神经网络工具箱的使用

天零ヮ冰釋

1. clear
   : _9 {, ]/ O8 _$ b( [, b0 |3 t4 r4 ?
2. load('trainset.mat');                   %加载训练集数据1 o5 X\" L4 V3 y; h. U# z& Q1 @8 @) `
   
3. load('testset.mat');                    %加载测试集数据
   - p  z( p# p! w! n% A
4. p1=trainsetdata([1:179],:);0 k2 R/ l4 S4 O% n  y. Q
   
5. p2=trainsetdata([180:300],:);3 E* p5 D2 X7 B$ R) z/ P8 j\" u
   
6. p=[p1;p2]';
   2 n\" v2 B0 N! U9 k\" [& `% J+ X
7. pr=minmax(p);: c, \% D) b8 K2 ?, K
   
8. goal=[ones(1,179),zeros(1,121);zeros(1,179),ones(1,121)];
   $ }- G9 M$ l( j& {9 n, [
9. net=newff(pr,[21,160],{'tansig','logsig'}); %隐含层节点和输出层节点% I\" ?% q  D, V5 Q- v$ f, [
   
10.                                             %取值不太确定，请指教
    ! u6 F) b7 L, t+ `! q
11. net.trainParam.show = 10;/ b0 u0 W0 g  q9 a
    
12. net.trainParam.lr = 0.05;7 r/ K2 `# [  V
    
13. net.trainParam.goal = 1e-10;1 c! N\" g4 G4 {5 _
    
14. net.trainParam.epochs = 50000;2 g! M: N1 s1 z8 s; x. I
    
15. net = train(net,p,goal);
    7 i1 p5 M  y2 j
16. x=testsetdata';
    8 `  C/ e8 h+ N8 {2 r' r
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. 8 G2 {2 H5 I' j! U8 m, b$ B/ i+ t
4. > In obs_use at 18
   
5. $ w6 W: ^# i& Z; [  z4 ?
6.   In newff>create_network at 127
   
7. % R  a# r! P2 n% N. T/ L
8.   In newff at 102
   
9. # w. s) d+ e* I
10.   In test at 9+ j( Y% R; ^* t: B! O
    
12.           See help for NEWFF to update calls to the new argument list.$ _. ^  M9 }, q0 P, d! m# C
    
14. ) h  O  T+ x' z$ T( F6 O8 x
    
16. ??? Error using ==> trainlm at 109
    
17.   v* A0 c7 H4 @9 k( H3 G9 O
18. Output data size does not match net.outputs{2}.size.
    
19. , d\\" Y: @5 Z# m
20. \\" m  N4 C4 t: B/ z0 d
    
22. Error in ==> network.train at 107
    
23. 0 X2 g3 u6 O. e\\" D) V2 A' B
24. [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);
    
25. & O; q( e1 j. a- G6 b
26. 
    
27. + q! K. [( [/ K. o- ^* m' R
28. Error in ==> test at 15
    
29. 5 l4 {. t- d& {3 S/ u, Y3 u
30. net = train(net,p,goal);

Warning: NEWFF used in an obsolete way. <br /> <span style=\\"display:none\\">8 G2 {2 H5 I' j! U8 m, b$ B/ i+ t</span> > In obs_use at 18<br /> <span style=\\"display:none\\">$ w6 W: ^# i& Z; [  z4 ?</span>   In newff>create_network at 127<br /> <span style=\\"display:none\\">% R  a# r! P2 n% N. T/ L</span>   In newff at 102<br /> <span style=\\"display:none\\"># w. s) d+ e* I</span>   In test at 9<font class=\\"jammer\\">+ j( Y% R; ^* t: B! O</font><br />           See help for NEWFF to update calls to the new argument list.<font class=\\"jammer\\">$ _. ^  M9 }, q0 P, d! m# C</font><br /> <font class=\\"jammer\\">) h  O  T+ x' z$ T( F6 O8 x</font><br /> ??? Error using ==> trainlm at 109<br /> <span style=\\"display:none\\">  v* A0 c7 H4 @9 k( H3 G9 O</span> Output data size does not match net.outputs{2}.size.<br /> <span style=\\"display:none\\">, d\\" Y: @5 Z# m</span> <font class=\\"jammer\\">\\" m  N4 C4 t: B/ z0 d</font><br /> Error in ==> network.train at 107<br /> <span style=\\"display:none\\">0 X2 g3 u6 O. e\\" D) V2 A' B</span> [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);<br /> <span style=\\"display:none\\">& O; q( e1 j. a- G6 b</span> <br /> <span style=\\"display:none\\">+ q! K. [( [/ K. o- ^* m' R</span> Error in ==> test at 15<br /> <span style=\\"display:none\\">5 l4 {. t- d& {3 S/ u, Y3 u</span> net = train(net,p,goal);

jiajinshan110

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