nninit.m

推荐一下老外的工具箱psotb-beta-.3

% BP网络初始化:给出网络的训练样本P、T,
% 输入、输出数及隐含神经元d数R,S2,S1
function [P,T,R,S1,S2,Q,S]=nninit
% p=[0 0 0 0 0;...
%    1 0 0 0 1;...
%    0 1 0 0 1;...
%    0 0 1 0 0;...
%    0 0 0 1 0];...
% P=p';
% t=[0 0 0 0;...
%    1 0 0 0;...
%    0 1 0 0;...
%    0 0 1 0;...
%    0 0 0 1];
% T=t';
t=[14 10.9 14 19 19 19.7 30 17 17 16 16 14.2 14.4 25 13 14 20.5 20.5 9.5 9.6 12 17 16.1 11 12.5 12.5 12.5 13 18.4 18 12.1 14 19 19.7 30 14.4 25 13 18.3;...
        350 350 400 400 400 400 400 350 350 350 350 400 400 400 400 400 400 400 400 400 400 400 400 400 350 350 350 350 400 400 350 400 400 400 400 400 400 400 400;...
        13.5 10.4 13 16 16 17.2 28 15.9 16 15.4 15.4 13.5 13.8 24.5 12.6 13.6 20.1 20.1 7.3 6.4 11.3 16.5 15.7 10.4 12 12 12 12.5 16 16 11.6 13 16 17.2 28 13.8 24.5 12.6 16;...
        36 31.9 15.5 46.1 46.1 39.8 45.1 30.1 30.1 29.1 27.6 17.5 17.5 22.2 29.1 29.1 26.1 26.1 49 49 51.1 41 43.4 53.7 68.8 68.8 68.8 66.1 52.1 52.1 33.6 15.5 46.1 39.8 45.1 17.5 22.2 29.1 52.1;...
        2.5 2.5 2.6 1.3 1.3 2.3 2.8 3 3 2.2 2.2 1.6 1.6 2.7 2.2 2.2 1.7 1.7 6.5 6.5 2.5 2.5 2.5 2.5 2.3 2.3 2.3 2.3 2.8 2.8 2.5 2.6 1.3 2.3 2.8 1.6 2.7 2.2 2.8;...
        1800 1000 910 1300 1200 1440 3250 1900 1800 1500 1700 850 750 1400 1050 1100 1800 2000 1600 1700 2160 2400 2160 2040 800 900 950 950 2600 2500 1400 840 1100 1560 3250 750 1600 1000 2600];
[pn,minp,maxp] = premnmx(t);
pn=(pn+1)/2*.8+.1;
P=pn(1:5,1:30);
T=pn(6,1:30);
[R,Q]=size(P);
[S2,Q]=size(T);
S1=11;
S=R*S1+S1*S2+S1+S2; % 遗传算法编码长度