OLSOFT Neural Network Library is a fully self-contained COM ActiveX control written in Visual C++ 6.0 and ATL. There are no external dependencies required for its operation such as other DLLs.
OLSOFT Neural Network Library is the class to create, learn and use Back Propagation neural networks and SOFM (Self-Organizing Feature Map). The library makes integration of neural networks functionality into your own applications easy and seamless. It enables your programs to handle data analysis, classification and forecasting needs.
The advanced learning and training capacities of OLSOFT Neural Network library allow using it for solving various needs and performing data mining, clusterization tasks in a variety of fields, such as finance, science, education, medicine, sport betting, etc.
Main features of the library: written as a COM object that may be used inside any application; library is fast because it is written in C++; supports multilayer artificial networks (Back Propagation and SOFM); forward Propagation Network with user defined synapses (use of "offset input" is supported); back Propagation Network training with adjustable traning parameters; SOFM network support (Euclid distance and Scalar Vector Multiplication algorythm); several different activation functions implemented (zero-threshold, hyperbolic tangens, exponent sigmoid); it is possible to save the network to a text file and restore later for use or further training; direct and simple use in VB6 and VBA applications (see examples); use in any programming language as a generic COM component.
OLSOFT Neural Network library is a proven technology tool that is implemented in NeuroXL products, namely NeuroXL Classifier and NeuroXL Predictor products.
Note that all data types are specific to Visual Basic 6. For Visual Basic.NET some data types will differ. A reference manual accompanies the library with examples and recommendations on how to use the library. Source Code Edition is also available.
Tags: neural network library, neural network, neural, back propagation, classification, prediction, data mining, clusterization, data analysis, artificial intelligence, neural network object, forecasting
Download
Get the Full Version of
|
