How can I use the PREMNMX function within the Neural Networks Toolbox to scale my data between 0 and 1, rather than -1 and 1?

I would like to use the PREMNMX function within Neural Networks Toolbox to scale my data between 0 and 1, rather than -1 and 1.

Answer 1

[MathWorks Support Team]

This can be achieved by scaling the data returned from the PREMNMX function. For example, if you want the data in the vector "X" to be scaled between 0 and 1, rather than -1 and 1, use the following code:

x = premnmx(X);

y = (x+1)/2

In general, if you want the data in "X" to be scaled between "a" and "b", use the generalized code:

x = premnmx(X);

y = (x+1)*(b-a)/2+a