Elsevier Science Home
Computer Physics Communications Program Library
Full text online from Science Direct
Programs in Physics & Physical Chemistry
CPC Home

[Licence| Download | New Version Template] acjm_v1_0.gz(90 Kbytes)
Manuscript Title: LVQNET 1.10, A program for neural net and statistical pattern recognition.
Authors: A. Cherubini, R. Odorico
Program title: LVQNET
Catalogue identifier: ACJM_v1_0
Distribution format: gz
Journal reference: Comput. Phys. Commun. 72(1992)249
Programming language: Fortran.
Computer: CRAY Y-MP.
Operating system: UNICOS, VMS, VM-CMS.
RAM: 1000K words
Word size: 32
Keywords: Computers, Neural networks, Quantization, Statistical Discrimination, Pattern classification.
Classification: 6.4.

Nature of problem:
Pattern classification.

Solution method:
Three different methods for pattern classification are employed:
i) a neural net based on an improved version of Learning Vector Quantization,
ii) Mahalanobis distance,
iii) Fisher's linear discriminant.


Running time:
It depends on the complexity of the problem. The test run was completed in 123 seconds of cpu time on a VAX 6000-320.