Difference between revisions of "Main Page"

From Artificial Neural Network for PHP
Line 3: Line 3:
 
This project realizes a '''neural network topology called ''multilayer perceptron'' for PHP 5.x''' environments. The source code is based on a work by ''Eddy Young'' in 2002. Several improvements and changes on this implementation are done by ''Thomas Wien'' in 2007.
 
This project realizes a '''neural network topology called ''multilayer perceptron'' for PHP 5.x''' environments. The source code is based on a work by ''Eddy Young'' in 2002. Several improvements and changes on this implementation are done by ''Thomas Wien'' in 2007.
   
  +
== Versions and Change-Log ==
----
 
   
 
'''Version 1.0 by Eddy Young''' (2002)
 
'''Version 1.0 by Eddy Young''' (2002)
Line 25: Line 25:
 
* Fixing bug: initializing inputs to all hidden layers
 
* Fixing bug: initializing inputs to all hidden layers
 
* Fixing bug: training for first hidden layer was skipped
 
* Fixing bug: training for first hidden layer was skipped
  +
  +
== Todo ==
  +
  +
* Separation of classes to several files
  +
* Graphical output of neural network values
   
 
== Overview ==
 
== Overview ==

Revision as of 21:44, 17 December 2007

ANN - Artificial Neural Network for PHP 5.x

This project realizes a neural network topology called multilayer perceptron for PHP 5.x environments. The source code is based on a work by Eddy Young in 2002. Several improvements and changes on this implementation are done by Thomas Wien in 2007.

Versions and Change-Log

Version 1.0 by Eddy Young (2002)

  • Initial version

Version 2.0 by Thomas Wien (17.12.2007)

  • PHP 5.x support
  • PHPDoc documentation
  • Momentum support
  • Linear / binary output
  • ANN_InputValue + ANN_OutputValue classes
  • Exceptions
  • Threshold function
  • Tangens hyperbolicus transfer function
  • Several performance issues
  • Avoiding array_keys & srand
  • Changes in saving and loading network
  • Printing network details to browser
  • Fixing bug: initializing inputs to all hidden layers
  • Fixing bug: training for first hidden layer was skipped

Todo

  • Separation of classes to several files
  • Graphical output of neural network values

Overview