radbas (Neural Network Toolbox)

Radial basis transfer function

A = radbas(N)
info = radbas(code)

radbas is a transfer function. Transfer functions calculate a layer's output from its net input.

radbas(N) takes one input,

N - S x Q matrix of net input (column) vectors.

and returns each element of N passed through a radial basis function.

radbas(code) returns useful information for each code string:

'deriv' - Name of derivative function.

'name' - Full name.

'output' - Output range.

'active' - Active input range.

Examples

Here we create a plot of the radbas transfer function.

n = -5:0.1:5;
a = radbas(n);
plot(n,a)

You can create a standard network that uses radbas by calling newpnn or newgrnn.

To change a network so that a layer uses radbas, set net.layers{i}.transferFcn to 'radbas'.

In either case, call sim to simulate the network with radbas. See newpnn or newgrnn for simulation examples.

radbas(N) calculates its output with according to:

a = exp(-n2)