Multilayer Perceptron

A multilayer perceptron (MLP) is a class of feedforward artificial neural network. An MLP consists of at least three layers of nodes. Except for the input nodes, each node is a neuron that uses a nonlinear activation function. MLP utilizes a supervised learning technique called backpropagation for training. Its multiple layers and non-linear activation distinguish MLP from a linear perceptron. ~Wikipedia

#

Configuration (Configurator class)

Configurator constructor:

    Configurator cfg = new Configurator(int input_count, int[] layers);

Layers structure (example):

    int[] layers = new int[] {
        3, // second layer (hidden) -> 3 neurons
        4, // third layer (hidden) -> 4 neurons,
        2, // fourth layer (output) -> 2 neurons
    }

Last element - output layer.

1st layer (input) - input_count
2nd layer (hidden) - layers[1]
3rd layer (hidden) - layers[2]
...
n layer (output) - layers[n - 1]

Options:

• setInput(double[])
• setExpected(double[])
• setRange(double, double) default: (-0.5, 0.5)
• setLearningFactor(double) default: (0.8)
• setMomentum(double) default: (0.2)
• setBias(boolean) default: (true)
• setInputRotation(boolean) default: (true)
• setEpochs default: 1000
• setErrorLogStep default: 10
• setError default: 0.01