[gecode-users] Help figuring out how to use float variables

[Mailing List Email]

Thanks. Yes, this is what I expected.
I will have to experiment more tomorrow.

But yes, the truncation problem I mention is essentially that Gecode finds
solution so close to each other that somehow, when put into the formula,
I'm guessing truncation errors cause the result to be the same for the
solutions. Mathematically, no matter how close to the value for the
solution, it is not possible to get the same result, which is why I said
they were the same solutions.


On 4 June 2013 16:26, Vincent Barichard <Vincent.Barichard at univ-angers.fr>wrote:

> Hi,
>
> It's not a truncating error, you get different solutions but very very
> close ones (only one
> float between two solutions), printing is not precise enough to
> distinguish them.
> Are you sure about your constraints ?
> It seems odd that you post only one constraint about your expression with
> NodeConfValues.
>
>
> I modified a little your script, maybe it's what you expected ?
>
> Best regards,
> Vincent
>
>
> 2013/6/4 Mailing List Email <mailinglistx at gmail.com>
>
>> I have a simple model which I am trying to use to find one, two or
>> possible three floating variables that satisfies certain solutions to a
>> formula. The problem I face is that Gecode keeps giving thousands of the
>> SAME solution, probably because of truncation errors.
>> I've looked in the MPG, but can't find any information regarding how to
>> do this, so if anyone has any suggestions or hints, I'd be grateful.
>> I'm using Gist to find and print the solutions.
>> I'm attaching the code.
>>
>> _______________________________________________
>> Gecode users mailing list
>> users at gecode.org
>> https://www.gecode.org/mailman/listinfo/gecode-users
>>
>>
>
>
> --
> Vincent Barichard         Université d'Angers (LERIA)
> Tel:  02 41 73 52 06      Département Informatique
> Fax: 02 41 73 50 73      H203
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.gecode.org/pipermail/users/attachments/20130604/53833ea9/attachment.html>