Fast Artificial Neural Network Library is a neural network library that implements multilayer artificial neural networks in C with support for both fully connected and sparsely connected networks. Cross-platform execution in both fixed and floating point are supported. It includes a framework for easy handling of training data sets. It is easy to use, versatile, well documented, and fast. C++, Perl, PHP, .NET, Python, Delphi, Octave, Pure Data, and Mathematica bindings are available. A reference manual accompanies the library with examples and recommendations on how to use the library. A graphical user interface is also available for the library.
|Tags||Software Development Libraries Scientific/Engineering Artificial Intelligence Internet Web Dynamic Content CGI Tools/Libraries|
|Operating Systems||OS Independent POSIX Mac OS X Linux Windows|
|Implementation||C Python PHP Delphi C++|
Release Notes: This release changes the create_array functions to indicate that the layer's array parameter is const (i.e., input only), adds a documented C++ wrapper and sample C++ program, and adds many more changes.
Release Notes: This release includes automatic building and training of ANN's using the Cascade2 algorithm. It also includes many new activation functions and a more flexible use of activation functions. A new file format is introduced that makes saving and loading of ANNs easier. This release should make it easier to use the FANN library and easier to get good results with the library.
Release Notes: This release adds RPROP, Quickprop, and batch training in addition to the current incremental training. More activation functions have been added and shortcut connections have been enabled.
Release Notes: This release adds PHP and Python bindings, MSVC++ project files, and deb/RPM packages. An extensive reference manual is added, as well as several new features and activation functions.
Release Notes: Some feature enhancements and changes as a result of collaboration with Evan Nemerson, who is porting the library to PHP.