It's pretty simple. You just need to define your own Problem java source
file (in tutorial 4 it's called MultiValuedRegression.java) in which the
evaluation takes place. In the setup-method you can read in your data and
store it in your current GPProblem instance so that you can use it for
evaluation in the evaluate-method.
On Wed, 23 May 2012 09:33:00 +0300, Simion Cristian
<[log in to unmask]> wrote:
>I am new to ECJ and GP and i would like to ask some question. I have a
>problem of symbolic regression similar with the one in tutorial 4 but i
>would like to use my data (i have experimental data in files).How can use
>my data from files in the problem.