repast-interest

[Repast-interest] colt web page

[Nick C. <nic...@ve...>]

Hi,

With the recent emails about the colt library I thought I'd post a
reference to the colt homepage:

http://tilde-hoschek.home.cern.ch/~hoschek/colt/

Nick

-- 
Nick Collier 
Social Science Research Computing 
University of Chicago 
http://repast.sourceforge.net

Re: [Repast-interest] Mersenne NormalDist wierdness.

[Nick C. <nic...@ve...>]

Hi,

I just ran your test case on my machine and I can't reproduce the
problem. At over 1.5 million iterations, it seems to work fine. 

The specifics:

[nick@trinco nick]$java -jar /home/nick/repast/lib/colt.jar

Version 1.0.2.133 (Thu Nov  8 15:45:10 CET 2001)
Includes patch for Normal.
Please report problems to wol...@ce... or
http://nicewww.cern.ch/~hoschek/colt/index.htm

[nick@trinco nick]$ java -cp /home/nick/repast/lib/repast.jar

uchicago.src.sim.engine.SimInit -v
RePast Version: 1.4.1p1


What do you get for the colt and repast versions? What machine and OS
are you running?

thanks,

Nick



On Tue, 2002-09-10 at 19:34, wbh wrote:
> Hi. I'm trying to generate a normally distributed value, but was getting
> nowhere. I've narrowed it down to:
> 
> import java.util.Date;
> import cern.jet.random.*;
> import cern.jet.random.engine.MersenneTwister;
> import edu.cornell.lassp.houle.RngPack.RandomElement;
> 
> public class tr
>   {
>   public static void
>   main (String[] args)
>     {
>     RandomElement   generator;
>     Normal          normaldist;
> 
>     generator=new MersenneTwister((int)(new Date()).getTime());
>     normaldist=new Normal(10.0f,5.0f,generator);
>     for(;;) System.out.println(normaldist.nextDouble());
>     }
>   }
> 
> Seems to go OK, but after a while (number 1750-1780) gets stuck and only
> ever gives out 10.0 again. Any ideas what I'm doing wrong? Thanks!
> Bill
> 
> 
> 
> -------------------------------------------------------
> In remembrance
> www.osdn.com/911/
> _______________________________________________
> Repast-interest mailing list
> Rep...@li...
> https://lists.sourceforge.net/lists/listinfo/repast-interest
-- 
Nick Collier 
Social Science Research Computing 
University of Chicago 
http://repast.sourceforge.net

Re: [Repast-interest] Mersenne NormalDist wierdness.

[Nick C. <nic...@ve...>]

For RNG RePast uses the colt library which is produced by CERN in
Switzerland. Its generally recommended as a good Java open source
numerics library. The version of colt distributed with repast 1.4
includes a patch to colt for the bug that Rick found last year. I'll
look into this current bug report and contact the colt people about it.

Nick

On Tue, 2002-09-10 at 20:00, Rick Riolo wrote:
> 
> I don't have an answer, but its scary to see
> (given i found a problem in another rng method last winter).
> 
> So...has anyone done an serious testing of the RNG 
> methods that RePast is using?
> Is there a test suite that has been used to exercise any of them?
> 
> If not, I think it should be moved up someone's
> ToDo list...perhaps even a group effort, if need be.
> 
> I think its *really* important to be able to
> have a lot of confidence in the rng methods being used
> for these kinds CAS of models.
> 
> - r
> 
> -- 
> Rick Riolo                           rl...@um...
> Center for Study of Complex Systems (CSCS)
> 4477 Randall Lab                
> University of Michigan         Ann Arbor MI 48109-1120
> Phone: 734 763 3323                  Fax: 734 763 9267
> http://www.pscs.umich.edu/PEOPLE/rlr-home.html
> 
> On Wed, 11 Sep 2002, wbh wrote:
> 
> > Date: Wed, 11 Sep 2002 00:34:34 +0100 (BST)
> > From: wbh <W.B...@ue...>
> > To: rep...@li...
> > Subject: [Repast-interest] Mersenne NormalDist wierdness.
> > 
> > Hi. I'm trying to generate a normally distributed value, but was getting
> > nowhere. I've narrowed it down to:
> > 
> > import java.util.Date;
> > import cern.jet.random.*;
> > import cern.jet.random.engine.MersenneTwister;
> > import edu.cornell.lassp.houle.RngPack.RandomElement;
> > 
> > public class tr
> >   {
> >   public static void
> >   main (String[] args)
> >     {
> >     RandomElement   generator;
> >     Normal          normaldist;
> > 
> >     generator=new MersenneTwister((int)(new Date()).getTime());
> >     normaldist=new Normal(10.0f,5.0f,generator);
> >     for(;;) System.out.println(normaldist.nextDouble());
> >     }
> >   }
> > 
> > Seems to go OK, but after a while (number 1750-1780) gets stuck and only
> > ever gives out 10.0 again. Any ideas what I'm doing wrong? Thanks!
> > Bill
> > 
> > 
> > 
> > -------------------------------------------------------
> > In remembrance
> > www.osdn.com/911/
> > _______________________________________________
> > Repast-interest mailing list
> > Rep...@li...
> > https://lists.sourceforge.net/lists/listinfo/repast-interest
> > 
> 
> 
> 
> -------------------------------------------------------
> In remembrance
> www.osdn.com/911/
> _______________________________________________
> Repast-interest mailing list
> Rep...@li...
> https://lists.sourceforge.net/lists/listinfo/repast-interest
-- 
Nick Collier 
Social Science Research Computing 
University of Chicago 
http://repast.sourceforge.net

Re: [Repast-interest] Mersenne NormalDist wierdness.

[Rick R. <rl...@fi...>]

I don't have an answer, but its scary to see
(given i found a problem in another rng method last winter).

So...has anyone done an serious testing of the RNG 
methods that RePast is using?
Is there a test suite that has been used to exercise any of them?

If not, I think it should be moved up someone's
ToDo list...perhaps even a group effort, if need be.

I think its *really* important to be able to
have a lot of confidence in the rng methods being used
for these kinds CAS of models.

- r

-- 
Rick Riolo                           rl...@um...
Center for Study of Complex Systems (CSCS)
4477 Randall Lab                
University of Michigan         Ann Arbor MI 48109-1120
Phone: 734 763 3323                  Fax: 734 763 9267
http://www.pscs.umich.edu/PEOPLE/rlr-home.html

On Wed, 11 Sep 2002, wbh wrote:

> Date: Wed, 11 Sep 2002 00:34:34 +0100 (BST)
> From: wbh <W.B...@ue...>
> To: rep...@li...
> Subject: [Repast-interest] Mersenne NormalDist wierdness.
> 
> Hi. I'm trying to generate a normally distributed value, but was getting
> nowhere. I've narrowed it down to:
> 
> import java.util.Date;
> import cern.jet.random.*;
> import cern.jet.random.engine.MersenneTwister;
> import edu.cornell.lassp.houle.RngPack.RandomElement;
> 
> public class tr
>   {
>   public static void
>   main (String[] args)
>     {
>     RandomElement   generator;
>     Normal          normaldist;
> 
>     generator=new MersenneTwister((int)(new Date()).getTime());
>     normaldist=new Normal(10.0f,5.0f,generator);
>     for(;;) System.out.println(normaldist.nextDouble());
>     }
>   }
> 
> Seems to go OK, but after a while (number 1750-1780) gets stuck and only
> ever gives out 10.0 again. Any ideas what I'm doing wrong? Thanks!
> Bill
> 
> 
> 
> -------------------------------------------------------
> In remembrance
> www.osdn.com/911/
> _______________________________________________
> Repast-interest mailing list
> Rep...@li...
> https://lists.sourceforge.net/lists/listinfo/repast-interest
> 

[Repast-interest] Mersenne NormalDist wierdness.

[wbh <W.B...@ue...>]

Hi. I'm trying to generate a normally distributed value, but was getting
nowhere. I've narrowed it down to:

import java.util.Date;
import cern.jet.random.*;
import cern.jet.random.engine.MersenneTwister;
import edu.cornell.lassp.houle.RngPack.RandomElement;

public class tr
  {
  public static void
  main (String[] args)
    {
    RandomElement   generator;
    Normal          normaldist;

    generator=new MersenneTwister((int)(new Date()).getTime());
    normaldist=new Normal(10.0f,5.0f,generator);
    for(;;) System.out.println(normaldist.nextDouble());
    }
  }

Seems to go OK, but after a while (number 1750-1780) gets stuck and only
ever gives out 10.0 again. Any ideas what I'm doing wrong? Thanks!
Bill

Re: [Repast-interest] drawing objects with a MultiObject2DDisplay

[Nick C. <nic...@ve...>]

Hi,

You'll have to make your own display, but this isn't too hard to do.
Your display would look something like:

public class MultiDisplay extends Display {

  private Multi2DGrid grid;

  public MultiDisplay(Multi2DGrid grid, int width, int height) {
    super(width, height);
    this.grid = grid;
  }

  public void drawDisplay(SimGraphics g) {
    if (view) {
      // iterate through the objects in your grid drawing them
      // however you want.
    }
  }
}

In drawDisplay you iterate through your grid and draw your rectangles
within the cells. You calculate the x and y pixel coordinate of the cell
by taking the row and col coordinate of your grid and multiplying that
by DisplayConstants.CELL_WIDTH and DisplayContants.CELL_HEIGHT. Those
will also give you the size of the drawing cell so you can draw your
mini rectangles within that. 

If you have any more questions, feel free to ask the list.

Nick


On Tue, 2002-09-10 at 08:54, ra...@li... wrote:
> hi,
> i have a grid with multiple objects in it, ( it's an environment with
> multiple resourses for site )
> 
> -----------------
> |a b|a  |   | ...
> |c d|  d|c  |....
> -----------------
> |a b|....
> ........
> 
> every resource into the single site has a specific production
> rate/quantityMax/currentQuantity, so the same resource name for differen
> sites have different values for the triple. 
> My Problem is this: how can i display the resources (for site there are
> at max 4 different type of resources) into the single site/cell? 
> I have set the resource class ( object of the Multi2DGrid) like Drawable
> , but in this way into the gui i can see only 1 resource. i Would be
> pobbsible to see all the resources into the cell. for es. like a rect
> made of other rects with different colors.
> 
> best regards
> 
> by
> R.J.P.B.
-- 
Nick Collier 
Social Science Research Computing 
University of Chicago 
http://repast.sourceforge.net

Re: [Repast-interest] Random compilation problems

[Nick C. <nic...@ve...>]

Hi,

If I follow your code correctly, you are calling Random.createUniform()
outside of a method body. Try calling it inside the constructor. That
should work, I think.

Nick

On Tue, 2002-09-10 at 05:03, ra...@li... wrote:
> hi,
>  i'm trying to use the random utility:
> 
>  in my .java there are
> 
>  import uchicago.src.sim.util.*;
>  import cern.jet.random.Uniform;
> 
> 
>  and into the code :
>         Random.createUniform();
> 
> the compilation give me these messages : 
> 
> EchoSpace.java:45: <identifier> expected
>         Random.createUniform();
>                            ^
> EchoSpace.java:45: cannot resolve symbol
> symbol  : class createUniform
> location: class uchicago.src.sim.util.Random
>  Random.createUniform();
>         ^
>  2 errors
> 
> by R.J.P.B.
-- 
Nick Collier 
Social Science Research Computing 
University of Chicago 
http://repast.sourceforge.net

[Repast-interest] =?iso-8859-1?Q?drawing_objects_with_a_MultiObject2DDisplay?=

[=?utf-8?Q?<ra...@li...>]
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[Repast-interest] =?iso-8859-1?Q?Random_compilation_problems?=

[=?utf-8?Q?<ra...@li...>]

aGksDQogaSdtIHRyeWluZyB0byB1c2UgdGhlIHJhbmRvbSB1dGlsaXR5Og0KDQogaW4gbXkg
LmphdmEgdGhlcmUgYXJlDQoNCiBpbXBvcnQgdWNoaWNhZ28uc3JjLnNpbS51dGlsLio7DQog
aW1wb3J0IGNlcm4uamV0LnJhbmRvbS5Vbmlmb3JtOw0KDQoNCiBhbmQgaW50byB0aGUgY29k
ZSA6DQogICAgICAgIFJhbmRvbS5jcmVhdGVVbmlmb3JtKCk7DQoNCnRoZSBjb21waWxhdGlv
biBnaXZlIG1lIHRoZXNlIG1lc3NhZ2VzIDogDQoNCkVjaG9TcGFjZS5qYXZhOjQ1OiA8aWRl
bnRpZmllcj4gZXhwZWN0ZWQNCiAgICAgICAgUmFuZG9tLmNyZWF0ZVVuaWZvcm0oKTsNCiAg
ICAgICAgICAgICAgICAgICAgICAgICAgIF4NCkVjaG9TcGFjZS5qYXZhOjQ1OiBjYW5ub3Qg
cmVzb2x2ZSBzeW1ib2wNCnN5bWJvbCAgOiBjbGFzcyBjcmVhdGVVbmlmb3JtDQpsb2NhdGlv
bjogY2xhc3MgdWNoaWNhZ28uc3JjLnNpbS51dGlsLlJhbmRvbQ0KIFJhbmRvbS5jcmVhdGVV
bmlmb3JtKCk7DQogICAgICAgIF4NCiAyIGVycm9ycw0KDQpieSBSLkouUC5CLg0K

[Repast-interest] request for repast project descriptions

[Nick C. <nic...@ve...>]

Hi all, 

I've asked this once or twice before, but the number of repast related 
projects listed on the repast project page is not representative of 
repast's actual use (I hope!!!). We had nearly 1700 downloads of 1.4
and there are quite few people on the mailing list. If even a small
portion of those are using repast, the number of projects on the project
page remains unrepresentative.

Consequently, I'd like to add to the list of projects and publications that use 
RePast on the repast web site. So, if y'all could send me a few lines 
about what you've done or are doing with repast, and your name, title, 
contact info, etc. I'd appreciate it. Also, if you have any publications
that use repast simulations, I could use an abstract as well as the
above. If you are unwilling to provide contact info that's fine as well
just a name will do.

thanks very much, 

Nick 

-- 
Nick Collier 
Social Science Research Computing 
University of Chicago 
http://repast.sourceforge.net 

[Repast-interest] Re: One more quick question...

[Nick C. <nic...@ve...>]

Hi,

This is harder than it should be, but certainly doable. The following
should turn a display off:

display.viewEventPerformed(new ViewEvent(dsurf, Displayable.TOGGLE_VIEW,
false));

and this will turn a display on:

display.viewEventPerformed(new ViewEvent(dsurf, Displayable.TOGGLE_VIEW,
true));

This should work for any display that extends Display2D such as
Object2DDisplay. "dsurf" should be replaced by whatever the variable
name for your display surface is. It and the Displayable.TOGGLE_VIEW are
not strictly necessary but including them mimics the event that is fired
when you turn off a display via the gui. And of course, "display" should
be replaced by your own display variable name.

ViewEvent is in the uchicago.src.sim.gui package so you'll have to
import that if you haven't already.

Lastly, this display will not be turned off until the next call to
update the display surface.

Nick

On Wed, 2002-09-04 at 10:17, George E. Danner wrote:
> Dear Nick-
>  
> Sorry to bug you again.I couldn't find a reference to this in the
> docs..is there any way to programmatically turn on and off the
> visibility of layers (i.e. an Object2DDisplay)?
>  
> Thanks!!
>  
> George E. Danner
> Industrial Science
>  
-- 
Nick Collier 
Social Science Research Computing 
University of Chicago 
http://repast.sourceforge.net

Re: [Repast-interest] Re-start problem

[Nick C. <nic...@ve...>]

Ugh. Typo in the docs which I've just fixed. Where they refer to
heatBugs.jar, they should read bugs.jar. Try it with that and it should
work.

You can also start heat bugs with 

java -cp c:/repast/lib/repast.jar -jar c:/repast/demo/bugs/bugs.jar

Thanks for catching this,

Nick

On Sun, 2002-09-01 at 00:46, Daniel Calhoun wrote:
> I am just getting back to Repast efforts, after long absence, and
> working from version 1.4.  In the "How to Run a RePast Simulation,"
> the example to run the HeatBugs example produces an "inititialization
> error" --
> "java.lang.IllegalArgumentException:Can't find
> uchicago.src.sim.heatBugs.HeatBugsModel."
> Since the jar file is sitting in the system, and works well enough
> when double-clicked to start it, what is not happening?
> 
> 
> 
> -------------------------------------------------------
> This sf.net email is sponsored by: OSDN - Tired of that same old
> cell phone?  Get a new here for FREE!
> https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390
> _______________________________________________
> Repast-interest mailing list
> Rep...@li...
> https://lists.sourceforge.net/lists/listinfo/repast-interest
-- 
Nick Collier 
Social Science Research Computing 
University of Chicago 
http://repast.sourceforge.net

[Repast-interest] Re-start problem

[Daniel C. <bus...@so...>]

I am just getting back to Repast efforts, after long absence, and
working from version 1.4.  In the "How to Run a RePast Simulation,"
the example to run the HeatBugs example produces an "inititialization
error" --
"java.lang.IllegalArgumentException:Can't find
uchicago.src.sim.heatBugs.HeatBugsModel."
Since the jar file is sitting in the system, and works well enough
when double-clicked to start it, what is not happening?

Re: [Repast-interest] Mac Users

[Max J C. <mxc...@sa...>]

However, I checked with some of my sources (I'm the Apple Campus Rep for
the university of Pennsylvania) and it is not necessarily the case that
Jaguar (10.2) will support java 1.4, at least not immediately when it is
released.  However, there is a developer preview of java 1.4 that has been
seeded, so if not immediately when Jaguar is released, java 1.4 will be
supported soon.

So, for the immediate time being, requiring 1.4 will make life with repast
difficult for Mac users wanting to use repast 2.0 for the time being.
However, even when 1.4 is supported, it will require the paid upgrade.

-Max Cantor





On Fri, 16 Aug 2002, Owen Densmore wrote:

>
> On Friday, August 16, 2002, at 09:11 AM, Nick Collier wrote:
>
> > Hi,
> >
> > I have a question for any Mac users out there. I've heard that for 0S X
> > the upgrade to Java 1.4 will be part of an OS upgrade that you'll have
> > to pay for. I can't tell from apple's web site if this upgrade will also
> > be available for free for earlier versions of OS X. Anyone know anything
> > about this?
> >
> > The reason I ask is there are useful additions to the java API only
> > available in 1.4. I'm hesitant to write any code dependent on these
> > additions until we see 1.4 for OS X. And if its an upgrade you'd have to
> > pay for we'd have to think hard about it as we don't want people to have
> > to pay, even indirectly, to use repast.
> >
> > Nick
>
> I asked the apple java developers list and got the attached.
> The archive is available:
>    http://www.lists.apple.com/mailman/listinfo/java-dev
>
> Apparently the answer is "Java 1.4 requires Jaguar"
>
> Owen
>
> To: jav...@li...
> Subject: Re: Fwd: [Repast-interest] Mac Users
> From: gf...@ep...
> Date: Fri, 16 Aug 2002 10:09:34 -0700
>
> Owen Densmore wrote:
> |Do we know if java 1.4 will only run on Jaguar?  See below:
>
> Apple has stated that Java 1.4 uses Jaguar-only features, so, yes, it
> will run only on Jaguar. (For the official word on things, check the
> archives for posts from Allen Denison [al...@ap...]; he's the head
> Java guy.)
>
> Glen Fisher
>
> Date: Fri, 16 Aug 2002 10:39:38 -0700
> To: jav...@li...
> From: Greg Guerin <glg...@am...>
> Subject: Re: 1.4 yet again
>
> Owen Densmore <ow...@su...> wrote:
>
> Do we know if java 1.4 will only run on Jaguar?
> [..concerns about using 1.4 features that require paid upgrade..]
>
> Apple has stated the following:
>    1) Mac OS 10.2 is not a free upgrade.
>    2) Java 1.4 will not run on less than 10.2.
>
> The obvious conclusion is that for users to run Java 1.4, a paid upgrade
> to
> 10.2 (at least) will be necessary.  Yet another reason for not
> necessarily
> jumping headlong into 1.4-requiring development without any means of
> retreat.
>
> If I were in a position where I had to choose, and the reason I wanted
> 1.4
> was for some new feature that there was no other way to obtain, I'd
> engineer my product to use abstractions and factories.  Make a 1.3
> factory
> and implementation, as well as a 1.4 one.  Then have the program pick the
> factory and imp suited to the platform.  The 1.3 imp may be slower or
> less
> featureful than the 1.4 imp, but at least I'm not locking the product
> entirely out of the 1.3 market just because I can't be bothered to
> engineer
> a little platform-sensing flexibility.
>
>    -- GG
>
> Owen Densmore           Sun Microsystems Labs   Work: 650-336-1678
> ow...@su...            2600 Casey Ave          Cell: 650-796-0652
> http://backspaces.net/  Mountain View, CA 94043 Home: 650-321-7210
> SantaFe: 505-988-3787   451 Camino Don Miguel   ow...@cy...
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by: OSDN - Tired of that same old
> cell phone?  Get a new here for FREE!
> https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390
> _______________________________________________
> Repast-interest mailing list
> Rep...@li...
> https://lists.sourceforge.net/lists/listinfo/repast-interest
>

Re: [Repast-interest] Mac Users

[Owen D. <ow...@su...>]

On Friday, August 16, 2002, at 09:11 AM, Nick Collier wrote:

> Hi,
>
> I have a question for any Mac users out there. I've heard that for 0S X
> the upgrade to Java 1.4 will be part of an OS upgrade that you'll have
> to pay for. I can't tell from apple's web site if this upgrade will also
> be available for free for earlier versions of OS X. Anyone know anything
> about this?
>
> The reason I ask is there are useful additions to the java API only
> available in 1.4. I'm hesitant to write any code dependent on these
> additions until we see 1.4 for OS X. And if its an upgrade you'd have to
> pay for we'd have to think hard about it as we don't want people to have
> to pay, even indirectly, to use repast.
>
> Nick

I asked the apple java developers list and got the attached.
The archive is available:
   http://www.lists.apple.com/mailman/listinfo/java-dev

Apparently the answer is "Java 1.4 requires Jaguar"

Owen

To: jav...@li...
Subject: Re: Fwd: [Repast-interest] Mac Users
From: gf...@ep...
Date: Fri, 16 Aug 2002 10:09:34 -0700

Owen Densmore wrote:
|Do we know if java 1.4 will only run on Jaguar?  See below:

Apple has stated that Java 1.4 uses Jaguar-only features, so, yes, it 
will run only on Jaguar. (For the official word on things, check the 
archives for posts from Allen Denison [al...@ap...]; he's the head 
Java guy.)

Glen Fisher

Date: Fri, 16 Aug 2002 10:39:38 -0700
To: jav...@li...
From: Greg Guerin <glg...@am...>
Subject: Re: 1.4 yet again

Owen Densmore <ow...@su...> wrote:

Do we know if java 1.4 will only run on Jaguar?
[..concerns about using 1.4 features that require paid upgrade..]

Apple has stated the following:
   1) Mac OS 10.2 is not a free upgrade.
   2) Java 1.4 will not run on less than 10.2.

The obvious conclusion is that for users to run Java 1.4, a paid upgrade 
to
10.2 (at least) will be necessary.  Yet another reason for not 
necessarily
jumping headlong into 1.4-requiring development without any means of
retreat.

If I were in a position where I had to choose, and the reason I wanted 
1.4
was for some new feature that there was no other way to obtain, I'd
engineer my product to use abstractions and factories.  Make a 1.3 
factory
and implementation, as well as a 1.4 one.  Then have the program pick the
factory and imp suited to the platform.  The 1.3 imp may be slower or 
less
featureful than the 1.4 imp, but at least I'm not locking the product
entirely out of the 1.3 market just because I can't be bothered to 
engineer
a little platform-sensing flexibility.

   -- GG

Owen Densmore           Sun Microsystems Labs   Work: 650-336-1678
ow...@su...            2600 Casey Ave          Cell: 650-796-0652
http://backspaces.net/  Mountain View, CA 94043 Home: 650-321-7210
SantaFe: 505-988-3787   451 Camino Don Miguel   ow...@cy...

Re: [Repast-interest] Limit on Parameters

[Nick C. <nic...@ve...>]

Hi,

There's no caching or maximum number of parameters. Did you remember to
include the new parameter names in the String[] returned by
getInitParams. I often forget to do this.

Nick


On Fri, 2002-08-16 at 16:13, Max J Cantor wrote:
> I just finished up creating a repast simulation, and after a few runs,
> changed the code to add more parameters (around 25).  However, I still
> only see the original 15 parameters in the gui.  I cleaned out the project
> and recompiled.  Is there some sort of caching going on, or is there
> simply a maximum number of parameters?
> 
> Max Cantor
> University of Pennsylvania
> 
> 
> 
> 
> 
> 
> 
> -------------------------------------------------------
> This sf.net email is sponsored by: OSDN - Tired of that same old
> cell phone?  Get a new here for FREE!
> https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390
> _______________________________________________
> Repast-interest mailing list
> Rep...@li...
> https://lists.sourceforge.net/lists/listinfo/repast-interest
-- 
Nick Collier 
Social Science Research Computing 
University of Chicago 
http://repast.sourceforge.net

Re: [Repast-interest] Limit on Parameters

[Gulyas L. <gu...@om...>]

Hi Max,

Could you attach your code? Without that it's really hard to tell.
In general, there should be no chaching anywhere...

Gulya

On Fri, 16 Aug 2002, Max J Cantor wrote:

> I just finished up creating a repast simulation, and after a few runs,
> changed the code to add more parameters (around 25).  However, I still
> only see the original 15 parameters in the gui.  I cleaned out the project
> and recompiled.  Is there some sort of caching going on, or is there
> simply a maximum number of parameters?
>
> Max Cantor
> University of Pennsylvania
>
>
>
>
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by: OSDN - Tired of that same old
> cell phone?  Get a new here for FREE!
> https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390
> _______________________________________________
> Repast-interest mailing list
> Rep...@li...
> https://lists.sourceforge.net/lists/listinfo/repast-interest
>

-- 
--
Laszlo Gulyas				las...@sz...
AI Laboratory				http://www.sztaki.hu/~gulyas/
Computer and Automation Research Inst.  H-1111, Budapest, Kende u. 13-17.
Hungarian Academy of Sciences		* 36 1 209-6194

--> On leave at Government Department, Harvard University      <--
--> 113b VanSerge, Mails: 34 Kirkland st., Cambridge, MA 02138 <--
--> gu...@fa..., (617) 384-9088                     <--

[Repast-interest] Limit on Parameters

[Max J C. <mxc...@sa...>]

I just finished up creating a repast simulation, and after a few runs,
changed the code to add more parameters (around 25).  However, I still
only see the original 15 parameters in the gui.  I cleaned out the project
and recompiled.  Is there some sort of caching going on, or is there
simply a maximum number of parameters?

Max Cantor
University of Pennsylvania

RE: [Repast-interest] Mac Users

[Christopher J. M. <cjmackie@Princeton.EDU>]

Nick;  Thanks for thinking of us!  We're expecting to have to pay for the
10.2 "Jaguar" upgrade, yes.  Everyone pays--including 10.1 users (you may
get a free pass if you buy 10.1 now, but anyone who got 10.1 when it came
out is on the hook).  It's also expected to be worth paying for.  I haven't
paid any attention to what version of Java is shipping in 10.2, but if they
say 1.4, I believe them.

I would encourage you to write for Java 1.4, either way.  The vast majority
of Mac people haven't yet switched to OS-X; if they switch after September,
they'll be switching to 10.2 at no additional cost.  Those who have switched
already are highly likely to want to pay for 10.2 on grounds having nothing
to do with Java and RePast support.  Either way, I don't think it's
realistic to set a "zero cost" standard in such matters: who's paying for
the hardware, after all?  RePast grows as its capabilities grow: if I have
to spend a few extra $$ on an OS upgrade to get more out of it, and I need
those capabilities, it makes sense for me to do it, just as I would pay a
few more $$$ for the RAM to run bigger models with it, etc.

Hope this helps, --Chris

-----Original Message-----
From: rep...@li...
[mailto:rep...@li...]On Behalf Of Nick
Collier
Sent: Friday, August 16, 2002 12:11 PM
To: rep...@li...
Subject: [Repast-interest] Mac Users


Hi,

I have a question for any Mac users out there. I've heard that for 0S X
the upgrade to Java 1.4 will be part of an OS upgrade that you'll have
to pay for. I can't tell from apple's web site if this upgrade will also
be available for free for earlier versions of OS X. Anyone know anything
about this?

The reason I ask is there are useful additions to the java API only
available in 1.4. I'm hesitant to write any code dependent on these
additions until we see 1.4 for OS X. And if its an upgrade you'd have to
pay for we'd have to think hard about it as we don't want people to have
to pay, even indirectly, to use repast.

Nick

--
Nick Collier
Social Science Research Computing
University of Chicago
http://repast.sourceforge.net



-------------------------------------------------------
This sf.net email is sponsored by: OSDN - Tired of that same old
cell phone?  Get a new here for FREE!
https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390
_______________________________________________
Repast-interest mailing list
Rep...@li...
https://lists.sourceforge.net/lists/listinfo/repast-interest

[Repast-interest] Mac Users

[Nick C. <nic...@ve...>]

Hi,

I have a question for any Mac users out there. I've heard that for 0S X
the upgrade to Java 1.4 will be part of an OS upgrade that you'll have
to pay for. I can't tell from apple's web site if this upgrade will also
be available for free for earlier versions of OS X. Anyone know anything
about this?

The reason I ask is there are useful additions to the java API only
available in 1.4. I'm hesitant to write any code dependent on these
additions until we see 1.4 for OS X. And if its an upgrade you'd have to
pay for we'd have to think hard about it as we don't want people to have
to pay, even indirectly, to use repast.

Nick

-- 
Nick Collier 
Social Science Research Computing 
University of Chicago 
http://repast.sourceforge.net

RE: [Repast-interest] compile RePast java on solaris

[Deffarges, B. {PRG~Basel} <BERNARD.DEFFARGES@Roche.COM>]

Hi Xiadong,=20

A library is missing in your classpath, the colt.jar library. You should =
just add it to the classpath and it should solve your problem.

Cheers,

Bernard Deffarges
=20

-----Original Message-----
From: Xiaodong Li [mailto:xia...@cs...]
Sent: Wednesday, August 14, 2002 11:38 AM
To: rl...@fi...
Cc: Rep...@li...
Subject: [Repast-interest] compile RePast java on solaris



Dear Prof. Rick Riolo,

I am wondering if you can offer me some clues about what went wrong here =

when I tried to compile HeatBugs (Repast version) on our server.  I have =
no=20
problems running the demos, but I did encounter errors when trying to=20
compile HeatBugs codes. I used:

javac -classpath /usr/local/pkg/repast-1.4.1/lib/repast.jar *.java

Is the above command correct?  I then got 14 errors:

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
...

HeatBugsModel.java:105: cannot resolve symbol
symbol  : variable Uniform
location: class uchicago.src.sim.heatBugs.HeatBugsModel
       int idealTemp =3D Uniform.staticNextIntFromTo(minIdealTemp, =
maxIdealTemp);
                       ^
HeatBugsModel.java:106: cannot resolve symbol
symbol  : variable Uniform
location: class uchicago.src.sim.heatBugs.HeatBugsModel
       int outputHeat =3D Uniform.staticNextIntFromTo(minOutputHeat,=20
maxOutputHeat);
                        ^
HeatBugsModel.java:110: cannot resolve symbol
symbol  : variable Uniform
location: class uchicago.src.sim.heatBugs.HeatBugsModel
         x =3D Uniform.staticNextIntFromTo(0, space.getSizeX() - 1);
             ^
HeatBugsModel.java:111: cannot resolve symbol
symbol  : variable Uniform
location: class uchicago.src.sim.heatBugs.HeatBugsModel
         y =3D Uniform.staticNextIntFromTo(0, space.getSizeY() - 1);
             ^
14 errors
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

Thanks very much.

Regards,

- Xiaodong

RMIT Uni. Melbourne, VIC, Australia



-------------------------------------------------------
This sf.net email is sponsored by: Dice - The leading online job board
for high-tech professionals. Search and apply for tech jobs today!
http://seeker.dice.com/seeker.epl?rel_code=3D31
_______________________________________________
Repast-interest mailing list
Rep...@li...
https://lists.sourceforge.net/lists/listinfo/repast-interest

[Repast-interest] compile RePast java on solaris

[Xiaodong Li <xia...@cs...>]

Dear Prof. Rick Riolo,

I am wondering if you can offer me some clues about what went wrong here 
when I tried to compile HeatBugs (Repast version) on our server.  I have no 
problems running the demos, but I did encounter errors when trying to 
compile HeatBugs codes. I used:

javac -classpath /usr/local/pkg/repast-1.4.1/lib/repast.jar *.java

Is the above command correct?  I then got 14 errors:

==============================================
...

HeatBugsModel.java:105: cannot resolve symbol
symbol  : variable Uniform
location: class uchicago.src.sim.heatBugs.HeatBugsModel
       int idealTemp = Uniform.staticNextIntFromTo(minIdealTemp, maxIdealTemp);
                       ^
HeatBugsModel.java:106: cannot resolve symbol
symbol  : variable Uniform
location: class uchicago.src.sim.heatBugs.HeatBugsModel
       int outputHeat = Uniform.staticNextIntFromTo(minOutputHeat, 
maxOutputHeat);
                        ^
HeatBugsModel.java:110: cannot resolve symbol
symbol  : variable Uniform
location: class uchicago.src.sim.heatBugs.HeatBugsModel
         x = Uniform.staticNextIntFromTo(0, space.getSizeX() - 1);
             ^
HeatBugsModel.java:111: cannot resolve symbol
symbol  : variable Uniform
location: class uchicago.src.sim.heatBugs.HeatBugsModel
         y = Uniform.staticNextIntFromTo(0, space.getSizeY() - 1);
             ^
14 errors
=======================================

Thanks very much.

Regards,

- Xiaodong

RMIT Uni. Melbourne, VIC, Australia

Re: [Repast-interest] Why is round 0 special?

[Nick C. <nic...@ve...>]

Well, the short story is that some were confused by scheduling at 0, so
we changed the schedule to begin with 1. Its one of those things that I
wish I hadn't changed, but so be it.

Anyway, the following may help with your problem. Schedule an action at
tick 1 that does the record then whatever the "regular" action for your
model is. At that point, you've done your record before anything else
and your regular runs are in synch with the tick count. You could also
play with the Schedule.LAST parameter to get what you want. Lastly, you
could do a record in begin() or buildModel() before anything scheduled
has executed, but after your model is completely setup. This should, I
think but I never tried it, record the data at tick 0.

hope this helps,

Nick


On Sun, 2002-08-11 at 21:51, A. Maurits van der Veen wrote:
> Hi all,
> 
> In my simulation, I would like to generate a report (i.e. send a 'record'
> to the recorder) after the simulation has been set up but prior to running
> the first round.
> 
> I tried to do this by scheduling a record at time 0, but Repast simply
> changes that to a 1 and then I get 2 records executed after the first
> timestep.
> 
> It turns out that I can solve the problem by having a record at time 1 and
> starting to run the rounds at time 2. But then I have to make internal
> adjustments for the fact that the tickcount is always 1 ahead of the
> number of rounds we have run, which is not elegant.
> 
> Is there any particular reason it is not possible to do something at round
> 0?
> 
> Maurits
> 
> ------------------------------------------------------------------------
> Maurits van der Veen                  ma...@sa...
> Department of Political Science       phone: (215) 898-6624
> University of Pennsylvania            http://www.sas.upenn.edu/~maurits/
> 210 Stiteler Hall
> Philadelphia, PA 19104-6215           Eppur si muove
> ------------------------------------------------------------------------
> 
> 
> 
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Repast-interest mailing list
> Rep...@li...
> https://lists.sourceforge.net/lists/listinfo/repast-interest
-- 
Nick Collier 
Social Science Research Computing 
University of Chicago 
http://repast.sourceforge.net

[Repast-interest] Why is round 0 special?

[A. M. v. d. V. <ma...@sa...>]

Hi all,

In my simulation, I would like to generate a report (i.e. send a 'record'
to the recorder) after the simulation has been set up but prior to running
the first round.

I tried to do this by scheduling a record at time 0, but Repast simply
changes that to a 1 and then I get 2 records executed after the first
timestep.

It turns out that I can solve the problem by having a record at time 1 and
starting to run the rounds at time 2. But then I have to make internal
adjustments for the fact that the tickcount is always 1 ahead of the
number of rounds we have run, which is not elegant.

Is there any particular reason it is not possible to do something at round
0?

Maurits

------------------------------------------------------------------------
Maurits van der Veen                  ma...@sa...
Department of Political Science       phone: (215) 898-6624
University of Pennsylvania            http://www.sas.upenn.edu/~maurits/
210 Stiteler Hall
Philadelphia, PA 19104-6215           Eppur si muove
------------------------------------------------------------------------

[Repast-interest] A genetic algorithm in Repast?

[Xiaodong Li <xia...@cs...>]

Does anyone know if there is a genetic algorithm developed in Repast? and 
also it is for free public access?  I am new to Repast, so any help is 
highly appreciated.

Cheers,

- Xiaodong

RMIT Uni. Melbourne, VIC, Australia