radbasn

Normalized radial basis transfer function

Graph and Symbol

Radial basis transfer function. The function has a maximum of 1 when its input is 0. As the distance between input n and 0 decreases, the output a increases.

Syntax

A = radbasn(N,FP)

Description

radbasn is a neural transfer function. Transfer functions calculate a layer’s output from its net input. This function is equivalent to radbas, except that output vectors are normalized by dividing by the sum of the pre-normalized values.

A = radbasn(N,FP) takes one or two inputs,

`N`	`S`-by-`Q` matrix of net input (column) vectors
`FP`	Struct of function parameters (ignored)

and returns A, an S-by-Q matrix of the radial basis function applied to each element of N.

Examples

Here six random 3-element vectors are passed through the radial basis transform and normalized.

n = rand(3,6)
a = radbasn(n)

Assign this transfer function to layer i of a network.

net.layers{i}.transferFcn = 'radbasn';

Algorithms

a = radbasn(n) = exp(-n^2) / sum(exp(-n^2))

Version History

Introduced in R2010b