ECJ-INTEREST-L Archives

June 2005

ECJ-INTEREST-L@LISTSERV.GMU.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Sean Luke <[log in to unmask]>
Reply To:
ECJ Evolutionary Computation Toolkit <[log in to unmask]>
Date:
Tue, 14 Jun 2005 08:50:52 -0400
Content-Type:
text/plain
Parts/Attachments:
text/plain (35 lines)
On Jun 14, 2005, at 1:07 PM, Artur Matos wrote:

> I tried to compile both Uniform.java and GPTreeConstraints.java but I
> am getting errors like this:
>
> GPTreeConstraints.java:94: cannot find symbol
> symbol  : variable treeConstraintRepository
> location: class ec.gp.GPInitializer

Oops, this is part of the new ECJ coming out in a bit.  Sorry!  Just
use your copy of GPTreeConstraints but change error(...) to
warning(...) as you had before.

> Exception in thread "main" java.lang.ArithmeticException: Distribution
> has all zero probabilities
>
>         at
> ec.util.RandomChoice.organizeDistribution(RandomChoice.java:139)
>         at
> ec.util.RandomChoice.organizeDistribution(RandomChoice.java:111)
>         at ec.gp.build.Uniform.preprocess(Uniform.java:283)

You're not using the new Uniform.java I provided in the email.

> Anyway, the number of trees reported with my params file (only 1 for
> type 0. size 3, and only 1 for type 1, size 1) are definitely strange.
> I will try to check what the problem is meanwhile. What exactly does
> the number of trees mean? Is the number of possible permutations for
> that size and type?

Yes.  It's the number of possible trees of the given size and whose
root returns the given type.

Sean

ATOM RSS1 RSS2