gusucode.com > 支持向量机工具箱 - LIBSVM OSU_SVM LS_SVM源码程序 > 支持向量机工具箱 - LIBSVM OSU_SVM LS_SVM\stprtool\linear\anderson\Contents.m
% Generalized Anderson's task. % % andrdemo - Demonstrates algorithms which find solution of % the Generalized Anderson's task (GAT). % % Algorithms: % oanders - Original Anderson's solution for two distributions. % eanders - Epsilon-solution of the Generalized Anderson's task. % gganders - Finds optimal solution of GAT using general gradient % optimization. % ganders - Finds optimal solution of the GAT, general algorithm. % ganders2 - Finds optimal solution of the GAT, upgraded version. % % Auxiliary functions: % andrerr - Computes upper limit of probability of bad classification % in the GAT. % mineps - Solves one subtask of the GAT which is minimalization of % special objective function. Uses discretization in % independent variable. % minepsvl - --//-- , uses discretization of function value. % minepsrt - --//-- , uses transformation to problem of finding % polynom roots. % pandr2d - Visualizes solution of the GAT in 2D. % pandr2df - Visualizes solution of the GAT in 2D, fulfiles ellipses. % testandr - Randomly create test examples for GAT. % % Statistical Pattern Recognition Toolbox, Vojtech Franc, Vaclav Hlavac % (c) Czech Technical University Prague, http://cmp.felk.cvut.cz % Written Vojtech Franc (diploma thesis) 02.01.2000 % Modifications % 24. 6.00 V. Hlavac, comments polished.