An implementation of a multi-layer perceptron in Scala
Switch branches/tags
Nothing to show

Scala Neural Network

An implementation in Scala written by Yann Nicolas

This Scala implementation of a neural network, allows the execution and the training of a multi-layer perceptron.


To build and run the XOR example you need Simple Build Tool (sbt).


To compile and run the XOR example use 'sbt run'.

  • First, a new Perceptron will be created with two hidden layers, one of 5 neurons and another of 10 neurons.

  • Then, 100 iteration will be used to train the perceptron to be able to solve XOR operations

  • The perceptron configuration will be saved in the XOR.xml file

  • The perceptron will be loaded from the XOR.xml file

  • XOR operations will be run to verify the perceptron outputs

You can also use the 'sbt eclipse' to generate an Eclipse project.


The use and distribution terms for this software are covered by the Apache Software Foundation License 2.0 (