ECJ-INTEREST-L Archives

July 2013

ECJ-INTEREST-L@LISTSERV.GMU.EDU

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

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

Print Reply
Sender:
ECJ Evolutionary Computation Toolkit <[log in to unmask]>
Date:
Mon, 22 Jul 2013 17:34:14 +0200
Reply-To:
ECJ Evolutionary Computation Toolkit <[log in to unmask]>
Subject:
MIME-Version:
1.0
Content-Transfer-Encoding:
8bit
In-Reply-To:
Content-Type:
text/plain; charset=utf-8
From:
Bojan Janisch <[log in to unmask]>
Parts/Attachments:
text/plain (37 lines)
Hey again,

I've found the problem. Like you warned earlier,
I got an infinite loop at my SimpleBreeder. 

I've checked for fitness, which would (even if the best of 
population has the worst fitness) never break out of the while(true) loop.

It could be that the standardized KozaFitness which I earlier used,
could break out of the loop, because it always had a fitness greater
than zero. That would explain why the problem occurred after I changed 
the fitness.

Well thanks again for your help Sean.

Greetings

Bojan

----- Ursprüngliche Mail -----
Von: "Bojan Janisch" <[log in to unmask]>
An: "ECJ Evolutionary Computation Toolkit" <[log in to unmask]>
Gesendet: Montag, 22. Juli 2013 11:44:56
Betreff: SizeFairCrossover-Breeding infinite

Hey Sean,

I've analyzed my infinite breeding problem with multiple ECJ runs. Do you know if the SizeFairCrossover-BreedingPipeline does reset the population?
Because if the best fitness of the second generation is the worst possible (means 0.0) it runs infinite long and breeds without stop. But the Breeding-Pipeline should
reset the entire Population if there are no Individuals with a FScore over zero, because all combinations would also generate a FScore of zero.

Do you know where I should modify the Breeding-Pipeline of SizeFairCrossover to reset the population if the FScore is the worst possible?

Greetings

Bojan

ATOM RSS1 RSS2