神经网络工具箱的使用

天零ヮ冰釋

1. clear- c5 C- ]; |' ?3 v9 k6 m; l8 ?, p: c
   
2. load('trainset.mat');                   %加载训练集数据* I3 w5 C6 Z& r5 g0 O3 W
   
3. load('testset.mat');                    %加载测试集数据
   \" `7 O# t5 z2 }. H- O) \/ Q
4. p1=trainsetdata([1:179],:);
   , d( G# d% B; j) V
5. p2=trainsetdata([180:300],:);
   ; K\" K' |/ ?- f6 U9 y) B
6. p=[p1;p2]';
   9 p# Y  B2 @# K6 J8 B
7. pr=minmax(p);
   # P9 G! Z$ F* P
8. goal=[ones(1,179),zeros(1,121);zeros(1,179),ones(1,121)];
   2 L! U- u/ D& t' p( U- y; Q/ I
9. net=newff(pr,[21,160],{'tansig','logsig'}); %隐含层节点和输出层节点
   : J9 M2 ~* l+ L9 w' [2 S. [& u
10.                                             %取值不太确定，请指教$ g1 s) u+ a$ D, w
    
11. net.trainParam.show = 10;2 H+ Y& a5 u\" B\" w0 h7 A
    
12. net.trainParam.lr = 0.05;: \4 ~5 `* G; \' c2 ~. ~
    
13. net.trainParam.goal = 1e-10;
    ' l& \* I$ N, C7 U- E' ^
14. net.trainParam.epochs = 50000;$ ^- P. l8 K- P0 b5 F
    
15. net = train(net,p,goal);
    & _) S9 F8 V: S) v5 k1 P
16. x=testsetdata';# j# l8 H. I8 K- 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. 1 M/ S* G: ]' o4 V. Q- e
   
4. > In obs_use at 18
   
5. - h  H8 k\\" w: a8 s% ~& O
6.   In newff>create_network at 127
   
7. ' l\\" T. B/ {' x, ]
8.   In newff at 102
   
9. + p) I# t3 |: \\\" w( h5 g# M\\" N
10.   In test at 9
    
11. & W; I$ i; d7 m# A
12.           See help for NEWFF to update calls to the new argument list.
    
13. 0 t) Z4 {; N+ f7 A$ x! X
14. 
    
15. \\" y. I7 u: [; y& _\\" G4 ?8 H# r
16. ??? Error using ==> trainlm at 109# K; X! J/ w& |, j
    
18. Output data size does not match net.outputs{2}.size.
    
19. + _1 G& ?' i) w6 M\\" v$ s\\" ]
20. 
    
21. $ y' s1 j+ y$ N$ j# N+ R
22. Error in ==> network.train at 107' n' u- j. m; ~6 H
    
24. [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);
    
25. & z1 T- s) w+ P+ P
26. ' ~  }1 ?- `  @& t% x2 ~3 K
    
28. Error in ==> test at 15; J5 I6 m3 y& ]! a+ C
    
30. net = train(net,p,goal);

Warning: NEWFF used in an obsolete way. <font class=\\"jammer\\">1 M/ S* G: ]' o4 V. Q- e</font><br /> > In obs_use at 18<br /> <span style=\\"display:none\\">- h  H8 k\\" w: a8 s% ~& O</span>   In newff>create_network at 127<br /> <span style=\\"display:none\\">' l\\" T. B/ {' x, ]</span>   In newff at 102<br /> <span style=\\"display:none\\">+ p) I# t3 |: \\\" w( h5 g# M\\" N</span>   In test at 9<br /> <span style=\\"display:none\\">& W; I$ i; d7 m# A</span>           See help for NEWFF to update calls to the new argument list.<br /> <span style=\\"display:none\\">0 t) Z4 {; N+ f7 A$ x! X</span> <br /> <span style=\\"display:none\\">\\" y. I7 u: [; y& _\\" G4 ?8 H# r</span> ??? Error using ==> trainlm at 109<font class=\\"jammer\\"># K; X! J/ w& |, j</font><br /> Output data size does not match net.outputs{2}.size.<br /> <span style=\\"display:none\\">+ _1 G& ?' i) w6 M\\" v$ s\\" ]</span> <br /> <span style=\\"display:none\\">$ y' s1 j+ y$ N$ j# N+ R</span> Error in ==> network.train at 107<font class=\\"jammer\\">' n' u- j. m; ~6 H</font><br /> [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);<br /> <span style=\\"display:none\\">& z1 T- s) w+ P+ P</span> <font class=\\"jammer\\">' ~  }1 ?- `  @& t% x2 ~3 K</font><br /> Error in ==> test at 15<font class=\\"jammer\\">; J5 I6 m3 y& ]! a+ C</font><br /> net = train(net,p,goal);

jiajinshan110

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