神经网络工具箱的使用

天零ヮ冰釋

1. clear% g. g7 G4 [! T- [
   
2. load('trainset.mat');                   %加载训练集数据
   : D9 i$ h1 |6 X, n( r+ V
3. load('testset.mat');                    %加载测试集数据
   ; o9 K3 Y- [) R+ U1 C; Z
4. p1=trainsetdata([1:179],:);
   \" i9 y: c# a/ r! L1 x! a# u
5. p2=trainsetdata([180:300],:);) Q* p- ]; c. A; u+ h2 u% i
   
6. p=[p1;p2]';- t) Q\" G& W7 W, m9 z. K
   
7. pr=minmax(p);
   4 I\" l! I- z) N8 Q: @5 b6 I
8. goal=[ones(1,179),zeros(1,121);zeros(1,179),ones(1,121)];
   5 @0 w' V: p1 e0 W, C( j; b
9. net=newff(pr,[21,160],{'tansig','logsig'}); %隐含层节点和输出层节点1 [5 J2 N  P* F  @$ H
   
10.                                             %取值不太确定，请指教
    . J6 }' B3 y5 F
11. net.trainParam.show = 10;
    \" p: s# s0 L; E9 r5 N  o
12. net.trainParam.lr = 0.05;) j; i( w% A& B8 Y0 X- }& l
    
13. net.trainParam.goal = 1e-10;' w) s: a* z* m6 s/ e
    
14. net.trainParam.epochs = 50000;
    ' ~( o1 s$ f, g' }( M: k+ q5 j# U
15. net = train(net,p,goal);
    & U! ]7 Y\" f0 V2 S6 W
16. x=testsetdata';3 x9 Y; g\" ^! i& \! U5 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. 5 g/ R* }2 n+ m' r
   
4. > In obs_use at 180 h( Z9 b7 X0 i) I9 Z; R
   
6.   In newff>create_network at 127
   
7. # K# d6 d, s& l
8.   In newff at 1026 f5 S7 u6 e3 ~) o\\" P
   
10.   In test at 93 C* [0 o+ t( L2 f4 u( x
    
12.           See help for NEWFF to update calls to the new argument list.
    
13. \\" ~2 w  y2 G7 I0 y' a  `
14. 
    
15. % c3 H1 x& U; F( [2 R2 c2 M0 F
16. ??? Error using ==> trainlm at 1093 a* y\\" w0 M3 B+ M
    
18. Output data size does not match net.outputs{2}.size.; H  J# J- I0 G. J* L4 Z. q
    
20. 
    
21. 7 ]) Z' J* h: c$ @9 W9 `/ ]
22. Error in ==> network.train at 1076 V) ^% M( O# `( ]5 t8 g/ {
    
24. [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);% a% E. s8 G7 Z
    
26. 
    
27. * @2 B) w+ i- i( }5 Z1 P
28. Error in ==> test at 15
    
29. % z! e& m  {. l# ^
30. net = train(net,p,goal);

Warning: NEWFF used in an obsolete way. <font class=\\"jammer\\">5 g/ R* }2 n+ m' r</font><br /> > In obs_use at 18<font class=\\"jammer\\">0 h( Z9 b7 X0 i) I9 Z; R</font><br />   In newff>create_network at 127<br /> <span style=\\"display:none\\"># K# d6 d, s& l</span>   In newff at 102<font class=\\"jammer\\">6 f5 S7 u6 e3 ~) o\\" P</font><br />   In test at 9<font class=\\"jammer\\">3 C* [0 o+ t( L2 f4 u( x</font><br />           See help for NEWFF to update calls to the new argument list.<br /> <span style=\\"display:none\\">\\" ~2 w  y2 G7 I0 y' a  `</span> <br /> <span style=\\"display:none\\">% c3 H1 x& U; F( [2 R2 c2 M0 F</span> ??? Error using ==> trainlm at 109<font class=\\"jammer\\">3 a* y\\" w0 M3 B+ M</font><br /> Output data size does not match net.outputs{2}.size.<font class=\\"jammer\\">; H  J# J- I0 G. J* L4 Z. q</font><br /> <br /> <span style=\\"display:none\\">7 ]) Z' J* h: c$ @9 W9 `/ ]</span> Error in ==> network.train at 107<font class=\\"jammer\\">6 V) ^% M( O# `( ]5 t8 g/ {</font><br /> [net,tr] = feval(net.trainFcn,net,X,T,Xi,Ai,EW,net.trainParam);<font class=\\"jammer\\">% a% E. s8 G7 Z</font><br /> <br /> <span style=\\"display:none\\">* @2 B) w+ i- i( }5 Z1 P</span> Error in ==> test at 15<br /> <span style=\\"display:none\\">% z! e& m  {. l# ^</span> net = train(net,p,goal);

jiajinshan110

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