Located in /ANN_Network.php (line 51)
ANN_Filesystem | --ANN_Network
[752] | protected | static | string : | getDefaultFilename () |
[1172] | public | static | void : | loadFromFile ([string $strFilename = null]) |
[1444] | public | static | ANN_Network : | loadFromHost (string $strUsername, string $strPassword, string $strHost) |
[406] | protected | void : | activate () | |
[1505] | protected | void : | adjustLearningRate () | |
[172] | public | ANN_Network : | __construct ([integer $intNumberOfHiddenLayers = 2], [integer $intNumberOfNeuronsPerLayer = 4], [integer $intNumberOfOutputs = 1]) | |
[1130] | protected | void : | calculateMaxTrainingLoops () | |
[366] | protected | void : | createHiddenLayers (integer $intNumberOfHiddenLayers, integer $intNumberOfNeuronsPerLayer) | |
[393] | protected | void : | createOutputLayer (integer $intNumberOfOutputs) | |
[1476] | protected | void : | detectOutputType () | |
[708] | protected | integer : | getCountInputs () | |
[1356] | protected | float : | getNetworkError () | |
[509] | protected | integer : | getNextIndexInputsToTrain ([boolean $boolReset = FALSE]) | |
[1217] | public | integer : | getNumberHiddenLayers () | |
[1231] | public | integer : | getNumberHiddens () | |
[1202] | public | integer : | getNumberInputs () | |
[1246] | public | integer : | getNumberOutputs () | |
[304] | public | array : | getOutputs () | |
[342] | public | array : | getOutputsByInputKey (integer $intKeyInput) | |
[538] | public | integer : | getTotalLoops () | |
[549] | protected | boolean : | isEpoch () | |
[614] | protected | boolean : | isTrainingComplete () | |
[652] | protected | boolean : | isTrainingCompleteByEpoch () | |
[669] | protected | boolean : | isTrainingCompleteByInputKey (integer $intKeyInput) | |
[1338] | protected | void : | logNetworkErrors () | |
[1283] | public | void : | logNetworkErrorsToFile (string $strFilename) | |
[1301] | protected | void : | logWeights () | |
[1264] | public | void : | logWeightsToFile (string $strFilename) | |
[796] | public | void : | printNetwork ([integer $intLevel = 2]) | |
[878] | protected | void : | printNetworkDetails1 () | |
[1039] | protected | void : | printNetworkDetails2 () | |
[1188] | public | void : | saveToFile ([string $strFilename = null]) | |
[1414] | public | void : | saveToHost (string $strUsername, string $strPassword, string $strHost) | |
[1592] | public | void : | setBackpropagationAlgorithm ([integer $intAlgorithm = self::ALGORITHM_BACKPROPAGATION]) | |
[1538] | public | void : | setDynamicLearningRate ([boolean $boolDynamicLearningRate = TRUE]) | |
[202] | protected | void : | setInputs (array $arrInputs) | |
[281] | protected | void : | setInputsToTrain (array $arrInputs) | |
[575] | public | void : | setLearningRate ([float $floatLearningRate = 0.5]) | |
[1143] | public | void : | setMaxTrainingLoopsFactor ([integer $intMaxTrainingLoopsFactor = 230]) | |
[596] | public | void : | setMomentum ([float $floatMomentum = 0.95]) | |
[1644] | public | void : | setOutputErrorTolerance ([float $floatOutputErrorTolerance = 0.02]) | |
[226] | protected | void : | setOutputs (array $arrOutputs) | |
[765] | protected | void : | setOutputType ([integer $intType = self::OUTPUT_LINEAR]) | |
[1626] | public | void : | setQuickPropMaxWeightChangeFactor ([float $floatQuickPropMaxWeightChangeFactor = 2.25]) | |
[267] | public | void : | setValues (ANN_Values $objValues) | |
[1570] | public | void : | setWeightDecay ([float $floatWeightDecay = 0.05]) | |
[1554] | public | void : | setWeightDecayMode ([boolean $boolWeightDecayMode = TRUE]) | |
[440] | public | boolean : | train () | |
[1379] | public | ANN_Network : | trainByHost (string $strUsername, string $strPassword, string $strHost) | |
[727] | protected | void : | training (array $arrOutputs) | |
[1157] | public | void : | __wakeup () |
Adjusting learning rate dynamically
If network error of current epoch is higher than the network error of the previous epoch the learning rate is adjusted by minus 1 per cent of current learning rate. Otherwise the learning rate is adjusted by plus 1 per cent of current learning rate. So, learning rate increases faster than decreasing does. But if learning rate reaches 0.9 it switches back to 0.5 to avoid endless training. The lowest learning rate is 0.5 also to avoid endless training.
Get the output values
Get the output values to the related input values set by setValues(). This method returns the output values as a two-dimensional array.
Log network errors while training in CSV format
Log weights while training in CSV format
Selecting propagation algorithm
EXPERIMENTAL
Setting the learning rate disables dynamic learning rate automatically.
Setting the percentage of output error in comparison to the desired output
Parameter setting for QuickProp algorithm
EXPERIMENTAL
Set Values for training or using network
Set Values of inputs and outputs for training or just inputs for using already trained network.
Inherited From ANN_Filesystem
ANN_Filesystem::loadFromFile()
ANN_Filesystem::saveToFile()
Back propagation (default)
Individual learning rate (EXPERIMENTAL)
iRProp- (EXPERIMENTAL)
iRProp+ (EXPERIMENTAL)
Quick propagation (EXPERIMENTAL)
RProp (EXPERIMENTAL)
RProp- (EXPERIMENTAL)
RProp+ (EXPERIMENTAL)
Binary output type
Linear output type
Documentation generated on Thu, 01 Jan 2009 16:01:52 +0100 by phpDocumentor 1.4.1