Back propagation neural network Java example source code

This example implements a multi-layer neural network, which will behave as an XOR gate after being trained using the back propagation algorithm.



The network has only two layers, the input layer and the output layer. It does not have any hidden layers. The input layer consist of two neurons and the output layer consist of only one neuron

You can find the NeuralNetwork.jar in the following link which includes the complete source code, compiled classes and images.
After downloading the jar file you can run the application by executing the following command.

java -classpath NeuralNetwork.jar com.neural.network.main.AppMain

First press Train button to calculate the weights. A message will be displayed saying you have successfully trained and the weights will get populated to the weight text boxes.

Afterwards you can select input values using the x1 an x2 combo boxes and Press Run button to test whether the neural network produces the correct output.