UNPKG

encog

Version:

Encog is a NodeJs ES6 framework based on the Encog Machine Learning Framework by Jeff Heaton, plus some the of basic data manipulation helpers.

118 lines (100 loc) 2.65 kB
/** * This interface defines a freeform neuron. By freeform that this neuron is not * necessarily part of a layer. */ class FreeformNeuron { /** * Add an input connection to this neuron. * * @param inputConnection {FreeformConnection} * The input connection. */ addInput(inputConnection) { } /** * Add an output connection to this neuron. * * @param outputConnection {FreeformConnection} * The output connection. */ addOutput(outputConnection) { } /** * @return The activation for this neuron. This is the final output after * the activation function has been applied. */ getActivation() { } /** * @return {InputSummation} The input summation method. */ getInputSummation() { } /** * @return {Array} The outputs from this neuron. */ getOutputs() { } /** * @return {Number} The output sum for this neuron. This is the output prior to the * activation function being applied. */ getSum() { } /** * @return {Boolean} True, if this is a bias neuron. */ isBias(){} /** * Perform the internal calculation for this neuron. */ performCalculation() { } /** * Set the activation, or final output for this neuron. * @param {Number} activation THe activation. */ setActivation(activation){} /** * Determine if this neuron is a bias neuron. * @param b {Boolean} True, if this neuron is considered a bias neuron. */ setBias(b){} /** * Set the input summation method. * @param {InputSummation} theInputSummation The input summation method. */ setInputSummation(theInputSummation){} /** * Update the context value for this neuron. */ updateContext(){} /** * Add to the specified temp value. * @param i {Number} The index. * @param value {Number} The value to add. */ addTempTraining(i, value){} /** * Allocate the specified length of temp training. * @param l {Number} The length. */ allocateTempTraining(l){} /** * Clear the temp training. */ clearTempTraining(){} /** * Get the specified temp training. * @param index {Number} The indfex. * @return {Number} The temp training value. */ getTempTraining(index){} /** * Set a temp training value. * @param index {Number} The index. * @param value {Number} The value. */ setTempTraining(index, value){} } module.exports = FreeformNeuron;