MATLAB BP神经网络训练P=[392,584,776,2713,3236,2783,2163;718,1043,1368,4318,5054,4483,3769];T=[1.4404,1.4722,1.4753,1.497,1.7663,1.6807,1.7426];net=newff(minmax(P),[3,1],{'tansig','purelin'},'traingdm');net.trainParam.show = 50;net.trainParam.lr
来源:学生作业帮助网 编辑:作业帮 时间:2024/06/18 00:42:55
![MATLAB BP神经网络训练P=[392,584,776,2713,3236,2783,2163;718,1043,1368,4318,5054,4483,3769];T=[1.4404,1.4722,1.4753,1.497,1.7663,1.6807,1.7426];net=newff(minmax(P),[3,1],{'tansig','purelin'},'traingdm');net.trainParam.show = 50;net.trainParam.lr](/uploads/image/z/6841802-2-2.jpg?t=MATLAB+BP%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C%E8%AE%AD%E7%BB%83P%3D%5B392%2C584%2C776%2C2713%2C3236%2C2783%2C2163%3B718%2C1043%2C1368%2C4318%2C5054%2C4483%2C3769%5D%3BT%3D%5B1.4404%2C1.4722%2C1.4753%2C1.497%2C1.7663%2C1.6807%2C1.7426%5D%3Bnet%3Dnewff%28minmax%28P%29%2C%5B3%2C1%5D%2C%7B%27tansig%27%2C%27purelin%27%7D%2C%27traingdm%27%29%3Bnet.trainParam.show+%3D+50%3Bnet.trainParam.lr)
MATLAB BP神经网络训练P=[392,584,776,2713,3236,2783,2163;718,1043,1368,4318,5054,4483,3769];T=[1.4404,1.4722,1.4753,1.497,1.7663,1.6807,1.7426];net=newff(minmax(P),[3,1],{'tansig','purelin'},'traingdm');net.trainParam.show = 50;net.trainParam.lr
MATLAB BP神经网络训练
P=[392,584,776,2713,3236,2783,2163;718,1043,1368,4318,5054,4483,3769];
T=[1.4404,1.4722,1.4753,1.497,1.7663,1.6807,1.7426];
net=newff(minmax(P),[3,1],{'tansig','purelin'},'traingdm');
net.trainParam.show = 50;
net.trainParam.lr = 0.05;
net.trainParam.mc = 0.9;
net.trainParam.epochs = 1000;
net.trainParam.goal = 1e-3;
[net,tr]=train(net,P,T);
A = sim(net,P)
得到A的结果全是一样的,是程序问题,
MATLAB BP神经网络训练P=[392,584,776,2713,3236,2783,2163;718,1043,1368,4318,5054,4483,3769];T=[1.4404,1.4722,1.4753,1.497,1.7663,1.6807,1.7426];net=newff(minmax(P),[3,1],{'tansig','purelin'},'traingdm');net.trainParam.show = 50;net.trainParam.lr
你的程序训练完毕后根本就没达到目标误差,就是说训练效果不好,不能进行预测,只有训练结果好了才能预测仿真,你再改一下隐含层神经元数或者训练和传递函数试试吧~
另外输入层的值可以归一化也可以不归一化,归一化后在仿真之前要反归一化.