From marbles-isi@mailman.isi.edu Fri Oct 5 18:53:33 2001 From: marbles-isi@mailman.isi.edu (Martin Frank) Date: Fri, 05 Oct 2001 10:53:33 -0700 Subject: [Marbles-isi] took randomtime out of cio compile order in solvers/centralized Makefile for now Message-ID: <5.1.0.14.2.20011005105143.00a92e68@tnt.isi.edu> Here are the errors. Cheers, yours, Martin C:/freeware/jdk1.3.1_01/bin/javac -classpath 'C:/code/source' *.java M2RtRandomTime.java:136: cannot resolve symbol symbol : method prettyPrint (java.io.PrintWriter) location: class edu.isi.dce.marbles2.solvers.centralized.randomtime.M2RtTaskSchedules tasksSchedules.prettyPrint(Util.err); ^ M2RtRandomTime.java:142: cannot resolve symbol symbol : constructor M2RtResourceSchedules (edu.isi.dce.marbles2.problemdefinition.M2PdResources) location: class edu.isi.dce.marbles2.solvers.centralized.randomtime.M2RtResourceSchedules iResSchedules = new M2RtResourceSchedules(iProblem.getResources()); ^ M2RtRandomTime.java:152: cannot resolve symbol symbol : method computeTotalEnergy (edu.isi.dce.marbles2.problemdefinition.M2PdResources) location: class edu.isi.dce.marbles2.solvers.centralized.randomtime.M2RtResourceActivityPairs long energy = resActPairs.computeTotalEnergy(iProblem.getResources()); ^ M2RtRandomTime.java:205: cannot resolve symbol symbol : method getSchedule (java.lang.String) location: class edu.isi.dce.marbles2.solvers.centralized.randomtime.M2RtResourceSchedules M2RtResourceSchedule rs = iResSchedules.getSchedule(res.getName()); ^ M2RtRandomTime.java:224: cannot resolve symbol symbol : method listResources () location: class edu.isi.dce.marbles2.problemdefinition.M2PdRequirementResourcePairs if(!reqResPairs.listResources().contains(res) && okToAppend){ ^ M2RtRandomTime.java:271: cannot resolve symbol symbol : method getSchedule (java.lang.String) location: class edu.isi.dce.marbles2.solvers.centralized.randomtime.M2RtResourceSchedules M2RtResourceSchedule resSchedule = iResSchedules.getSchedule(rrp.getResource().getName()); ^ M2RtRandomTime.java:287: cannot resolve symbol symbol : method prettyPrint (java.io.PrintWriter) location: class edu.isi.dce.marbles2.solvers.centralized.randomtime.M2RtTaskInterval if(debug){tInterval.prettyPrint(Util.err);} ^ M2RtRequirementResourcePair.java:19: edu.isi.dce.marbles2.solvers.centralized.randomtime.M2RtRequirementResourcePair should be declared abstract; it does not define prettyPrint(java.io.PrintWriter) in edu.isi.dce.marbles2.solvers.centralized.randomtime.M2RtRequirementResourcePair public class M2RtRequirementResourcePair implements M2PdRequirementResourcePair { ^ From marbles-isi@mailman.isi.edu Fri Oct 5 19:02:18 2001 From: marbles-isi@mailman.isi.edu (Alejandro Bugacov) Date: Fri, 05 Oct 2001 11:02:18 -0700 Subject: [Marbles-isi] took randomtime out of cio compile order insolvers/centralized Makefile for now References: <5.1.0.14.2.20011005105143.00a92e68@tnt.isi.edu> Message-ID: <3BBDF5AA.95027761@isi.edu> oops! i forgot to add a few files! sorry. alejandro. Martin Frank wrote: > > Here are the errors. > > Cheers, yours, Martin > > C:/freeware/jdk1.3.1_01/bin/javac -classpath 'C:/code/source' *.java > M2RtRandomTime.java:136: cannot resolve symbol > symbol : method prettyPrint (java.io.PrintWriter) > location: class > edu.isi.dce.marbles2.solvers.centralized.randomtime.M2RtTaskSchedules > tasksSchedules.prettyPrint(Util.err); > ^ > M2RtRandomTime.java:142: cannot resolve symbol > symbol : constructor > M2RtResourceSchedules (edu.isi.dce.marbles2.problemdefinition.M2PdResources) > location: class > edu.isi.dce.marbles2.solvers.centralized.randomtime.M2RtResourceSchedules > iResSchedules = new M2RtResourceSchedules(iProblem.getResources()); > ^ > M2RtRandomTime.java:152: cannot resolve symbol > symbol : method > computeTotalEnergy (edu.isi.dce.marbles2.problemdefinition.M2PdResources) > location: class > edu.isi.dce.marbles2.solvers.centralized.randomtime.M2RtResourceActivityPairs > long energy = resActPairs.computeTotalEnergy(iProblem.getResources()); > ^ > M2RtRandomTime.java:205: cannot resolve symbol > symbol : method getSchedule (java.lang.String) > location: class > edu.isi.dce.marbles2.solvers.centralized.randomtime.M2RtResourceSchedules > M2RtResourceSchedule rs = > iResSchedules.getSchedule(res.getName()); > ^ > M2RtRandomTime.java:224: cannot resolve symbol > symbol : method listResources () > location: class > edu.isi.dce.marbles2.problemdefinition.M2PdRequirementResourcePairs > if(!reqResPairs.listResources().contains(res) && okToAppend){ > ^ > M2RtRandomTime.java:271: cannot resolve symbol > symbol : method getSchedule (java.lang.String) > location: class > edu.isi.dce.marbles2.solvers.centralized.randomtime.M2RtResourceSchedules > M2RtResourceSchedule resSchedule = > iResSchedules.getSchedule(rrp.getResource().getName()); > ^ > M2RtRandomTime.java:287: cannot resolve symbol > symbol : method prettyPrint (java.io.PrintWriter) > location: class > edu.isi.dce.marbles2.solvers.centralized.randomtime.M2RtTaskInterval > if(debug){tInterval.prettyPrint(Util.err);} > ^ > M2RtRequirementResourcePair.java:19: > edu.isi.dce.marbles2.solvers.centralized.randomtime.M2RtRequirementResourcePair > should be declared abstract; it does not define > prettyPrint(java.io.PrintWriter) in > edu.isi.dce.marbles2.solvers.centralized.randomtime.M2RtRequirementResourcePair > public class M2RtRequirementResourcePair implements > M2PdRequirementResourcePair { > ^ > > _______________________________________________ > marbles-isi mailing list > marbles-isi@mailman.isi.edu > http://mailman.isi.edu/mailman/listinfo/marbles-isi -- Alejandro Bugacov USC Information Sciences Institute 4676 Admiralty Way, Suite 1001 Marina del Rey, CA 90292 Voice: (310) 448-8269 Fax: (310) 823-6714 http://www.isi.edu/~bugacov From marbles-isi@mailman.isi.edu Mon Oct 8 23:19:15 2001 From: marbles-isi@mailman.isi.edu (Seongbin Park) Date: Mon, 08 Oct 2001 15:19:15 -0700 Subject: [Marbles-isi] marbles 2 problem generation package Message-ID: <200110082219.f98MJFg09379@tnt.isi.edu> Hi, I committed the current version of marbles 2 problem generation package. You can run it by "sh run.sh" under the directory ... marbles2/problemgen. Seongbin From marbles-isi@mailman.isi.edu Mon Oct 8 23:19:15 2001 From: marbles-isi@mailman.isi.edu (Seongbin Park) Date: Mon, 08 Oct 2001 15:19:15 -0700 Subject: [Marbles-isi] marbles 2 problem generation package Message-ID: <200110082219.f98MJFg09379@tnt.isi.edu> Hi, I committed the current version of marbles 2 problem generation package. You can run it by "sh run.sh" under the directory ... marbles2/problemgen. Seongbin From marbles-isi@mailman.isi.edu Thu Oct 11 00:34:43 2001 From: marbles-isi@mailman.isi.edu (Min Cai) Date: Wed, 10 Oct 2001 16:34:43 -0700 Subject: [Marbles-isi] checked in new version of problemverifier Message-ID: <003801c151e4$2426a3f0$b7800980@bugatti> This is a multi-part message in MIME format. ------=_NextPart_000_0035_01C151A9.778C2280 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable I have checked in the new version of problemverifier with the function = to check the problemdefinition correctness. I also added the code to = check the correctness after we parse the problemdefintion from xml files in the main package. - Min ------=_NextPart_000_0035_01C151A9.778C2280 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
I have checked in the new version of=20 problemverifier with the function to check the problemdefinition = correctness. I=20 also added the code to check the correctness after we parse the=20 problemdefintion
from xml files in the main = package.
 
 
- Min
------=_NextPart_000_0035_01C151A9.778C2280-- From marbles-isi@mailman.isi.edu Thu Oct 11 23:54:35 2001 From: marbles-isi@mailman.isi.edu (Seongbin Park) Date: Thu, 11 Oct 2001 15:54:35 -0700 Subject: [Marbles-isi] marbles2/problemgen Message-ID: <200110112254.f9BMsZg12640@tnt.isi.edu> Hi, I made some changes and commited the problemgen codes -- please do cvs updat -d at the directory .../marbles2/problemgen. Seongbin From marbles-isi@mailman.isi.edu Sat Oct 13 01:06:50 2001 From: marbles-isi@mailman.isi.edu (Min Cai) Date: Fri, 12 Oct 2001 17:06:50 -0700 Subject: [Marbles-isi] checked in:solutionverifier Message-ID: <005801c1537a$f57fe640$b7800980@bugatti> This is a multi-part message in MIME format. ------=_NextPart_000_0055_01C15340.4903E950 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable I checked in the solutionverifier and modified the runner to verifiy the = solution using this package. - Min ------=_NextPart_000_0055_01C15340.4903E950 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
I checked in the solutionverifier and = modified the=20 runner to verifiy the solution using this package.
 
- Min
------=_NextPart_000_0055_01C15340.4903E950-- From marbles-isi@mailman.isi.edu Mon Oct 15 17:54:26 2001 From: marbles-isi@mailman.isi.edu (Martin Frank) Date: Mon, 15 Oct 2001 09:54:26 -0700 Subject: [Marbles-isi] randomtime solver returns invalid solutions Message-ID: <5.1.0.14.2.20011015094508.00ae95c8@tnt.isi.edu> --=====================_3874200==_ Content-Type: text/plain; charset="us-ascii"; format=flowed All - I hooked in the new solutionverifier that Min wrote to check on the results of the solvers. It turns out that the randomtime solver can return invalid solutions. A simple 1-task 2-requirements problem and its randomtime-proposed solution is attached. Note that resource B is not available at the time that randomtime scheduled the task. Alejandro and Donghan, can you look into this. Probably a trivial bug. Thanks, yours, Martin P.S.: You can reproduce the problem by checking out from CVS and then saying "sh gentest.sh" in marbles2/main. --=====================_3874200==_ Content-Type: application/xml; name="current.m2problem.gen.xml" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="current.m2problem.gen.xml" PFByb2JsZW1EZWZpbml0aW9uPg0KICA8Tmlja25hbWU+TWFyYmxlcyBTY2hlbWE8L05pY2tuYW1l Pg0KICA8RGVzY3JpcHRpb24+TWFyYmxlcyBkZXNjcmlwdGlvbjwvRGVzY3JpcHRpb24+DQogIDxT cFJlc291cmNlcz4NCiAgICA8UmVzb3VyY2U+DQogICAgICA8TmFtZT5BPC9OYW1lPg0KICAgICAg PEF2YWlsYWJpbGl0eT4NCiAgICAgICAgPFg+DQogICAgICAgICAgPEludGVydmFsPg0KICAgICAg ICAgICAgPEZyb20+MjAwMS0wOC0yOFQxODowOToxMS0wNzowMDwvRnJvbT4NCiAgICAgICAgICAg IDxUbz4yMDAxLTA4LTI5VDA3OjQ4OjI5LTA3OjAwPC9Ubz4NCiAgICAgICAgICA8L0ludGVydmFs Pg0KICAgICAgICAgIDxJbnRlcnZhbD4NCiAgICAgICAgICAgIDxGcm9tPjIwMDEtMDgtMjlUMDk6 NTc6MzgtMDc6MDA8L0Zyb20+DQogICAgICAgICAgICA8VG8+MjAwMS0wOC0yOVQxNzo0MjoyMC0w NzowMDwvVG8+DQogICAgICAgICAgPC9JbnRlcnZhbD4NCiAgICAgICAgICA8SW50ZXJ2YWw+DQog ICAgICAgICAgICA8RnJvbT4yMDAxLTA4LTI5VDIyOjU2OjQ5LTA3OjAwPC9Gcm9tPg0KICAgICAg ICAgICAgPFRvPjIwMDEtMDgtMzBUMDY6Mjk6MzktMDc6MDA8L1RvPg0KICAgICAgICAgIDwvSW50 ZXJ2YWw+DQogICAgICAgICAgPEludGVydmFsPg0KICAgICAgICAgICAgPEZyb20+MjAwMS0wOC0z MFQxNjo0NDo0OS0wNzowMDwvRnJvbT4NCiAgICAgICAgICAgIDxUbz4yMDAxLTA4LTMxVDAxOjQx OjE1LTA3OjAwPC9Ubz4NCiAgICAgICAgICA8L0ludGVydmFsPg0KICAgICAgICAgIDxJbnRlcnZh bD4NCiAgICAgICAgICAgIDxGcm9tPjIwMDEtMDgtMzFUMDk6NTA6MTktMDc6MDA8L0Zyb20+DQog ICAgICAgICAgICA8VG8+MjAwMS0wOC0zMVQxNTozMjo0Ni0wNzowMDwvVG8+DQogICAgICAgICAg PC9JbnRlcnZhbD4NCiAgICAgICAgPC9YPg0KICAgICAgPC9BdmFpbGFiaWxpdHk+DQogICAgPC9S ZXNvdXJjZT4NCiAgICA8UmVzb3VyY2U+DQogICAgICA8TmFtZT5CPC9OYW1lPg0KICAgICAgPEF2 YWlsYWJpbGl0eT4NCiAgICAgICAgPFg+DQogICAgICAgICAgPEludGVydmFsPg0KICAgICAgICAg ICAgPEZyb20+MjAwMS0wOC0yOFQyMzo0MToxNy0wNzowMDwvRnJvbT4NCiAgICAgICAgICAgIDxU bz4yMDAxLTA4LTI5VDA5OjI5OjA1LTA3OjAwPC9Ubz4NCiAgICAgICAgICA8L0ludGVydmFsPg0K ICAgICAgICAgIDxJbnRlcnZhbD4NCiAgICAgICAgICAgIDxGcm9tPjIwMDEtMDgtMjlUMTI6NDM6 NDYtMDc6MDA8L0Zyb20+DQogICAgICAgICAgICA8VG8+MjAwMS0wOC0yOVQyMzoyNzo1NC0wNzow MDwvVG8+DQogICAgICAgICAgPC9JbnRlcnZhbD4NCiAgICAgICAgICA8SW50ZXJ2YWw+DQogICAg ICAgICAgICA8RnJvbT4yMDAxLTA4LTMwVDEwOjI4OjI0LTA3OjAwPC9Gcm9tPg0KICAgICAgICAg ICAgPFRvPjIwMDEtMDgtMzBUMTA6NDY6MTgtMDc6MDA8L1RvPg0KICAgICAgICAgIDwvSW50ZXJ2 YWw+DQogICAgICAgICAgPEludGVydmFsPg0KICAgICAgICAgICAgPEZyb20+MjAwMS0wOC0zMFQy Mzo0NToxNi0wNzowMDwvRnJvbT4NCiAgICAgICAgICAgIDxUbz4yMDAxLTA4LTMxVDE2OjI0OjIx LTA3OjAwPC9Ubz4NCiAgICAgICAgICA8L0ludGVydmFsPg0KICAgICAgICA8L1g+DQogICAgICA8 L0F2YWlsYWJpbGl0eT4NCiAgICA8L1Jlc291cmNlPg0KICA8L1NwUmVzb3VyY2VzPg0KICA8U3BU YXNrcz4NCiAgICA8VGFzaz4NCiAgICAgIDxOYW1lPlE8L05hbWU+DQogICAgICA8VmFsdWU+MjAw PC9WYWx1ZT4NCiAgICAgIDxTcFNlZ21lbnRzPg0KICAgICAgICA8U2VnbWVudD4NCiAgICAgICAg ICA8TmFtZT5zZWcwPC9OYW1lPg0KICAgICAgICAgIDxUeXBpY2FsRHVyYXRpb24+DQogICAgICAg ICAgICA8RHVyYXRpb25Db21wb25lbnRzPg0KICAgICAgICAgICAgICA8RHVyYXRpb25Db21wb25l bnQ+DQogICAgICAgICAgICAgICAgPENvdW50PjQ5PC9Db3VudD4NCiAgICAgICAgICAgICAgICA8 VW5pdD5taW51dGU8L1VuaXQ+DQogICAgICAgICAgICAgIDwvRHVyYXRpb25Db21wb25lbnQ+DQog ICAgICAgICAgICA8L0R1cmF0aW9uQ29tcG9uZW50cz4NCiAgICAgICAgICA8L1R5cGljYWxEdXJh dGlvbj4NCiAgICAgICAgPC9TZWdtZW50Pg0KICAgICAgICA8U2VnbWVudD4NCiAgICAgICAgICA8 TmFtZT5zZWcxPC9OYW1lPg0KICAgICAgICAgIDxUeXBpY2FsRHVyYXRpb24+DQogICAgICAgICAg ICA8RHVyYXRpb25Db21wb25lbnRzPg0KICAgICAgICAgICAgICA8RHVyYXRpb25Db21wb25lbnQ+ DQogICAgICAgICAgICAgICAgPENvdW50PjQwPC9Db3VudD4NCiAgICAgICAgICAgICAgICA8VW5p dD5taW51dGU8L1VuaXQ+DQogICAgICAgICAgICAgIDwvRHVyYXRpb25Db21wb25lbnQ+DQogICAg ICAgICAgICA8L0R1cmF0aW9uQ29tcG9uZW50cz4NCiAgICAgICAgICA8L1R5cGljYWxEdXJhdGlv bj4NCiAgICAgICAgPC9TZWdtZW50Pg0KICAgICAgPC9TcFNlZ21lbnRzPg0KICAgICAgPEFuY2hv clNlZ21lbnROYW1lPnNlZzE8L0FuY2hvclNlZ21lbnROYW1lPg0KICAgICAgPFZhbGlkQmVnaW5u aW5nc09mQW5jaG9yU2VnbWVudD4NCiAgICAgICAgPFg+DQogICAgICAgICAgPEludGVydmFsPg0K ICAgICAgICAgICAgPEZyb20+MjAwMS0wOC0yOFQwMjo1Njo0NC0wNzowMDwvRnJvbT4NCiAgICAg ICAgICAgIDxUbz4yMDAxLTA4LTI4VDIyOjQ3OjQzLTA3OjAwPC9Ubz4NCiAgICAgICAgICA8L0lu dGVydmFsPg0KICAgICAgICA8L1g+DQogICAgICA8L1ZhbGlkQmVnaW5uaW5nc09mQW5jaG9yU2Vn bWVudD4NCiAgICAgIDxTcFJlcXVpcmVtZW50cz4NCiAgICAgICAgPFJlcXVpcmVtZW50Pg0KICAg ICAgICAgIDxOYW1lPjE8L05hbWU+DQogICAgICAgICAgPFJlcXVpcmVkRm9yU2VnbWVudE5hbWVz Pg0KICAgICAgICAgICAgPFN0cmluZz5zZWcwPC9TdHJpbmc+DQogICAgICAgICAgICA8U3RyaW5n PnNlZzE8L1N0cmluZz4NCiAgICAgICAgICA8L1JlcXVpcmVkRm9yU2VnbWVudE5hbWVzPg0KICAg ICAgICA8L1JlcXVpcmVtZW50Pg0KICAgICAgICA8UmVxdWlyZW1lbnQ+DQogICAgICAgICAgPE5h bWU+MjwvTmFtZT4NCiAgICAgICAgICA8UmVxdWlyZWRGb3JTZWdtZW50TmFtZXM+DQogICAgICAg ICAgICA8U3RyaW5nPnNlZzE8L1N0cmluZz4NCiAgICAgICAgICA8L1JlcXVpcmVkRm9yU2VnbWVu dE5hbWVzPg0KICAgICAgICA8L1JlcXVpcmVtZW50Pg0KICAgICAgPC9TcFJlcXVpcmVtZW50cz4N CiAgICAgIDxQb3NzaWJsZUZpbGxlcnM+DQogICAgICAgIDxQb3NzaWJsZUZpbGxlcj4NCiAgICAg ICAgICA8UmVxdWlyZW1lbnRzQWZmZWN0ZWQ+DQogICAgICAgICAgICA8U3RyaW5nPjE8L1N0cmlu Zz4NCiAgICAgICAgICAgIDxTdHJpbmc+MjwvU3RyaW5nPg0KICAgICAgICAgIDwvUmVxdWlyZW1l bnRzQWZmZWN0ZWQ+DQogICAgICAgICAgPFBvc3NpYmxlUmVzb3VyY2VDb21iaW5hdGlvbnM+DQog ICAgICAgICAgICA8UG9zc2libGVSZXNvdXJjZUNvbWJpbmF0aW9uPg0KICAgICAgICAgICAgICA8 UmVzb3VyY2VzPg0KICAgICAgICAgICAgICAgIDxTdHJpbmc+QjwvU3RyaW5nPg0KICAgICAgICAg ICAgICAgIDxTdHJpbmc+QTwvU3RyaW5nPg0KICAgICAgICAgICAgICA8L1Jlc291cmNlcz4NCiAg ICAgICAgICAgIDwvUG9zc2libGVSZXNvdXJjZUNvbWJpbmF0aW9uPg0KICAgICAgICAgIDwvUG9z c2libGVSZXNvdXJjZUNvbWJpbmF0aW9ucz4NCiAgICAgICAgPC9Qb3NzaWJsZUZpbGxlcj4NCiAg ICAgIDwvUG9zc2libGVGaWxsZXJzPg0KICAgIDwvVGFzaz4NCiAgPC9TcFRhc2tzPg0KPC9Qcm9i bGVtRGVmaW5pdGlvbj4NCg== --=====================_3874200==_ Content-Type: application/xml; name="current.m2solution.gen.xml" Content-Transfer-Encoding: base64 Content-Disposition: attachment; filename="current.m2solution.gen.xml" PFNvbHV0aW9uRGVmaW5pdGlvbj4NCiAgPEZpbGxlZFRhc2tzPg0KICAgIDxGaWxsZWRUYXNrPg0K ICAgICAgPE5hbWU+UTwvTmFtZT4NCiAgICAgIDxUaW1lT2ZUYXNrPjIwMDEtMDgtMjhUMTg6Mzc6 NTgtMDc6MDA8L1RpbWVPZlRhc2s+DQogICAgICA8UmVzb3VyY2VOYW1lcz4NCiAgICAgICAgPFN0 cmluZz5CPC9TdHJpbmc+DQogICAgICAgIDxTdHJpbmc+QTwvU3RyaW5nPg0KICAgICAgPC9SZXNv dXJjZU5hbWVzPg0KICAgIDwvRmlsbGVkVGFzaz4NCiAgPC9GaWxsZWRUYXNrcz4NCjwvU29sdXRp b25EZWZpbml0aW9uPg0K --=====================_3874200==_-- From marbles-isi@mailman.isi.edu Mon Oct 15 18:04:06 2001 From: marbles-isi@mailman.isi.edu (Alejandro Bugacov) Date: Mon, 15 Oct 2001 10:04:06 -0700 Subject: [Marbles-isi] randomtime solver returns invalid solutions References: <5.1.0.14.2.20011015094508.00ae95c8@tnt.isi.edu> Message-ID: <3BCB1706.1537D035@isi.edu> yes, we are aware of that! at this randomtime (not finnished yet) solutions can be incorrect. we haven't writen the part to filter the correct ones yet. i'd comment out the calling of the randomtime solver. alejandro. Martin Frank wrote: > > All - > > I hooked in the new solutionverifier that Min wrote to check on the results > of the solvers. It turns out that the randomtime solver can return invalid > solutions. > > A simple 1-task 2-requirements problem and its randomtime-proposed solution > is attached. > > Note that resource B is not available at the time that randomtime scheduled > the task. > > Alejandro and Donghan, can you look into this. Probably a trivial bug. > > Thanks, yours, Martin > > P.S.: You can reproduce the problem by checking out from CVS and then > saying "sh gentest.sh" in marbles2/main. > > ------------------------------------------------------------------------ > Name: current.m2problem.gen.xml > current.m2problem.gen.xml Type: XML Document (text/xml) > Encoding: base64 > > Name: current.m2solution.gen.xml > current.m2solution.gen.xml Type: XML Document (text/xml) > Encoding: base64 -- Alejandro Bugacov bugacov@isi.edu Voice: (310) 448-8269 Fax: (310) 823-6714 http://www.isi.edu/~bugacov From marbles-isi@mailman.isi.edu Thu Oct 18 02:12:42 2001 From: marbles-isi@mailman.isi.edu (Martin Frank) Date: Wed, 17 Oct 2001 18:12:42 -0700 Subject: [Marbles-isi] proposed enhancements/changes to the Marbles 2 problem definition Message-ID: <5.1.0.14.2.20011017180758.00ab27f8@tnt.isi.edu> Proposed enhancement are checked in, commented out for now; Min will implement for SimpleProblem. These allow a solver to find out from the problem definition which requirements are interdependent with which other requirements, then ask the problem definition for the estimated total number of possible combinations, and ask for individual combinations. This enables the solver author to write a more intelligent solver, having to rely less on a blind "generate-and-test" strategy. The satisfyingResourcesFor() is obviously not going away until (a) no solver uses it anymore. Does this make sense to you? Cheers, yours, Martin // call below is obsolete - new solvers should not use it public ResourceBonusPairs satisfyingResourcesFor( Requirement, Task taskContainingTheRequirement, RequirementResourcePairs alreadyAssignedResourcesOfThisTask, Interval /*during this entire time interval*/) /* Give me all resources that are qualified and available to fill this requirement during the entire given time. (Note that even though a resource may not be scheduled for anything it may still not be available for a task, because of crew day restrictions for example.) The returned bonus is the _minimum_ bonus (or malus) for using the resource -- it can turn out to be higher than that based on the exact time point of usage. (dynamic) */; // the satisfyingResourcesFor cal below is replaced by these three calls // Min: comment in to implement // public DependencyGroup dependencyGroupFor(Task,Requirement,Interval) // /* For example, if you call this on Pilot1 in SNAP, it will return // (Pilot1,Pilot2). If you call on the Range, it will return (Range). Will // never be NULL, will never be empty. */; // public int estimatedNumberOfCombinationsFor(DependencyGroup); // public ResourceBonusPairs returnCombinationNumberFor( // int /*between 1 and estimated number of combinations*/, // DependencyGroup) // /* The return value has the same number of list item as the requirements in // the DependencyGroup, and fills them in order. If there is an estimated // 10000 number of combinations, the problem API implementor should attempt to // return the highest-sum-bonus resources for the lower-numbered combinations, // but without making the search very expensive. Thus, to find the optimal // solution, the solver has to ask for every combination number after all, but // searching the lower numbers should generally be more fruitful.*/; From marbles-isi@mailman.isi.edu Thu Oct 18 07:09:30 2001 From: marbles-isi@mailman.isi.edu (Seongbin Park) Date: Wed, 17 Oct 2001 23:09:30 -0700 Subject: [Marbles-isi] marbles 2 problemgen Message-ID: <200110180609.f9I69Ug13064@tnt.isi.edu> Hi, While I was trying to run a marbles 2 solver with the data synthetically generated from the problem generation package, I found that there was a bug about XML structures (PossibleFillers). It was fixed and I just now commited it to the marbles2/problemgen directory. Sorry about the bug and please do cvs update -d under the directory .../marbles2/problemgen. Seongbin From marbles-isi@mailman.isi.edu Thu Oct 18 07:23:22 2001 From: marbles-isi@mailman.isi.edu (Seongbin Park) Date: Wed, 17 Oct 2001 23:23:22 -0700 Subject: [Marbles-isi] proposed enhancements/changes to the Marbles 2 problem definition In-Reply-To: Your message of "Wed, 17 Oct 2001 18:12:42 PDT." <5.1.0.14.2.20011017180758.00ab27f8@tnt.isi.edu> Message-ID: <200110180623.f9I6NMg15487@tnt.isi.edu> Hi Martin, > Proposed enhancement are checked in, commented out for now; Min will > implement for SimpleProblem. > > These allow a solver to find out from the problem definition which > requirements are interdependent with which other requirements, then ask the > problem definition for the estimated total number of possible combinations, > and ask for individual combinations. > > This enables the solver author to write a more intelligent solver, having > to rely less on a blind "generate-and-test" strategy. > > The satisfyingResourcesFor() is obviously not going away until (a) no > solver uses it anymore. > > Does this make sense to you? I believe this question is for solver writers, and I have not written a solver, so cannot tell you much. However, from the description above, my feeling about the proposed enhancement is that it will certainly make a solver more intelligent since it allows the solver to preprocess with the problem definition first and proceed to find real combinations with time availability information later. Or, if you think my understanding does not appear to be what you were intending to say, please let me know. Thanks, Seongbin From marbles-isi@mailman.isi.edu Wed Oct 31 04:29:03 2001 From: marbles-isi@mailman.isi.edu (Seongbin Park) Date: Tue, 30 Oct 2001 20:29:03 -0800 Subject: [Marbles-isi] marbles 2 problemgen Message-ID: <200110310429.f9V4T3g04465@tnt.isi.edu> Hi, I just now checked in the modified codes for marbles 2 problem generation. Please do cvs update -d under the directory .../marbles2/problemgen. Seongbin