[eclipse-clp-users] FW: code problem

From: Bogdan Tanasa <g-bogta_at_ida.liu.se>
Date: Fri, 8 Oct 2010 22:53:29 +0200
Now it seems to work a little bit better but I have problem with the gcd /3
predicate.

 

According to the manual: In coroutining mode, if Number1 or Number2 are
uninstantiated, the call is delayed until these variables are instantiated. 

 

.. I still get the instantiation fault error. How can I switch to
coroutining mode and what that means ?

 

Best regards,

Bogdan.

 

From: Bogdan Tanasa [mailto:g-bogta_at_ida.liu.se] 
Sent: 8 octombrie 2010 18:08
To: 'eclipse-clp-users_at_lists.sourceforge.net'
Subject: FW: code problem

 

I come back with my problemJ. I was able to find an workaround to obtain
what I want.

 

I am running the bb_min to get the optimal solution but when asking the
solver for other solutions (MORE) I get solutions which a lower cost the the
initial one. For me it seems that bb_min is not actually computing the
optimal solution. 

 

What can be the problem?

 

Best regards,

Bogdan.

 

From: Bogdan Tanasa [mailto:g-bogta_at_ida.liu.se] 
Sent: 8 octombrie 2010 14:20
To: 'eclipse-clp-users_at_lists.sourceforge.net'
Subject: code problem

 

Hi guys,

 

I am trying to solve a problem and I have written the constraints but I
observe a strange behavior.

I have some Boolean variables and I compute some periods like this: T / x
where x is the Boolean variable. I am doing this because I want to get the
minimum period for only those variables which are equal with 1. So in the
case when all are 0 I am expecting to get +1.0Inf as the minimum value.

When I print the constraints I get an interval instead of one value.

 

What can be the cause?

 

 

Best regards,

Bogdan.






Received on Fri Oct 08 2010 - 20:55:16 CEST

This archive was generated by hypermail 2.2.0 : Thu Feb 02 2012 - 02:31:58 CET