神经网络工具箱的使用

天零ヮ冰釋

1. clear  f- b  \( K# a: R: l
   
2. load('trainset.mat');                   %加载训练集数据) w8 T$ C8 H! p8 U0 v9 g' |
   
3. load('testset.mat');                    %加载测试集数据/ e/ B' o* |5 L- |! a& e
   
4. p1=trainsetdata([1:179],:);9 o) G1 C6 M1 a, C
   
5. p2=trainsetdata([180:300],:);
   + `% ^. `/ b) ]: ?; y
6. p=[p1;p2]';( @% t# c, ?) C3 G* V1 ]
   
7. pr=minmax(p);
   4 Z+ c6 Y$ b) y5 R: \
8. goal=[ones(1,179),zeros(1,121);zeros(1,179),ones(1,121)];
   . ^% o: f9 b( G. O% V
9. net=newff(pr,[21,160],{'tansig','logsig'}); %隐含层节点和输出层节点' ~* Y% }  u% h  ?: ^' o% [
   
10.                                             %取值不太确定，请指教6 I, b  z* |) V3 q) ~% E3 j
    
11. net.trainParam.show = 10;
    6 L2 }/ j* K( S: m: C1 m
12. net.trainParam.lr = 0.05;9 k9 x5 g6 @  o
    
13. net.trainParam.goal = 1e-10;+ Z' F5 F; Q9 j9 ]8 o6 o  A
    
14. net.trainParam.epochs = 50000;! f) n1 l$ C4 _( f& v4 E/ Z; s  z
    
15. net = train(net,p,goal);
      c/ D9 @- D; t\" u6 \6 ^; i% t- N
16. x=testsetdata';
    * \1 M6 h$ a; N4 A% k7 \
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 u: X7 S% }& A) @
   
4. > In obs_use at 18
   
5.   Z: j  ^( z- o; Y  I8 r' u* X: T
6.   In newff>create_network at 1275 P. _0 U9 Y' x: g
   
8.   In newff at 102
   
9.   d7 x9 P7 Y5 r5 j* X
10.   In test at 9- z2 @1 d+ `; K\\" u9 p7 w
    
12.           See help for NEWFF to update calls to the new argument list.7 E: K! k\\" ~% a% t  j
    
14. 9 \' R0 ^: ]; q% G4 e# W+ F0 G/ j+ K
    
16. ??? Error using ==> trainlm at 109
    
17. 5 S7 e) J+ L8 y
18. Output data size does not match net.outputs{2}.size.
    
19. , j9 [; m! B5 P6 R* R# {& s
20. . R( E\\" z/ L6 ]$ P/ C
    
22. Error in ==> network.train at 1076 \  F8 W& ]9 m\\" M
    
24. [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);
    
25. ( `9 D; F( c& |) \. `) W
26. 
    
27. 1 {! A0 l; L6 b! B8 O+ o% J% u* [
28. Error in ==> test at 15
    
29. ) u, L) T$ _; i+ p1 r2 y& {& w
30. net = train(net,p,goal);

Warning: NEWFF used in an obsolete way. <font class=\\"jammer\\">- ?2 u: X7 S% }& A) @</font><br /> > In obs_use at 18<br /> <span style=\\"display:none\\">  Z: j  ^( z- o; Y  I8 r' u* X: T</span>   In newff>create_network at 127<font class=\\"jammer\\">5 P. _0 U9 Y' x: g</font><br />   In newff at 102<br /> <span style=\\"display:none\\">  d7 x9 P7 Y5 r5 j* X</span>   In test at 9<font class=\\"jammer\\">- z2 @1 d+ `; K\\" u9 p7 w</font><br />           See help for NEWFF to update calls to the new argument list.<font class=\\"jammer\\">7 E: K! k\\" ~% a% t  j</font><br /> <font class=\\"jammer\\">9 \' R0 ^: ]; q% G4 e# W+ F0 G/ j+ K</font><br /> ??? Error using ==> trainlm at 109<br /> <span style=\\"display:none\\">5 S7 e) J+ L8 y</span> Output data size does not match net.outputs{2}.size.<br /> <span style=\\"display:none\\">, j9 [; m! B5 P6 R* R# {& s</span> <font class=\\"jammer\\">. R( E\\" z/ L6 ]$ P/ C</font><br /> Error in ==> network.train at 107<font class=\\"jammer\\">6 \  F8 W& ]9 m\\" M</font><br /> [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);<br /> <span style=\\"display:none\\">( `9 D; F( c& |) \. `) W</span> <br /> <span style=\\"display:none\\">1 {! A0 l; L6 b! B8 O+ o% J% u* [</span> Error in ==> test at 15<br /> <span style=\\"display:none\\">) u, L) T$ _; i+ p1 r2 y& {& w</span> net = train(net,p,goal);

jiajinshan110

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