Main Page | Namespace List | Class Hierarchy | Alphabetical List | Class List | File List | Namespace Members | Class Members | File Members | Related Pages

annie - Neural Network Library


Last Modified On: January 13, 2003


annie is an ANN, ie, Artificial Neural Network library developed in C++. It can be used to implement various kinds of neural networks like Multi-Layer Perceptron, Radial basis function networks, Hopfield networks etc.

To avoid possible conflict with any other library that you may be using, all classes here are packaged into a namespace annie.

This library is intended to be ideal for beginners who need to use neural networks in some particular application, and want to concentrate on the application rather than the network implementation itself, so such users can treat this library as a black-box, not worrying about how it's working.

You can also download the Matlab Extensions Pack from the website (, which will help you use the features of Matlab to train and modify networks and yet be able to use them in your C++ programs.

Developed by : Asim Shankar

Getting Started

Starting up

All functionality is packaged in the namespace 'annie'. So, if you're writing a program, just do the following:

 #include <annie.h>
 using namespace annie;

You can ignore the second step if you explicitly use annie:: before any member of the library. For example,

 annie::TwoLayerNetwork net(2,3,1); 
to create a 2 layer feed-forward neural network with 2 inputs, 3 hidden nodes and 1 output.

Ofcourse, you have to make sure that your compiler knows where to find annie.h and should be able to link to the library file. See the README.txt file in the annie distribution for details. etc...

Generated on Fri Jun 18 13:18:10 2004 for Annie by doxygen 1.3.5