[rules-users] Re: drools-solver XML schema/detailed description?

Geoffrey De Smet ge0ffrey.spam at gmail.com
Tue Feb 10 14:16:07 EST 2009


With kind regards,
Geoffrey De Smet


Andrew Waterman schreef:
> Great thanks to you Geoffrey for the information.  This will really help 
> in our implementation, as the game uses several different token types, 
> so it would be nice to be able to build factories for each type and let 
> your solver handle the mix.
> 
> Could you let me know which class implements the configuration for your 
> XML?  That would help me to find out what else may be available.
> 

It's XStream based at the moment.
The xml config classes are under the package org.drools.solver.config
They follow the builder pattern to build a solver.

> best wishes,
> 
> Andrew
> 
> ---------------------------------
> Andrew Waterman
> San Cristóbal de las Casas, Chiapas, Mexico
> +52 1 967 107 5902
> +1 510 342 5693
> 
> On Feb 10, 2009, at 1:16 AM, Geoffrey De Smet wrote:
> 
>> I need to document those in the manual, if I haven't already in 
>> subversion.
>>
>> relativeSelection means that only a percentage of all moves will be 
>> randomly selected to be evaluated into a score (and possibly accepted 
>> and picked as the next step).
>> You should experiment with different values for it.
>> You can use the benchmarker class (see the examples) to play out 
>> different values against each other.
>>
>> topSize means that the top x scoring moves from the last step will 
>> always be selected. My experiments haven't proven topSize to be 
>> usefull yet...
>>
>> With kind regards,
>> Geoffrey De Smet
>>
>> Andrew Waterman schreef:
>>> Hi,
>>> My apologies for my earlier double post, I'm having some troubles 
>>> with my mail client.  I'm writing to see if anyone has any detailed 
>>> information on an XML-schema, or, a more detailed break-down of the 
>>> drools-solver configuration files.  Basically, I'm trying to find out 
>>> the set of available tags in drools-solver without going through the 
>>> XML parsing code.  In particular, I'm curious about the options with 
>>> regards to the "<selector>" tag.  In the 
>>> "curriculumCourseSolverConfig.xml" file for the curriculum 
>>> drools-solver example, the sub-tag "topSize' is used in the selector 
>>> configuration:
>>>    <selector>
>>>        <selector>
>>>            <topSize>10</topSize>
>>>        </selector>
>>>        <selector>
>>>        <selector>
>>>            <moveFactoryClass>
>>> org.drools.solver.examples.itc2007.curriculumcourse.solver.move.factory.RoomChangeMoveFactory 
>>>
>>>    </moveFactoryClass>
>>>            <relativeSelection>0.002</relativeSelection>
>>>        </selector>
>>> .... [list of further move factories]     </selector>
>>> In addition, the "relativeSelection" subtag is used as well.
>>> What do these subtags mean?  Is there a way to let the selector move 
>>> through different weighted MoveFactories? Thanks!
>>> best wishes,
>>> Andrew
>>> ---------------------------------
>>> Andrew Waterman
>>> San Cristóbal de las Casas, Chiapas, Mexico
>>> +52 1 967 107 5902
>>> +1 510 342 5693
>>> ------------------------------------------------------------------------
>>> _______________________________________________
>>> rules-users mailing list
>>> rules-users at lists.jboss.org
>>> https://lists.jboss.org/mailman/listinfo/rules-users
>>
>> _______________________________________________
>> rules-users mailing list
>> rules-users at lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/rules-users
> 
> 
> _______________________________________________
> rules-users mailing list
> rules-users at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/rules-users
> 




More information about the rules-users mailing list