神经网络工具箱的使用

天零ヮ冰釋

1. clear
   ' a% q+ v8 t6 N% ~0 p
2. load('trainset.mat');                   %加载训练集数据) q9 L* \; t1 ?( j2 [- X: l
   
3. load('testset.mat');                    %加载测试集数据
   ! Z5 B( Z! b. ~: W% y
4. p1=trainsetdata([1:179],:);3 r: _7 J8 ^: \. P, W+ [
   
5. p2=trainsetdata([180:300],:);
   ( O' t/ p! v0 q& N. E' h) n5 P
6. p=[p1;p2]';
   9 l) j$ }6 h, b& x. T: b
7. pr=minmax(p);, k& Z\" T8 t4 ]9 z& e
   
8. goal=[ones(1,179),zeros(1,121);zeros(1,179),ones(1,121)];
   % z\" V( s5 j- B) V4 N* M$ j; h+ y% u
9. net=newff(pr,[21,160],{'tansig','logsig'}); %隐含层节点和输出层节点( q8 }\" K* e! z' Z0 {' d
   
10.                                             %取值不太确定，请指教' U# d+ g) w% D( z( d; P
    
11. net.trainParam.show = 10;2 T( m\" J7 K1 d( w4 D
    
12. net.trainParam.lr = 0.05;
    ! |* R- i2 i, v+ Z. J2 T
13. net.trainParam.goal = 1e-10;' `2 F* s9 y) u! X\" H
    
14. net.trainParam.epochs = 50000;' ^- K7 ]/ w% \6 I\" \! D
    
15. net = train(net,p,goal);0 u1 Y! ^' w% n0 ^' M
    
16. x=testsetdata';
    $ Y/ R$ \- s; B. Z7 v
17. y0=sim(net,p)

复制代码

下面是运行结果，请各位高手帮忙改改啊，主要是节点数那块儿不知道该怎么取值

2. Warning: NEWFF used in an obsolete way. 5 ?( o* f- C$ ]: s; }: f
   
4. > In obs_use at 18  E3 b% u6 G3 `0 D: n
   
6.   In newff>create_network at 127
   
7. % p  w\\" b( h9 Q8 y: {) ^# @% ~
8.   In newff at 102, S# @3 K( H% d4 V8 Y
   
10.   In test at 9
    
11. : }0 u. l3 {+ m, c# {& k
12.           See help for NEWFF to update calls to the new argument list.! M7 Y$ ~7 l3 _/ v. s! I/ o3 Z
    
14. 
    
15. 5 l6 G, t\\" Y. R  \
16. ??? Error using ==> trainlm at 109
    
17.   a) ^) r' {8 O# \$ ~# T) K6 p
18. Output data size does not match net.outputs{2}.size.4 l3 N9 p3 M5 q, ]
    
20. 
    
21. 8 Y0 f( T5 v8 x' F
22. Error in ==> network.train at 107
    
23. 1 z0 U1 @4 O' f1 Z, _, b
24. [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);
    
25. & ^( O7 i6 K6 w$ I
26. 
    
27. ! L, |5 z0 U+ h* P! M$ K
28. Error in ==> test at 159 c\\" \, _. r2 l& ]7 p
    
30. net = train(net,p,goal);

Warning: NEWFF used in an obsolete way. <font class=\\"jammer\\">5 ?( o* f- C$ ]: s; }: f</font><br /> > In obs_use at 18<font class=\\"jammer\\">  E3 b% u6 G3 `0 D: n</font><br />   In newff>create_network at 127<br /> <span style=\\"display:none\\">% p  w\\" b( h9 Q8 y: {) ^# @% ~</span>   In newff at 102<font class=\\"jammer\\">, S# @3 K( H% d4 V8 Y</font><br />   In test at 9<br /> <span style=\\"display:none\\">: }0 u. l3 {+ m, c# {& k</span>           See help for NEWFF to update calls to the new argument list.<font class=\\"jammer\\">! M7 Y$ ~7 l3 _/ v. s! I/ o3 Z</font><br /> <br /> <span style=\\"display:none\\">5 l6 G, t\\" Y. R  \</span> ??? Error using ==> trainlm at 109<br /> <span style=\\"display:none\\">  a) ^) r' {8 O# \$ ~# T) K6 p</span> Output data size does not match net.outputs{2}.size.<font class=\\"jammer\\">4 l3 N9 p3 M5 q, ]</font><br /> <br /> <span style=\\"display:none\\">8 Y0 f( T5 v8 x' F</span> Error in ==> network.train at 107<br /> <span style=\\"display:none\\">1 z0 U1 @4 O' f1 Z, _, b</span> [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);<br /> <span style=\\"display:none\\">& ^( O7 i6 K6 w$ I</span> <br /> <span style=\\"display:none\\">! L, |5 z0 U+ h* P! M$ K</span> Error in ==> test at 15<font class=\\"jammer\\">9 c\\" \, _. r2 l& ]7 p</font><br /> net = train(net,p,goal);

jiajinshan110

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