求

lyfsgdhr

P=[0.1093,0.1110,0.1127,0.1141,0.1154,0.1164,0.1171,0.1175,0.1178,0.1179,0.1179,0.1179,0.1179,0.1180,0.1182];
5 F# A5 @! X0 qT=[0.1110,0.1127,0.1141,0.1154,0.1164,0.1171,0.1175,0.1178,0.1179,0.1179,0.1179,0.1179,0.1180,0.1182,0.1185];
; L( X: |4 T' V' ^* t5 H/ {* b%  创建一个新的前向神经网络
net_1=newff(minmax(P),[10,1],{'tansig','purelin'},'traingdm')
%  当前输入层权值和阈值
$ J5 e( w" I; yinputWeights=net_1.IW{1,1}
3 T# I; h" W) k* u$ @, ninputbias=net_1.b{1}
%  当前网络层权值和阈值
layerWeights=net_1.LW{2,1}
layerbias=net_1.b{2}
- E4 d( X) t: ~& q%  设置训练参数
3 ~% E) u' y2 u, Mnet_1.trainParam.show = 50;
net_1.trainParam.lr = 0.05;
+ K' i8 r3 \3 Gnet_1.trainParam.mc = 0.9;
, w" G' ]0 {; ]) E+ xnet_1.trainParam.epochs = 10000;
net_1.trainParam.goal = 1e-3;
%  调用 TRAINGDM 算法训练 BP 网络
7 V, y% {) ]' h4 Q. Z. k[net_1,tr]=train(net_1,P,T);
%  对 BP 网络进行仿真
+ W3 _) V4 ~+ s! P; v1 r; N4 E( bA = sim(net_1,P);
5 P) {! i- P  q0 |2 \4 d* I/ O%  计算仿真误差
: k% r! i9 S& x: hE = T - A;
* J5 v2 u3 ?$ d4 A. nMSE=mse(E)
( Z3 K! R5 E0 a5 n" Mx=[0.1110,0.1127,0.1141,0.1154,0.1164,0.1171,0.1175,0.1178,0.1179,0.1179,0.1179,0.1179,0.1180,0.1182,0.1185]';%测试
( w+ z6 E: I% K$ Dsim(net_1,x)
+ W6 ]# L2 x; z* a- R. r这段程序是根据14年的数据，来预测下一年的，怎么算不出来啊 。

小青菜

看上去好烦哦

zhou2yong888

看不懂·····

fffffffddf

...........................................

kxhu

哦～～

hbchina99

鉴定完毕！  

ask4kevin

看起来好~~像啊~~~~~

talent212121

鉴定完毕！  

hyxie616

楼主，你写得实在是太好了。我惟一能做的，就只有把这个帖子顶上去这件事了

pankg

我基本上是采用看英语文章的办法，先泛读，再精读，再一句一句看，最后再提纲挈领，总算是明白一点了，当然，也可能还是领悟错了。最后要说的一句话是：楼主，你很牛叉，希望你不是真的有病。