神经网络工具箱的使用

天零ヮ冰釋

1. clear  v1 p# C5 m. @- {
   
2. load('trainset.mat');                   %加载训练集数据$ a% ~# \: L' y
   
3. load('testset.mat');                    %加载测试集数据8 W8 r4 ^9 \% `' N, k% W
   
4. p1=trainsetdata([1:179],:);
   9 B: i3 _* l5 B9 y: T* u( l* A* {
5. p2=trainsetdata([180:300],:);
   * |# w$ M  g8 ]
6. p=[p1;p2]';+ u- G. ]0 A9 f1 {
   
7. pr=minmax(p);4 o7 k; A; L) w% c% h
   
8. goal=[ones(1,179),zeros(1,121);zeros(1,179),ones(1,121)];2 ~# F7 p3 G# i- Z% x! ]' r8 J7 {, v) n
   
9. net=newff(pr,[21,160],{'tansig','logsig'}); %隐含层节点和输出层节点
   5 T+ j) K9 e6 {8 j! v
10.                                             %取值不太确定，请指教
    ! [2 o' e; ^% Q4 m5 v0 R' x, }
11. net.trainParam.show = 10;+ U; u- N, f, H8 M
    
12. net.trainParam.lr = 0.05;
    * V& @/ {& W0 @# D! B
13. net.trainParam.goal = 1e-10;! r) Y: R. |' c3 ~0 Q& [4 @1 C6 Y
    
14. net.trainParam.epochs = 50000;
    $ T# S6 p% y; T' Z( q
15. net = train(net,p,goal);9 w' u* x# ^. C6 x
    
16. x=testsetdata';
    , p: I2 Y. x/ Z. l) S: |! P
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.
   
3. 8 s) t6 y: `6 }- {+ M
4. > In obs_use at 187 \  ~- Q! R; t% D7 {9 A
   
6.   In newff>create_network at 127
   
7. ! i4 R; g! l, |7 p
8.   In newff at 102, W) C) u0 ~) E
   
10.   In test at 9) A% l& {$ K, f5 c# S' Q+ f0 O  U( Q
    
12.           See help for NEWFF to update calls to the new argument list.: n: ^/ e9 M9 b; ^4 g+ S
    
14. 
    
15. - S; `+ J& j' x9 c: l6 m7 f. H
16. ??? Error using ==> trainlm at 109% _4 U3 g% ~+ ?' b2 _9 Q+ r
    
18. Output data size does not match net.outputs{2}.size.( ~% P% s% v: `3 j/ p8 Y) i6 C# o
    
20. 
    
21. + K  f. ]/ G, S( c* ], y2 `
22. Error in ==> network.train at 107
    
23. * _1 m! h# V6 H
24. [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);
    
25. + {$ E5 A% P; C+ p3 m
26. - _+ b/ S% y) P5 g6 u
    
28. Error in ==> test at 15* V. Z5 ~+ i  y
    
30. net = train(net,p,goal);

Warning: NEWFF used in an obsolete way. <br /> <span style=\\"display:none\\">8 s) t6 y: `6 }- {+ M</span> > In obs_use at 18<font class=\\"jammer\\">7 \  ~- Q! R; t% D7 {9 A</font><br />   In newff>create_network at 127<br /> <span style=\\"display:none\\">! i4 R; g! l, |7 p</span>   In newff at 102<font class=\\"jammer\\">, W) C) u0 ~) E</font><br />   In test at 9<font class=\\"jammer\\">) A% l& {$ K, f5 c# S' Q+ f0 O  U( Q</font><br />           See help for NEWFF to update calls to the new argument list.<font class=\\"jammer\\">: n: ^/ e9 M9 b; ^4 g+ S</font><br /> <br /> <span style=\\"display:none\\">- S; `+ J& j' x9 c: l6 m7 f. H</span> ??? Error using ==> trainlm at 109<font class=\\"jammer\\">% _4 U3 g% ~+ ?' b2 _9 Q+ r</font><br /> Output data size does not match net.outputs{2}.size.<font class=\\"jammer\\">( ~% P% s% v: `3 j/ p8 Y) i6 C# o</font><br /> <br /> <span style=\\"display:none\\">+ K  f. ]/ G, S( c* ], y2 `</span> Error in ==> network.train at 107<br /> <span style=\\"display:none\\">* _1 m! h# V6 H</span> [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);<br /> <span style=\\"display:none\\">+ {$ E5 A% P; C+ p3 m</span> <font class=\\"jammer\\">- _+ b/ S% y) P5 g6 u</font><br /> Error in ==> test at 15<font class=\\"jammer\\">* V. Z5 ~+ i  y</font><br /> net = train(net,p,goal);

jiajinshan110

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