第二节：RBF、GRNN、PNN神经网络

1、RBF

输入层、隐含层、输出层

>> doc newrbe

Radial Basis Neural Networks

 

newrbe

Syntax

net = newrbe(P,T,spread)

2、GRNN (广义回归神经网络)

PNN（概率神经网络）

竞争层

 

重点函数解读：

• newrbe

– Design exact radial basis network

– net = newrbe(P,T,spread)

• newgrnn

– Design generalized regression neural network

– net = newgrnn(P,T,spread)

• newpnn

– Design probabilistic neural network

– net = newpnn(P,T,spread)

• cputime

– Elapsed CPU time

• round(ceil、fix、floor)

– Round to nearest integer

– Y = round(X)

• length(size)

– Length of vector

– n = length(X)

• find

– Find indices and values of nonzero elements

– [row,col] = find(X, ...)

• .* ./ .\ .^ …… vs * / \ ^ ……

– Multiplication (.*) right division (./) left division (.\) power (.^)

– matrix multiplication (*) matrix right division (/) matrix left division (\) matrix power (^)

 

示例:

RBF——近红外光谱汽油辛烷值预测

GRNN、PNN——鸢尾花种类识别

 

 

4个特征，150个对象

ind2vec(I) 创建稀疏矩阵每一列第几个为1

>> edit newgrnn

>> edit newpnn