octoleo/php-ml
1
0
Fork 0
mirror of https://github.com/Llewellynvdm/php-ml.git synced 2024-11-21 20:45:10 +00:00
Code Issues Packages Projects Releases Wiki Activity

Change from theta to learning rate var name in NN (#159)

Browse Source
This commit is contained in:
David Monllaó 2017-11-20 23:39:50 +01:00 committed by Arkadiusz Kondas
parent 333598b472
commit b1d40bfa30
3 changed files with 11 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
docs/machine-learning/neural-network/multilayer-perceptron-classifier.md
View File

@ -8,7 +8,7 @@ A multilayer perceptron (MLP) is a feedforward artificial neural network model t
* $hiddenLayers (array) - array with the hidden layers configuration, each value represent number of neurons in each layers
* $classes (array) - array with the different training set classes (array keys are ignored)
* $iterations (int) - number of training iterations
* $theta (int) - network theta parameter
* $learningRate (float) - the learning rate
* $activationFunction (ActivationFunction) - neuron activation function
```

10
src/Phpml/NeuralNetwork/Network/MultilayerPerceptron.php
View File

@ -46,9 +46,9 @@ abstract class MultilayerPerceptron extends LayeredNetwork implements Estimator,
protected $activationFunction;
/**
* @var int
* @var float
*/
private $theta;
private $learningRate;
/**
* @var Backpropagation
@ -58,7 +58,7 @@ abstract class MultilayerPerceptron extends LayeredNetwork implements Estimator,
/**
* @throws InvalidArgumentException
*/
public function __construct(int $inputLayerFeatures, array $hiddenLayers, array $classes, int $iterations = 10000, ?ActivationFunction $activationFunction = null, int $theta = 1)
public function __construct(int $inputLayerFeatures, array $hiddenLayers, array $classes, int $iterations = 10000, ?ActivationFunction $activationFunction = null, float $learningRate = 1)
{
if (empty($hiddenLayers)) {
throw InvalidArgumentException::invalidLayersNumber();
@ -73,7 +73,7 @@ abstract class MultilayerPerceptron extends LayeredNetwork implements Estimator,
$this->inputLayerFeatures = $inputLayerFeatures;
$this->hiddenLayers = $hiddenLayers;
$this->activationFunction = $activationFunction;
$this->theta = $theta;
$this->learningRate = $learningRate;
$this->initNetwork();
}
@ -87,7 +87,7 @@ abstract class MultilayerPerceptron extends LayeredNetwork implements Estimator,
$this->addBiasNodes();
$this->generateSynapses();
$this->backpropagation = new Backpropagation($this->theta);
$this->backpropagation = new Backpropagation($this->learningRate);
}
public function train(array $samples, array $targets): void

10
src/Phpml/NeuralNetwork/Training/Backpropagation.php
View File

@ -10,9 +10,9 @@ use Phpml\NeuralNetwork\Training\Backpropagation\Sigma;
class Backpropagation
{
/**
* @var int
* @var float
*/
private $theta;
private $learningRate;
/**
* @var array
@ -24,9 +24,9 @@ class Backpropagation
*/
private $prevSigmas = null;
public function __construct(int $theta)
public function __construct(float $learningRate)
{
$this->theta = $theta;
$this->learningRate = $learningRate;
}
/**
@ -43,7 +43,7 @@ class Backpropagation
if ($neuron instanceof Neuron) {
$sigma = $this->getSigma($neuron, $targetClass, $key, $i == $layersNumber);
foreach ($neuron->getSynapses() as $synapse) {
$synapse->changeWeight($this->theta * $sigma * $synapse->getNode()->getOutput());
$synapse->changeWeight($this->learningRate * $sigma * $synapse->getNode()->getOutput());
}
}
}