Op 07-09-12 17:28, Michiel Vermandel schreef:
Thanks for the response.

Can you tell me if the fix will have any effect on the end-result?
I wouldn't expect it to get much better, unless the tabu size was too high and stifling the move selection.
But it could be better, try it out. If you do, please let us the know results here (good or bad).

Thanks

 
-----------------
http://www.codessentials.com - Your essential software, for free!
Follow us at http://twitter.com/#!/Codessentials
From: Geoffrey De Smet <ge0ffrey.spam@gmail.com>
To: Michiel Vermandel <mvermand@yahoo.com>; Rules Users List <rules-users@lists.jboss.org>
Sent: Friday, September 7, 2012 4:01 PM
Subject: Re: Cancelled step ... : there is no doable move. Terminating phase early.


Op 07-09-12 15:03, Michiel Vermandel schreef:
Hi,

I see a warning while solving and I don't really know where to start in finding the cause:

Here is the debug output of the solver:

>>> I see that step 0 already results in a hard-score 0 (!?)

INFO : [2012-09-07 14:56:21,969] Solving started: time spend (90), score (null), new best score (null), random seed (0).
DEBUG: [2012-09-07 14:56:22,376]     Step index (0), time spend (499), score (0hard/99soft), initialized planning entity (Inspection of 112474.I#0 -1934359865).
DEBUG: [2012-09-07 14:56:22,704]     Step index (1), time spend (827), score (0hard/198soft), initialized planning entity (Inspection of 112474.I#1 -1934359864).
DEBUG: [2012-09-07 14:56:22,990]     Step index (2), time spend (1113), score (0hard/297soft), initialized planning entity (Inspection of 112474.I#2 -1934359863).

>>> then after a few hundred steps the hard-score starts to drop?!

DEBUG: [2012-09-07 14:56:30,166]     Step index (260), time spend (8289), score (0hard/24974soft), initialized planning entity (Inspection of 61457.I#0 1155198787).
DEBUG: [2012-09-07 14:56:30,189]     Step index (261), time spend (8312), score (0hard/25049soft), initialized planning entity (Inspection of 61457.I#1 1155198788).
DEBUG: [2012-09-07 14:56:30,210]     Step index (262), time spend (8333), score (0hard/25148soft), initialized planning entity (Inspection of 46045.I#0 1313733879).
DEBUG: [2012-09-07 14:56:30,234]     Step index (263), time spend (8357), score (-1hard/25223soft), initialized planning entity (Inspection of 46045.I#1 1313733880).
DEBUG: [2012-09-07 14:56:30,255]     Step index (264), time spend (8378), score (-1hard/25298soft), initialized planning entity (Inspection of 23226.I#0 1418884381).
DEBUG: [2012-09-07 14:56:30,281]     Step index (265), time spend (8403), score (-1hard/25373soft), initialized planning entity (Inspection of 23226.I#1 1418884382).

>>> hardscore keeps dropping

DEBUG: [2012-09-07 14:56:31,672]     Step index (321), time spend (9795), score (-15hard/28951soft), initialized planning entity (Inspection of 8748.I#1 1923067852).
DEBUG: [2012-09-07 14:56:31,703]     Step index (322), time spend (9826), score (-15hard/29026soft), initialized planning entity (Inspection of 124206.I#0 2065209261).
DEBUG: [2012-09-07 14:56:31,719]     Step index (323), time spend (9842), score (-15hard/29101soft), initialized planning entity (Inspection of 124206.I#1 2065209262).
INFO : [2012-09-07 14:56:31,719] Phase constructionHeuristic ended: step total (324), time spend (9842), best score (-15hard/29101soft).

>>> then I get this warning:

WARN : [2012-09-07 14:56:40,252]     Cancelled step index (0), time spend (18375): there is no doable move. Terminating phase early.
This should be fixed for 5.5.0.Beta1 (hopefully released next week).
IIRC, It only happened on small datasets, usually when they are close to optimality. Nevertheless, it's a bug (now fixed).

INFO : [2012-09-07 14:56:40,252] Phase localSearch ended: step total (0), time spend (18375), best score (-15hard/29101soft).
INFO : [2012-09-07 14:56:40,252] Solving ended: time spend (18375), best score (-15hard/29101soft), average calculate count per second (20824).

I know I give little details b ut I don't know what is relevant.
If someone can point me into some direction I can search and provide more info.

Thanks a lot!

Michiel

 
-----------------
http://www.codessentials.com - Your essential software, for free!
Follow us at http://twitter.com/#!/Codessentials


_______________________________________________
rules-users mailing list
rules-users@lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users






_______________________________________________
rules-users mailing list
rules-users@lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users