[ya1108]

我设计一个有一个隐含层的BP网络，但是调试总有错误，哪位高手帮忙指点一下，不胜感激~~

clear
P=[1,0.05,500;
1,0.1,500;
1,0.15,500;
1.2,0.12,800;
1.8,0.08,600;
2,0.15,1000;
2,0.2,1000;
2,0.2,1000;
3,0.25,1500;
3,0.3,1500;
3,0.2,1500];%网络输入矢量
T=[1.6 3.2 3.1 1.6 1.6 6.3 6.3 5.4 25 12.5 13.7;
0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 0 0 0];%网络输出矢量
P=P';
T=T';
[PN,minp,maxp,TN,mint,maxt]=premnmx(P,T);
net=newff(minmax(P),[3,7,3],{'tansig','tansig','tansig'},'traingdx','learngdm','sse');%设置网络结构、激活函数、学习函数等
net=init(net);
net.trainparam.goal=0.1;%设置误差为0.1
net.trainparam.epochs=20000;%训练次数为20000
net.trainparam.learnrate=0.2;%学习速率为0.2
net.trainparam.show=50;%每50次显示训练描述结果
net = train(net,PN,TN);
y2=sim(net,PN);
[P,T]=postmnmx(PN,minp,maxp,TN,mint,maxt)
plot(P,T,'o',P,y2,'*');
在matlab中运行后，提示以下错误：
??? Error using ==> unknown
Matrix dimensions must agree.

Error in ==> premnmx at 84
tn = 2*(t-mint0*oneQ)./((maxt0-mint0)*oneQ) - 1;

Error in ==> ra4 at 18
[PN,minp,maxp,TN,mint,maxt]=premnmx(P,T);

哪位高手指点一下，谢谢了