神经网络工具箱的使用

天零ヮ冰釋

1. clear$ C5 s# t) k+ Y6 |
   
2. load('trainset.mat');                   %加载训练集数据- Q$ [( q7 t9 y/ ~
   
3. load('testset.mat');                    %加载测试集数据
   1 u$ \- x! V: @( j
4. p1=trainsetdata([1:179],:);
   # ~  O: e3 J7 P# e$ G
5. p2=trainsetdata([180:300],:);: F2 x* B# L( c/ r5 q- e
   
6. p=[p1;p2]';
   3 d7 M9 n' d! A8 ]  V! W
7. pr=minmax(p);
   1 }! z! S+ Q\" N1 {0 L
8. goal=[ones(1,179),zeros(1,121);zeros(1,179),ones(1,121)];( D% c  B8 l$ \, a; V
   
9. net=newff(pr,[21,160],{'tansig','logsig'}); %隐含层节点和输出层节点
   3 Z\" @9 m9 f& n\" [
10.                                             %取值不太确定，请指教
    4 O3 ~/ }6 w3 \8 P8 H' F3 B
11. net.trainParam.show = 10;
    \" ~. q' i& ~( C0 x
12. net.trainParam.lr = 0.05;) u' Q4 X: }* Z! h9 R( i
    
13. net.trainParam.goal = 1e-10;
    5 u. m$ A' E7 w0 \; L\" L! h
14. net.trainParam.epochs = 50000;
    8 T8 h0 s# b7 h) c
15. net = train(net,p,goal);
    * M5 y/ q- \- r- @. N* j5 D
16. x=testsetdata';
    : K5 Z\" W\" ]: V: J' M
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. ! U2 X! ^3 G4 {6 s# O, K
   
4. > In obs_use at 184 R& A5 o; }/ ^4 g& |: z1 x
   
6.   In newff>create_network at 127, L& B- c( t8 r* J8 E
   
8.   In newff at 102
   
9. 8 L5 I5 y) P7 Q* ?2 r# C2 P
10.   In test at 97 k\\" ^# E. o/ Z$ R+ a1 _3 @$ ^
    
12.           See help for NEWFF to update calls to the new argument list.
    
13. 5 r4 P% k6 k  O- M0 `
14. - Y) g+ x8 l# f9 [+ y. b& u9 G; i) j
    
16. ??? Error using ==> trainlm at 1092 A4 z1 A. L\\" \# Y) z* A* Q
    
18. Output data size does not match net.outputs{2}.size.  p) t1 j7 e8 f8 g! V+ q: T
    
20. 
    
21. + G4 ^+ ^4 \/ z$ ^
22. Error in ==> network.train at 107\\" B4 f. [: T' k1 {7 f: E
    
24. [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);
    
25. : S1 l% e3 n5 a
26. , ^; C, H; {) e6 l  t& a
    
28. Error in ==> test at 15
    
29. & ]% g9 Q$ V2 i; `4 p0 X\\" O: ^
30. net = train(net,p,goal);

Warning: NEWFF used in an obsolete way. <font class=\\"jammer\\">! U2 X! ^3 G4 {6 s# O, K</font><br /> > In obs_use at 18<font class=\\"jammer\\">4 R& A5 o; }/ ^4 g& |: z1 x</font><br />   In newff>create_network at 127<font class=\\"jammer\\">, L& B- c( t8 r* J8 E</font><br />   In newff at 102<br /> <span style=\\"display:none\\">8 L5 I5 y) P7 Q* ?2 r# C2 P</span>   In test at 9<font class=\\"jammer\\">7 k\\" ^# E. o/ Z$ R+ a1 _3 @$ ^</font><br />           See help for NEWFF to update calls to the new argument list.<br /> <span style=\\"display:none\\">5 r4 P% k6 k  O- M0 `</span> <font class=\\"jammer\\">- Y) g+ x8 l# f9 [+ y. b& u9 G; i) j</font><br /> ??? Error using ==> trainlm at 109<font class=\\"jammer\\">2 A4 z1 A. L\\" \# Y) z* A* Q</font><br /> Output data size does not match net.outputs{2}.size.<font class=\\"jammer\\">  p) t1 j7 e8 f8 g! V+ q: T</font><br /> <br /> <span style=\\"display:none\\">+ G4 ^+ ^4 \/ z$ ^</span> Error in ==> network.train at 107<font class=\\"jammer\\">\\" B4 f. [: T' k1 {7 f: E</font><br /> [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);<br /> <span style=\\"display:none\\">: S1 l% e3 n5 a</span> <font class=\\"jammer\\">, ^; C, H; {) e6 l  t& a</font><br /> Error in ==> test at 15<br /> <span style=\\"display:none\\">& ]% g9 Q$ V2 i; `4 p0 X\\" O: ^</span> net = train(net,p,goal);

jiajinshan110

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