神经网络工具箱的使用

天零ヮ冰釋

1. clear# Z4 f: `2 b. ^2 K$ X; V3 j# M
   
2. load('trainset.mat');                   %加载训练集数据
     W; ]' p7 |/ ]2 X! d$ U  V
3. load('testset.mat');                    %加载测试集数据
   + ]# w, [5 M; h  K/ y& f
4. p1=trainsetdata([1:179],:);6 l# K. C! o+ r5 @( I9 l( N
   
5. p2=trainsetdata([180:300],:);. A$ d9 W* Y8 t2 d. a
   
6. p=[p1;p2]';
     C1 ?+ a# N) H$ l! u7 o
7. pr=minmax(p);: H% f% t  O& N% F9 @$ ]
   
8. goal=[ones(1,179),zeros(1,121);zeros(1,179),ones(1,121)];1 W9 r+ c: o& D: J: f+ |
   
9. net=newff(pr,[21,160],{'tansig','logsig'}); %隐含层节点和输出层节点0 x# a: {# A& N8 p# q
   
10.                                             %取值不太确定，请指教
    7 F  r* g/ c\" x, e' [
11. net.trainParam.show = 10;
    * z. l6 a. O# }, A$ S2 c0 @
12. net.trainParam.lr = 0.05;: z% s3 @+ h: J1 R+ K( S
    
13. net.trainParam.goal = 1e-10;
    0 t# @# b- \7 H/ e( v
14. net.trainParam.epochs = 50000;
    / S; f  _) s) r
15. net = train(net,p,goal);+ n. ^& p/ K2 m
    
16. x=testsetdata';
      q( D5 H/ m8 }) L+ P: z
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. 2 P7 a: Y& o& k+ A\\" C5 J7 T& Q
   
4. > In obs_use at 18/ |. u\\" o9 X  D8 P2 c% m- G4 a
   
6.   In newff>create_network at 127
   
7. 8 ~3 ]' i+ L1 K
8.   In newff at 102
   
9. / [! F- A\\" I6 N& z3 ]- B
10.   In test at 9
    
11. $ W2 w6 D7 L4 v
12.           See help for NEWFF to update calls to the new argument list.) K8 G1 z9 X. H! Z% }9 Y
    
14. 
    
15. 7 r1 H  U  f/ O4 m: v
16. ??? Error using ==> trainlm at 109$ s7 b- O0 y4 l/ G* ?
    
18. Output data size does not match net.outputs{2}.size.
    
19. 1 D: N; }* l9 E0 U- A
20. 7 n/ Y. }6 U; A! N
    
22. Error in ==> network.train at 107
    
23. : v1 `, |  B1 @$ c$ W7 _
24. [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);
    
25. 9 A% j  k: G; Q% Z
26. / ]! \2 V\\" t9 E- ?& d2 i& ~
    
28. Error in ==> test at 15
    
29. $ F5 {! L) |6 D& N1 L% M7 v
30. net = train(net,p,goal);

Warning: NEWFF used in an obsolete way. <font class=\\"jammer\\">2 P7 a: Y& o& k+ A\\" C5 J7 T& Q</font><br /> > In obs_use at 18<font class=\\"jammer\\">/ |. u\\" o9 X  D8 P2 c% m- G4 a</font><br />   In newff>create_network at 127<br /> <span style=\\"display:none\\">8 ~3 ]' i+ L1 K</span>   In newff at 102<br /> <span style=\\"display:none\\">/ [! F- A\\" I6 N& z3 ]- B</span>   In test at 9<br /> <span style=\\"display:none\\">$ W2 w6 D7 L4 v</span>           See help for NEWFF to update calls to the new argument list.<font class=\\"jammer\\">) K8 G1 z9 X. H! Z% }9 Y</font><br /> <br /> <span style=\\"display:none\\">7 r1 H  U  f/ O4 m: v</span> ??? Error using ==> trainlm at 109<font class=\\"jammer\\">$ s7 b- O0 y4 l/ G* ?</font><br /> Output data size does not match net.outputs{2}.size.<br /> <span style=\\"display:none\\">1 D: N; }* l9 E0 U- A</span> <font class=\\"jammer\\">7 n/ Y. }6 U; A! N</font><br /> Error in ==> network.train at 107<br /> <span style=\\"display:none\\">: v1 `, |  B1 @$ c$ W7 _</span> [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);<br /> <span style=\\"display:none\\">9 A% j  k: G; Q% Z</span> <font class=\\"jammer\\">/ ]! \2 V\\" t9 E- ?& d2 i& ~</font><br /> Error in ==> test at 15<br /> <span style=\\"display:none\\">$ F5 {! L) |6 D& N1 L% M7 v</span> net = train(net,p,goal);

jiajinshan110

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