Lightweight Neural Network

This is a lightweight implementation of a neural network for use in C and C++ programs. It is intended for use in applications that just happen to need a simple neural network and do not want to use needlessly complex neural network libraries. Features Multilayer perceptron neural network. Backpropagation training. Trainable bias. Small. Fast. Easy to incorporate in own application. Easy to extend. Licenced under LGPL. Includes example application to train a network to recognize handwritten digits. More Visit the SourceForge project page, where the project is hosted and where you can browse the CVS repository, or visit the Freshmeat project page. Download the latest version (0.8), released 28 July 2005. You may also want to download the training data (version 1.1) for the OCR example. Report a bug or request a feature, or just browse the bug database, using SourceForge's bug tracking system. Read the manual page, view the internal documentation or look at the internal structure of a neural network. Look at various code snippets explaining the usage of the library. Links comp.ai.neural-nets FAQ UCI Machine Learning Repository Lightweight Neural Network++, a fork of this project written in C++ Peter van Rossum's homepage

Copyright © 2003-2005 Peter van Rossum since 29 September 2003 Updated 2 August 2005