Package Exports
- ml-fnn
This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (ml-fnn) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.
Readme
Feedforward Neural Network
A implementation of feedforward neural networks in javascript based on the mrbo answer found here:
[Implementation] (http://stackoverflow.com/questions/9951487/implementing-a-neural-network-in-java-training-and-backpropagation-issues)
Methods
new FeedforwardNeuralNetwork([LayersSize])
Arguments
layersSize- Array of numbers with sizes of each layer.
Example
var FNN = new FeedforwardNeuralNetwork([2, 4, 1]);train(trainingSet, predictions, learningRate, momentum)
Train the Neural Network with a given training set, predictions, learning rate and a momentum (Regularization term).
Arguments
trainingSet- A matrix of the training set.predictions- A matrix of predictions with the same size of rows of the trainingSet.learningRate- The learning rate (number).momentum- The regularization term (number).
Example
var trainingSet = [[0, 0], [0, 1], [1, 0], [1, 1]];
var predictions = [[0], [0], [0], [1]];
FNN.train(trainingSet, predictions, 0.3, 0.3);predict(dataset)
Predict the values of the dataset.
Arguments
dataset- A matrix that contains the dataset.
Example
var dataset = [[0, 0], [0, 1], [1, 0], [1, 1]];
var ans = FNN.predict(dataset);export()
Exports the actual Neural Network to an Javascript Object.
load(model)
Returns a new Neural Network with the given model.
Arguments
model- Javascript Object generated from export() function.