神经网络的算法程序.docVIP

!rs-输入层节点数 cs-输出层节点数 ys-隐层节点数 !t-学习速率 r-动量项 nx-循环次数 p-样本数 !ec-最小误差 pd-学习为0，验证为1 integer rs,cs,ys,p,q,l1,l2,l3,l4,l5,l6,l7,l8,l9,l10,pdx real*8 a(100000) LOGICAL look open(6,file=lwl.txt,status=old) INQUIRE(FILE=out.txt,EXIST=look) IF(look)THEN OPEN(8,FILE=out.txt,STATUS=OLD) CLOSE(8,STATUS=DELETE) end if OPEN(8,FILE=out.txt,STATUS=NEW) read(6,*)rs,cs,ys,p,t,r,nx,ec q=0 l1=2*rs*ys+1+ys l2=l1+2*cs*ys+cs l3=l2+rs*p l4=l3+cs*p l5=l4+rs*p+p l6=l5+cs*p l7=l6+ys*p+p l8=l7+cs*p l9=l8+p l10=l9+p*cs l11=l10+p*ys l12=l11+cs*p l13=l12+cs l14=l13+cs l15=l14+cs l16=l15+cs write(8,*)l1,l2,l3,l4,l5,l6,l7,l8,l9,l10 call input1(rs,cs,ys,a(1),a(l1)) call input2(rs,cs,p,a(l2),a(l3)) call input22(rs,cs,p,a(l2),a(l3),a(l4),a(l5),a(l12),a(l13),a(l14),a(l15)) do 5 i=1,nx q=q+1 write(8,(a))q= write(8,*)q do 10 j=1,p call sup1(rs,cs,ys,j,p,a(1),a(l1),a(l4),a(l6),a(l7)) call sup2(cs,ys,j,p,a(l5),a(l7),a(l6),a(l1),a(l8),a(l9),a(l10)) 10 continue call sup3(rs,cs,ys,p,t,r,a(l4),a(l6),a(l1),a(1),a(l9),a(l10)) do 30 j=1,p call sup1(rs,cs,ys,j,p,a(1),a(l1),a(l4),a(l6),a(l7)) 30 continue call fgyh(cs,p,a(l7),a(l12),a(l13),a(l14),a(l15),a(l11)) call pde(cs,p,ec,a(l3),a(l11),pdx) if(pdx.eq.1)then write(8,(a))the calculate is end write(8,*)q end if 5 continue write(8,(A))n=10000 end end !权值初始化 subroutine input1(rs,cs,ys,v,w) integer rs,cs,ys real*8 v(rs+1,ys,2),w(ys+1,cs,2) do 5 j=1,ys do 15 i=0,rs v(i,j,1)=0.1 15 continue 5 continue do 25 n=1,cs do 35 m=0,ys w(m,n,1)=0.1 35 continue 25 continue return end !输入训练样本 subroutine input2(rs,cs,p,x,d) integer rs,cs,p real*8 x(rs,p),d(cs,p) do 10 m=1,p read(6,*)(x(i,m),i=1,rs) 10 continue do 20 n=1,p read(6,*)(d(i,n),i=1,cs) 20 continue return end !归一化数据处理 subroutine input22(rs,cs,p,x,d,x1,d1,ax,ai,adx,adi) integer rs,cs,p real*8 x(rs,p),d(cs,p),x1(rs+1,p),d1(cs,p),ax(cs),ai(cs),adx(cs),adi(cs) do 10 i=1,rs ai(i)=100d0 ax(i)=0d0 do 20 j=1,p a=x(i,j) if(a.ge.ax(i))ax(i)=a if(a.le.ai(i))ai(i)=a 20 continue do 30 j=1,p x1(i,j)=(x(i,j)-ai(i))/(ax(i)-ai(i)