Tse Guan Tan wrote:
> How can I set the real valued vector problem without any range?
Stupidly, ECJ doesn't actually have a parameter that specifies that the
min and max values are unconstrained. But it's easy enough to get
around. Set the min value to:
And the max value to:
These are the lowest and highest possible double values (unless you get
> (unconstraint assignment problem)
> or any suggestion for the library I can use to evolve the Neural Network in
> JAVA programming?
That I do not. Though Weka is very well regarded for general machine