基于MATLAB的径向基网络源程序.docx

1、%一维输入，一维输出，逼近效果很好！1.基于聚类的RBF 网设计算法SamNum = 100; % 总样本数TestSamNum = 101; % 测试样本数InDim = 1; % 样本输入维数ClusterNum = 10; % 隐节点数，即聚类样本数Overlap = 1.0; % 隐节点重叠系数% 根据目标函数获得样本输入输出rand(state,sum(100*clock)NoiseVar = 0.1;Noise = NoiseVar*randn(1,SamNum);SamIn = 8*rand(1,SamNum)-4;SamOutNoNoise = 1.1*(1-SamIn+2*S

2、amIn.2).*exp(-SamIn.2/2);SamOut = SamOutNoNoise + Noise;TestSamIn = -4:0.08:4;TestSamOut = 1.1*(1-TestSamIn+2*TestSamIn.2).*exp(-TestSamIn.2/2);figurehold ongridplot(SamIn,SamOut,k+)plot(TestSamIn,TestSamOut,k-)xlabel(Input x);ylabel(Output y);Centers = SamIn(:,1:ClusterNum);NumberInClusters = zeros

3、(ClusterNum,1); % 各类中的样本数，初始化为零IndexInClusters = zeros(ClusterNum,SamNum); % 各类所含样本的索引号while 1,NumberInClusters = zeros(ClusterNum,1); % 各类中的样本数，初始化为零IndexInClusters = zeros(ClusterNum,SamNum); % 各类所含样本的索引号% 按最小距离原则对所有样本进行分类for i = 1:SamNumAllDistance = dist(Centers,SamIn(:,i);MinDist,Pos = min(AllD

4、istance);NumberInClusters(Pos) = NumberInClusters(Pos) + 1;IndexInClusters(Pos,NumberInClusters(Pos) = i;end% 保存旧的聚类中心OldCenters = Centers;for i = 1:ClusterNumIndex = IndexInClusters(i,1:NumberInClusters(i);Centers(:,i) = mean(SamIn(:,Index);end% 判断新旧聚类中心是否一致，是则结束聚类EqualNum = sum(sum(Centers=OldCent

5、ers);if EqualNum = InDim*ClusterNum,break,endend% 计算各隐节点的扩展常数（宽度）AllDistances = dist(Centers,Centers); % 计算隐节点数据中心间的距离（矩阵）Maximum = max(max(AllDistances); % 找出其中最大的一个距离for i = 1:ClusterNum % 将对角线上的0 替换为较大的值AllDistances(i,i) = Maximum+1;endSpreads = Overlap*min(AllDistances); % 以隐节点间的最小距离作为扩展常数% 计算各隐

6、节点的输出权值Distance = dist(Centers,SamIn); % 计算各样本输入离各数据中心的距离SpreadsMat = repmat(Spreads,1,SamNum);HiddenUnitOut = radbas(Distance./SpreadsMat); % 计算隐节点输出阵HiddenUnitOutEx = HiddenUnitOut ones(SamNum,1); % 考虑偏移W2Ex = SamOut*pinv(HiddenUnitOutEx); % 求广义输出权值W2 = W2Ex(:,1:ClusterNum); % 输出权值B2 = W2Ex(:,Clus

7、terNum+1); % 偏移% 测试TestDistance = dist(Centers,TestSamIn);TestSpreadsMat = repmat(Spreads,1,TestSamNum);TestHiddenUnitOut = radbas(TestDistance./TestSpreadsMat);TestNNOut = W2*TestHiddenUnitOut+B2;plot(TestSamIn,TestNNOut,k-)W2B22.基于梯度法的RBF 网设计算法SamNum = 100; % 训练样本数TargetSamNum = 101; % 测试样本数InDim

8、= 1; % 样本输入维数UnitNum = 10; % 隐节点数MaxEpoch = 5000; % 最大训练次数E0 = 0.9; % 目标误差% 根据目标函数获得样本输入输出rand(state,sum(100*clock)NoiseVar = 0.1;Noise = NoiseVar*randn(1,SamNum);SamIn = 8*rand(1,SamNum)-4;SamOutNoNoise = 1.1*(1-SamIn+2*SamIn.2).*exp(-SamIn.2/2);SamOut = SamOutNoNoise + Noise;TargetIn = -4:0.08:4;T

9、argetOut = 1.1*(1-TargetIn+2*TargetIn.2).*exp(-TargetIn.2/2);figurehold ongridplot(SamIn,SamOut,k+)plot(TargetIn,TargetOut,k-)xlabel(Input x);ylabel(Output y);Center = 8*rand(InDim,UnitNum)-4;SP = 0.2*rand(1,UnitNum)+0.1;W = 0.2*rand(1,UnitNum)-0.1;lrCent = 0.001; % 隐节点数据中心学习系数lrSP = 0.001; % 隐节点扩展常数学习系数lrW = 0.001; % 隐节点输出权值学习系数ErrHistory = ; % 用于记录每次参数调整后的训练误差for epoch = 1:MaxEpochAllDist = dist(Center,SamIn);SPMat = repmat(SP,1,SamNum);UnitOut = radbas(AllDist./SPMat);NetOut = W*UnitOut;Error = SamOut-NetOut;%停止学习判断SSE = sumsqr(Error)% 记录每次权值调整后的训练误差ErrHistory = ErrHistory SSE;if SSE