rbf4.m

This procedure is well used in classifying based on matlab. And this time 3 RBF procedure are upload

%散布常数对径向基函数网络设计的影响
%散布常数很大，网络进行函数逼近时，出现了不适性。
p=-1:0.1:1;
t=[-.9602 -.5770 -.0729 .3771 .6405 .6600 .4609 .1336 -.2013 -.4344 -.5000 -.3930 -.1647 .0988 .3072 .3960 .3449 .1816 -.0312 -.2189 -.3201];
plot(p,t,'+');
title('训练样本');
xlabel('输入矢量p');
ylabel('目标矢量t');
eg=0.02;  %均放差精度
sc=100;     %散布常数
net=newrb(p,t,eg,sc);
X=-1:0.01:1;
Y=sim(net,X);
hold on
plot(X,Y);
hold off
legend({'目标','输出'})