神经网络工具箱的使用

天零ヮ冰釋

1. clear1 R) O0 D1 S  `1 _1 C
   
2. load('trainset.mat');                   %加载训练集数据
   3 j\" S1 ^; Z9 D\" `- L
3. load('testset.mat');                    %加载测试集数据
   - S, b8 H+ ~5 e0 T5 ]$ m4 `( N
4. p1=trainsetdata([1:179],:);
   , I6 h. ^( A& L# J8 z1 M3 A& I
5. p2=trainsetdata([180:300],:);
   : E& b( Z1 l+ W' I
6. p=[p1;p2]';5 w& T9 H8 A+ I6 e
   
7. pr=minmax(p);\" t( W$ T\" T% [& F+ g
   
8. goal=[ones(1,179),zeros(1,121);zeros(1,179),ones(1,121)];' m, C7 _$ B# f- s
   
9. net=newff(pr,[21,160],{'tansig','logsig'}); %隐含层节点和输出层节点
   0 [) k# c# T5 g( Y\" F4 H: s: s+ R
10.                                             %取值不太确定，请指教6 y9 A3 I% ?) w3 F% ?1 y( o, ~
    
11. net.trainParam.show = 10;% z; R9 P5 s3 ~
    
12. net.trainParam.lr = 0.05;
    8 ^6 U3 W+ S( z, i3 l/ U1 \
13. net.trainParam.goal = 1e-10;1 q. ~' Q8 T, x+ ~3 g9 O) X( c
    
14. net.trainParam.epochs = 50000;
    % r' W5 u\" V9 a! J' Y$ ~) i- Q% O
15. net = train(net,p,goal);7 B1 R\" z7 m3 _( _
    
16. x=testsetdata';! a2 I  `' Q1 a7 F\" s
    
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. + {! X% s: E2 a
4. > In obs_use at 18
   
5. 6 o+ T0 O: H' d* c2 B5 v
6.   In newff>create_network at 127
   
7. : n( c2 Q' ?$ c  ?9 }* |
8.   In newff at 102
   
9. ' y6 u$ i- [( e% ]4 l$ l# v
10.   In test at 9& F# J6 y; F  r0 _- I
    
12.           See help for NEWFF to update calls to the new argument list.
    
13. 7 v# x( |( ?% G9 G, }
14. 2 e% _6 }, @( W) z( e
    
16. ??? Error using ==> trainlm at 109
    
17. 3 Z/ Y9 ?0 v! _# X6 ?' d9 U. _' z
18. Output data size does not match net.outputs{2}.size.
    
19.   X6 d0 J& `) G% j\\" w4 U4 Y- b
20. 
    
21. 0 Q2 P9 Z( u$ K2 z4 N: M( @& C\\" W
22. Error in ==> network.train at 107% }9 s9 r% k# ]
    
24. [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);
    
25. + ^( \\\" s, G4 d  o( O& P; w! s2 D( u
26. 9 ?! B+ ?% O8 C
    
28. Error in ==> test at 15
    
29. ! t; [0 S2 R  z- J  ^+ ^: v
30. net = train(net,p,goal);

Warning: NEWFF used in an obsolete way. <br /> <span style=\\"display:none\\">+ {! X% s: E2 a</span> > In obs_use at 18<br /> <span style=\\"display:none\\">6 o+ T0 O: H' d* c2 B5 v</span>   In newff>create_network at 127<br /> <span style=\\"display:none\\">: n( c2 Q' ?$ c  ?9 }* |</span>   In newff at 102<br /> <span style=\\"display:none\\">' y6 u$ i- [( e% ]4 l$ l# v</span>   In test at 9<font class=\\"jammer\\">& F# J6 y; F  r0 _- I</font><br />           See help for NEWFF to update calls to the new argument list.<br /> <span style=\\"display:none\\">7 v# x( |( ?% G9 G, }</span> <font class=\\"jammer\\">2 e% _6 }, @( W) z( e</font><br /> ??? Error using ==> trainlm at 109<br /> <span style=\\"display:none\\">3 Z/ Y9 ?0 v! _# X6 ?' d9 U. _' z</span> Output data size does not match net.outputs{2}.size.<br /> <span style=\\"display:none\\">  X6 d0 J& `) G% j\\" w4 U4 Y- b</span> <br /> <span style=\\"display:none\\">0 Q2 P9 Z( u$ K2 z4 N: M( @& C\\" W</span> Error in ==> network.train at 107<font class=\\"jammer\\">% }9 s9 r% k# ]</font><br /> [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);<br /> <span style=\\"display:none\\">+ ^( \\\" s, G4 d  o( O& P; w! s2 D( u</span> <font class=\\"jammer\\">9 ?! B+ ?% O8 C</font><br /> Error in ==> test at 15<br /> <span style=\\"display:none\\">! t; [0 S2 R  z- J  ^+ ^: v</span> net = train(net,p,goal);

jiajinshan110

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