神经网络工具箱的使用

天零ヮ冰釋

1. clear
   \" K4 u  s; V. ?) c) Z9 q
2. load('trainset.mat');                   %加载训练集数据
   ; K- n2 }  R' T- Z  |) T8 D! A, a
3. load('testset.mat');                    %加载测试集数据
   # K# A9 ?4 _' d# I8 y; l5 V
4. p1=trainsetdata([1:179],:);
   7 q1 ~- i6 m3 {, W2 \0 w3 s
5. p2=trainsetdata([180:300],:);: l3 @7 k. o) W* C' o8 ^: P; y
   
6. p=[p1;p2]';% r! e6 ^0 t0 e2 Y
   
7. pr=minmax(p);\" s5 _  F3 z) ^/ l3 \3 l\" S
   
8. goal=[ones(1,179),zeros(1,121);zeros(1,179),ones(1,121)];
   + K1 O; v! l) f: f( d\" A
9. net=newff(pr,[21,160],{'tansig','logsig'}); %隐含层节点和输出层节点$ ~2 F# ~5 T( I5 m& b0 a
   
10.                                             %取值不太确定，请指教1 W( J1 \2 X; K  E/ H
    
11. net.trainParam.show = 10;\" w$ w( x! W7 D$ V  m- x2 m& T% n
    
12. net.trainParam.lr = 0.05;
    \" `( P7 ^* q; ]$ p5 B
13. net.trainParam.goal = 1e-10;* _2 K\" o  R- N\" V
    
14. net.trainParam.epochs = 50000;
    2 V  b  G, M2 P
15. net = train(net,p,goal);
    & G, V$ H& F6 N! q& E0 c% }; p3 P. Q
16. x=testsetdata';
    - L$ Q8 F6 \5 S, x+ [! b
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.   x4 S) p$ s2 P& Z& F* c
4. > In obs_use at 18
   
5. ( v, h2 ?( M* L9 @. r
6.   In newff>create_network at 127
   
7. ) _, p0 O  s& h! y. f
8.   In newff at 102
   
9. ; S0 m1 J) M9 ]\\" p\\" G( |' n# F5 M* p
10.   In test at 9/ |! C/ m2 C* Q5 Z
    
12.           See help for NEWFF to update calls to the new argument list.
    
13.   O$ p' M0 B/ ~! c. e
14. ( A, U) I% F/ F* q- q5 N$ ?3 U
    
16. ??? Error using ==> trainlm at 109  ?# `1 B# f2 J8 Z1 Q$ i
    
18. Output data size does not match net.outputs{2}.size.+ K! K/ t$ P8 M* N
    
20. 
    
21. 0 K: y, |- W+ D( Z& J2 X- S8 b
22. Error in ==> network.train at 107
    
23. : _3 |5 ~- o- o% l( t7 b
24. [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);
    
25. \\" l) V$ r\\" m) ?\\" i1 f$ |  ~. p, B8 `
26. 
    
27. $ j3 {  S6 j0 p7 o0 D) \5 p9 T6 N
28. Error in ==> test at 153 z8 _, z5 Z* Y/ @# x
    
30. net = train(net,p,goal);

Warning: NEWFF used in an obsolete way. <br /> <span style=\\"display:none\\">  x4 S) p$ s2 P& Z& F* c</span> > In obs_use at 18<br /> <span style=\\"display:none\\">( v, h2 ?( M* L9 @. r</span>   In newff>create_network at 127<br /> <span style=\\"display:none\\">) _, p0 O  s& h! y. f</span>   In newff at 102<br /> <span style=\\"display:none\\">; S0 m1 J) M9 ]\\" p\\" G( |' n# F5 M* p</span>   In test at 9<font class=\\"jammer\\">/ |! C/ m2 C* Q5 Z</font><br />           See help for NEWFF to update calls to the new argument list.<br /> <span style=\\"display:none\\">  O$ p' M0 B/ ~! c. e</span> <font class=\\"jammer\\">( A, U) I% F/ F* q- q5 N$ ?3 U</font><br /> ??? Error using ==> trainlm at 109<font class=\\"jammer\\">  ?# `1 B# f2 J8 Z1 Q$ i</font><br /> Output data size does not match net.outputs{2}.size.<font class=\\"jammer\\">+ K! K/ t$ P8 M* N</font><br /> <br /> <span style=\\"display:none\\">0 K: y, |- W+ D( Z& J2 X- S8 b</span> Error in ==> network.train at 107<br /> <span style=\\"display:none\\">: _3 |5 ~- o- o% l( t7 b</span> [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);<br /> <span style=\\"display:none\\">\\" l) V$ r\\" m) ?\\" i1 f$ |  ~. p, B8 `</span> <br /> <span style=\\"display:none\\">$ j3 {  S6 j0 p7 o0 D) \5 p9 T6 N</span> Error in ==> test at 15<font class=\\"jammer\\">3 z8 _, z5 Z* Y/ @# x</font><br /> net = train(net,p,goal);

jiajinshan110

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