Basic Prop is born of the need to teach the basics of artificial neural networks to students without substantial programming experience. There are many neural network simulators out there, almost all of which are more powerful and versatile than Basic Prop. But Basic Prop is simple. It does not overwhelm the user with options. It provides simple graphing and analysis tools that allow the user to quickly survey the state of the network.
The code for Basic Prop is written in java by Fred Cummins. It is based on an original simulator written by three German students, Martin Brunninger, Stefan Wasserl, and Alexander Pichler. The original code and applet are still available (as of 2011) at this link. The original code has been rewritten more or less from scratch, and several additional functions have been added, most notably the implementation of a Simple Recurrent Network architecture.
The code is being made freely available for educational purposes. It may not be resold or used in a commercial product. The source code will be made available and may be modified, subject to similar licensing conditions being attached to amy modified code. If in doubt, please contact the maintainer. If you do modify the code, consider contributing your modification to the project.
Basic Prop: A Neural Network Simulator for Educational Purposes by Fred Cummins is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.