紧急求助：关于bp神经网络的输出向量问题
小弟编辑一段bp神经网络程序，运行时，系统报错提示：Targets are incorrectly sized for network.Matrix must have 1 rows. 是否bp神经网络的目标输出都只能是一维向量？如果多维可以，此程序应如何修改？请大家多指教，期待尽快回复，急！！！pp=; ％pp为540*21矩阵
T=[1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0;
0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0];
p=pp';
t=T;
% 归一化训练样本
for i=1:21
p(:,i)=(p(:,i)-min(p(:,i)))/(max(p(:,i))-min(p(:,i)));
end
%----------------------确定隐层神经元---------------------------------------------
net=newff(minmax(p),,{'tansig','logsig'},'traingdx'); %最速下降BP法，节省机器内存
inputWeights=net.iw{1,1};
inputbias=net.b{1,1};
layerWeights=net.iw{2,1};
layerbias=net.b{2,1};
net.trainParam.lr = 0.05;
net.trainParam.show = 50;
net.trainParam.epochs = 10000;
net.trainParam.goal = 0.0002;
net=train(net,p,t);
YY=sim(net,p);
页:
[1]