ( 1));
$n->addTestData(array ( 1,  1, 1), array (-1));
// we try training the network for at most $max times
$max = 3;
// train the network in max 1000 epochs, with a max squared error of 0.01
while (!($success = $n->train(1000, 0.01)) && ++$i<$max) {
echo "Round $i: No success...
";
}
// print a message if the network was succesfully trained
if ($success) {
$epochs = $n->getEpoch();
echo "Success in $epochs training rounds!
";
}
// in any case, we print the output of the neural network
echo "
End result";
for ($i = 0; $i < count($n->trainInputs); $i ++) {
$output = $n->calculate($n->trainInputs[$i]);
echo "
Testset $i; ";
echo "expected output = (".implode(", ", $n->trainOutput[$i]).") ";
echo "output from neural network = (".implode(", ", $output).")\n";
}

The resulting output could for example be something along the following lines:

Success in 719 training rounds! Testset 0; expected output = (-1) output from neural network = (-0.986415991978) Testset 1; expected output = (1) output from neural network = (0.992121412998) Testset 2; expected output = (1) output from neural network = (0.992469534962) Testset 3; expected output = (-1) output from neural network = (-0.990224120384)

...which indicates the network has learned the task.

Author: E. Akerboom
Author: {@link http://www.tremani.nl/ Tremani}, {@link http://maps.google.com/maps?f=q&hl=en&q=delft%2C+the+netherlands&ie=UTF8&t=k&om=1&ll=53.014783%2C4.921875&spn=36.882665%2C110.566406&z=4 Delft}, The Netherlands
License: BSD License
Version: 1.1

Properties

array $controlDataID = array()

DataID control history

Default valuearray()Details

Type: array

array $controlInputs = array()

Input control history

Default valuearray()Details

Type: array

array $controlOutput = array()

Output control history

Default valuearray()Details

Type: array

array $edgeWeight = array()

Edge weights

Default valuearray()Details

Type: array

integer $epoch

NN epoch

Details

Type: integer

float $errorControlset

Control error

Details

Type: float

float $errorTrainingset

Error epoch

Details

Type: float

boolean $isVerbose = true

Is verbose

Default valuetrueDetails

Type: boolean

array<mixed,integer> $layerCount = 0

Layers count

Default value0Details

Type: array<mixed,integer>

array<mixed,float> $learningRate = array(0.1)

Learning rate

Default valuearray(0.1)Details

Type: array<mixed,float>

float $momentum = 0.8

Momentumn

Default value0.8Details

Type: float

array $nodeCount = array()

Amount of nodes

Default valuearray()Details

Type: array

array $nodeThreshold = array()

Threshold of nodes

Default valuearray()Details

Type: array

array $nodeValue = array()

Values of nodes

Default valuearray()Details

Type: array

array $previousWeightCorrection = array()

Fallback position

Default valuearray()Details

Type: array

boolean $success

Success

Details

Type: boolean

array $trainDataID = array()

Training ID

Default valuearray()Details

Type: array

array $trainInputs = array()

Input story

Default valuearray()Details

Type: array

array $trainOutput = array()

Output story

Default valuearray()Details

Type: array

boolean $weightsInitialized = false

Is inite

Default valuefalseDetails

Type: boolean

Methods

__construct(array $nodeCount) : void

Creates a neural network.

Example: // create a network with 4 input nodes, 10 hidden nodes, and 4 output nodes $n = new NeuralNetwork(4, 10, 4);


// create a network with 4 input nodes, 1 hidden layer with 10 nodes,
// another hidden layer with 10 nodes, and 4 output nodes
$n = new NeuralNetwork(4, 10, 10, 4);

// alternative syntax $n = new NeuralNetwork(array(4, 10, 10, 4));

Parameters

Name	Type	Description
$nodeCount	array	The number of nodes in the consecutive layers.

activation(float $value) : float

Implements the standard (default) activation function for backpropagation networks, the 'tanh' activation function.

Parameters

Name	Type	Description
$value	float	The preliminary output to apply this function to

Returns

Type	Description
float	The final output of the node

addControlData(array $input, array $output, integer $id = null) : void

Add a set of control data to the network.

This set of data is used to prevent 'overlearning' of the network. The network will stop training if the results obtained for the control data are worsening.

The data added as control data is not used for training.

Parameters

Name	Type	Description
$input	array	An input vector
$output	array	The corresponding output
$id	integer	(optional) An identifier for this piece of data

addTestData(array $input, array $output, integer $id = null) : void

Add a test vector and its output

Parameters

Name	Type	Description
$input	array	An input vector
$output	array	The corresponding output
$id	integer	(optional) An identifier for this piece of data

backpropagate(array $output, array $desired_output) : void

Performs the backpropagation algorithm. This changes the weights and thresholds of the network.

Parameters

Name	Type	Description
$output	array	The output obtained by the network
$desired_output	array	The desired output

calculate(array $input) : mixed

Calculate the output of the neural network for a given input vector

Parameters

Name	Type	Description
$input	array	The vector to calculate

Returns

Type	Description
mixed	The output of the network

clear() : void

Resets the state of the neural network, so it is ready for a new round of training.

derivativeActivation(float $value) : \HelpClasses\$float

Implements the derivative of the activation function. By default, this is the inverse of the 'tanh' activation function: 1.0 - tanh($value)*tanh($value);

Parameters

Name	Type	Description
$value	float	'X'

Returns

Type	Description
\HelpClasses\$float

export() : void

Exports the neural network

Details

Returns: array

fitLine(array $data) : array

Finds the least square fitting line for the given data.

This function is used to determine if the network is overtraining itself. If the line through the controlset's most recent squared errors is going 'up', then it's time to stop training.

Parameters

Name	Type	Description
$data	array	The points to fit a line to. The keys of this array represent the 'x'-value of the point, the corresponding value is the 'y'-value of the point.

Returns

Type	Description
array	An array containing, respectively, the slope and the offset of the fitted line.

getControlDataIDs() : array

Returns the identifiers of the control data used during the training of the network (if available)

Returns

Type	Description
array	An array of identifiers

getEpoch() : integer

Gets the number of epochs the network needed for training.

Returns

Type	Description
integer	The number of epochs.

getErrorControlSet() : float

Gets the squared error between the desired output and the obtained output of the control data.

Returns

Type	Description
float	The squared error of the control data

getErrorTrainingSet() : float

Gets the squared error between the desired output and the obtained output of the training data.

Returns

Type	Description
float	The squared error of the training data

getLearningRate(integer $layer) : float

Gets the learning rate for a specific layer

Parameters

Name	Type	Description
$layer	integer	The layer to obtain the learning rate for

Returns

Type	Description
float	The learning rate for that layer

getMomentum() : float

Gets the momentum.

Returns

Type	Description
float	The momentum

getRandomWeight( $layer) : float

Gets a random weight between [-0.25 .

. 0.25]. Used to initialize the network.

Parameters

Name	Type	Description
$layer

Returns

Type	Description
float	A random weight

getTestDataIDs() : array

Returns the identifiers of the data used to train the network (if available)

Returns

Type	Description
array	An array of identifiers

getTrainingSuccessful() : boolean

Determines if the training was successful.

Returns

Type	Description
boolean	'true' if the training was successful, 'false' otherwise

import(array $nn_array) : void

Import a neural network

Parameters

Name	Type	Description
$nn_array	array	An array of the neural network parameters

initWeights() : void

Randomise the weights in the neural network

isVerbose() : boolean

Returns whether or not the network displays status and error messages.

Returns

Type	Description
boolean	'true' if status and error messages are displayed, 'false' otherwise

load(string $filename) : boolean

Loads a neural network from a file saved by the 'save()' function. Clears the training and control data added so far.

Parameters

Name	Type	Description
$filename	string	The filename to load the network from

Returns

Type	Description
boolean	'true' on success, 'false' otherwise

save(string $filename) : boolean

Saves a neural network to a file

Parameters

Name	Type	Description
$filename	string	The filename to save the neural network to

Returns

Type	Description
boolean	'true' on success, 'false' otherwise

setEpoch(integer $epoch) : void

After training, this function is used to store the number of epochs the network needed for training the network. An epoch is defined as the number of times the complete trainingset is used for training.

Parameters

Name	Type	Description
$epoch	integer

setErrorControlSet(float $error) : void

After training, this function is used to store the squared error between the desired output and the obtained output of the control data.

Parameters

Name	Type	Description
$error	float	The squared error of the control data

setErrorTrainingSet(float $error) : void

After training, this function is used to store the squared error between the desired output and the obtained output of the training data.

Parameters

Name	Type	Description
$error	float	The squared error of the training data

setLearningRate(array $learningRate) : void

Sets the learning rate between the different layers.

Parameters

Name	Type	Description
$learningRate	array	An array containing the learning rates [range 0.0 - 1.0]. The size of this array is 'layerCount - 1'. You might also provide a single number. If that is the case, then this will be the learning rate for the whole network.

setMomentum(float $momentum) : void

Sets the 'momentum' for the learning algorithm. The momentum should accelerate the learning process and help avoid local minima.

Parameters

Name	Type	Description
$momentum	float	The momentum. Must be between 0.0 and 1.0; Usually between 0.5 and 0.9

setTrainingSuccessful(boolean $success) : void

After training, this function is used to store whether or not the training was successful.

Parameters

Name	Type	Description
$success	boolean	'true' if the training was successful, 'false' otherwise

setVerbose(boolean $isVerbose) : void

Determines if the neural network displays status and error messages. By default, it does.

Parameters

Name	Type	Description
$isVerbose	boolean	'true' if you want to display status and error messages, 'false' if you don't

showWeights(boolean $force = false) : void

Shows the current weights and thresholds

Parameters

Name	Type	Description
$force	boolean	Force the output, even if the network is {@link setVerbose() not verbose}.

squaredError(array $input, array $desired_output) : float

Calculate the root-mean-squared error of the output, given the desired output.

Parameters

Name	Type	Description
$input	array	The input to test
$desired_output	array	The desired output

Returns

Type	Description
float	The root-mean-squared error of the output compared to the desired output

squaredErrorControlSet() : float

Calculate the root-mean-squared error of the output, given the controldata.

Returns

Type	Description
float	The root-mean-squared error of the output

squaredErrorEpoch() : float

Calculate the root-mean-squared error of the output, given the trainingdata.

Returns

Type	Description
float	The root-mean-squared error of the output

train(integer $maxEpochs = 500, float $maxError = 0.01) : boolean

Start the training process

Parameters

Name	Type	Description
$maxEpochs	integer	The maximum number of epochs
$maxError	float	The maximum squared error in the training data

Returns

Type	Description
boolean	'true' if the training was successful, 'false' otherwise

inc\classes\help\classes\NeuralNetwork.php

\HelpClasses\NeuralNetworkjump to top

End result

Properties

Methods

\HelpClasses\NeuralNetwork
jump to top