Delphi 7.0x freeware [Maths, Stats & Algorithms]

Delphi 7.0x freeware
Maths, Stats & Algorithms

[ amrandom.zip ] [ 384,340 bytes ]

[ Freeware ] [ With Source ]
[ D4 | D5 | D6 | D7 ]

This aims to supply a Borland Delphi translation of Alan Miller's Random Module for FORTRAN-90. This translation has been done with Dr Miller's approval and is being made FREELY available to all Delphi Developers, though we do ask the Alan Miller and ESB Consultancy be given due credit. It includes the following Random Number Generators: - Normal (Gaussian) - Gamma - Chi-squared - Exponential - Weibull - Beta - t - Multivariate Normal - Generalized inverse Gaussian - Binomial (2 different ones) - Negative Binomial - von Mises - Cauchy Includes full Delphi Source and Demo.

Added: 15-10-2001/9-08-2004 | Version: 3.0.6 | Downloads: 3176/3116/14

Author:

Glenn Crouch ESB Consultancy [Homepage] [DSP files]

[ LeastSquareFitting.zip ] [ 176,753 bytes ]

[ Freeware ] [ With Source ]
[ D6 | D7 ]

A mathematical component for finding the best fitting curve to a given set of points by minimizing the sum of the squares of the offsets ("the residuals") of the points from the curve. Demo Applications with sourcecode (linear, nonlinear, multiple regression) are included. Use of multiple regression involves the discovering of the relationship between the values and then finding an equation that satisfies that relationship.

Added: 16-05-2005 | Version: 0.0 | Downloads: 264/264/22

Author:

Josef Schützenberger [Homepage] [DSP files]

[ TEDSoft_Neural_MLPNet.zip ] [ 21,140 bytes ]

[ Freeware ] [ With Source ]
[ D5 | D6 | D7 ]

This component realizes a neural multi-layer perceptron (MLP) with three or four layers (input, one or two hidden, output). The weights of the hidden and output layers can be trained using the backpropagation-of-error algorithm. The input of training and test data is possible via textfiles. The .zip-file contains the component as well as a short description (info.txt) and a small demo.

Added: 3-03-2003 | Version: 1.0 | Downloads: 2016/2016/15

Author:

Stefan Meuser TEDSoft GmbH [Homepage] [DSP files]

[ TEDSoft_Neural_RBFNet.zip ] [ 20,428 bytes ]

[ Freeware ] [ With Source ]
[ D5 | D6 | D7 ]

This component realizes a neural radial basisfunction-network (RBF-network) with three layers (input, RBF, output). The centers of the RBF-layer can be adjusted randomly, as set of the training data, or with the k-means-clustering algorithm. The training of the output weights is done using delta rule. The input of training and test data is possible via textfiles. The .zip-file contains the component as well as a short description (info.txt) and a small demo.

Added: 18-11-2002/3-03-2003 | Version: 1.1 | Downloads: 2523/2522/13

Author:

Stefan Meuser TEDSoft GmbH [Homepage] [DSP files]