Personal tools
Document Actions

StatPatternRecognition

by narsky last modified 2006-10-04 11:44

C++

Library/Module

Categorization

Wide variety of tools to separate signal from background

I. Narsky

The package implements linear and quadratic discriminant analysis, decision trees, bump hunting (PRIM), boosting (AdaBoost and arc-x4), bagging and random forest algorithms, a multi-class learner, and interfaces to the standard backpropagation neural net and radial basis function neural net implemented in the Stuttgart Neural Network Simulator. Supplemental tools such as bootstrap, estimation of data moments, a test of zero correlation between two variables with a joint elliptical distribution, and a multivariate goodness-of-fit method are also provided. The package offers a convenient set of tools for imposing requirements on input data, storing output into Root or Hbook, and handling multi-class data. Integrated in the BaBar computing environment, the package maintains a minimal set of external dependencies and can be easily adapted to any other HEP environment. All publications and talks about the package are collected on my home page http://www.hep.caltech.edu/~narsky/spr.html If you are about to download a copy of the package, I encourage you to send me email for inclusion in the StatPatternRecognition mailing list. You will receive timely updates on the status of the package.

narsky@hep.caltech.edu
spr_tarfile.tar by File Uploader Account — last modified 2006-10-04 11:46
TAR file, -- windows users can use http://gnuwin32.sourceforge.net/packages/tar.htm
spr_v03-01-02-tar.gz by Site Administrator — last modified 2006-05-30 16:04
 

Powered by Plone, the Open Source Content Management System